Modification of low dimensional nanostructured
TiO2 for energy application
(Modifizierung von niedrigdimensionalem nanostrukturiertem TiO2
für Energieanwendungen)
Der Technischen Fakultät
der Friedrich-Alexander-Universität Erlangen-Nürnberg
zur Erlangung des Doktorgrades
Doktor der Ingenieurwissenschaften (Dr.-Ing.)
vorgelegt von
Fahimeh Shahvaranfard
aus Shiraz, Iran
Als Dissertation genehmigt
von der Technischen Fakultät
der Friedrich-Alexander-Universität Erlangen-Nürnberg
Tag der mündlichen Prüfung: 21.01.2022
Vorsitzender des Promotionsorgans: Prof. Dr.-Ing. Knut Graichen
Gutachter: Prof. Dr. Christoph J. Brabec
Prof. Dr. Kyle G. Webber
Acknowledgements
Firstly, I would like to show my gratitude and appreciation to my advisor, Prof. Dr. Christoph
J. Brabec, for great supporting and sharing his knowledge. I would also like to thank Prof. Dr.
Patrik Schmuki for allowing me to have the opportunity to access the laboratory and research
facilities at LKO.
I sincerely thank Prof. Dr. Knut Graichen for his effective support.
I would like to thank my thesis committee members for their valuable comments and helpful
suggestions.
I also want to acknowledge all my colleagues at i-MEET and LKO. My special thanks go to
Prof. Marco Altomare. My PhD would not have been completed smoothly without his scientific
guidance. I would also like to express my gratitude to Dr. Ning Liu for his wonderful advice.
I would also like to show my appreciation to Dr. Saman Hosseinpour, Dr. Shiva Mohajernia,
Prof. Nhat Truong Nguyen and Dr. Sina Hejazi for their great scientific support and their
friendliness.
Last but not least, I would like to express my deepest gratitude to my parents, Mohammadali
and Zari; my brothers, Hossein and Mohsen; my sisters in law, Negar and Mozhgan; and dear
Ava, Arash and Arman for their love, support and motivation.
i
Contents
Acknowledgments ....................................................................................................................... i
Abstract ...................................................................................................................................... 1
1 Introduction and motivation .................................................................................................... 3
2 Background and state of the art ............................................................................................... 5
2.1 Nanostructured TiO2 ........................................................................................................ 5
2.1.1
Titanium dioxides................................................................................................. 5
2.1.2
Different TiO2 nanostructure forms ..................................................................... 7
2.2 Energy conversion with nanostructured TiO2 ................................................................ 18
2.2.1 Solar energy conversion to electricity .................................................................... 18
2.2.2 Solar energy conversion to valuable chemicals ...................................................... 28
3 Scientific objectives .............................................................................................................. 39
4 Thesis overview..................................................................................................................... 40
5 Author contributions ............................................................................................................. 42
6 Summary ............................................................................................................................... 43
7 Outlook .................................................................................................................................. 45
Reference .................................................................................................................................. 46
Abbreviations and symbols ...................................................................................................... 52
Appendix A publication I ......................................................................................................... 54
Appendix B publication II ........................................................................................................ 74
Appendix C publication III ...................................................................................................... 96
Appendix D publication IV .................................................................................................... 117
ii
Abstract
Among different semiconductors, the largest number of studies for solar energy conversion
are being conducted on titanium dioxide (TiO2). TiO2 is a semiconductor with high
photoactivity which is inexpensive, non-toxic, and stable in chemical and photochemical
reactions. TiO2 is mainly used in nanostructure forms such as zero-dimensional (e.g.
nanoparticles), one-dimensional (e.g. nanotubes and nanorods) and two-dimensional (e.g.
nanosheets). Nanostructured TiO2 can provide unique advantages such as high surface area and
directional electron transfer properties for photovoltaic and photocatalysis reactions. In this
thesis we focus on synthesizing nanostructured TiO2 and employing novel approaches for
modifying its properties for use in photovoltaic solar cells and photocatalysis H2 evolution.
Firstly, we engineered the interface of the electron transport layer (ETL) and perovskite
light absorber in perovskite solar cells (PSCs). Hydrothermally synthesized TiO2 nanorods were
used as ETL in methylammonium lead iodide (MAPI)-based perovskite solar cells. TiO2
nanorods show excellent charge carrier mobility and high electron lifetime which are two key
properties of an efficient electron transport layer. For the first time, the effect of dual
modification of TiO2 nanorods (NRs) array surface with TiCl4 treatment and PC61BM
monolayer deposition on the overall cell performance was studied. Under dual modifications,
synergistic effects were observed which result in significant improvement in cell efficiency and
reduction of device hysteresis. The performance enhancement is attributed to the efficient
electron mobility at the interface of TiO2 ETL with perovskite film due to the passivation of
defects.
Secondly, the effect of the crystallographic phase structure of TiO2 thin film as ETL on the
cell efficiency of MAPI-based perovskite solar cells is explored. Anatase and rutile TiO2 were
deposited by magnetron sputtering technique under controlled conditions. The results show the
comparable performance of the devices fabricated with both anatase and rutile ETLs that are
deposited under similar conditions.
In the third part, we investigated the effectiveness of PtCu alloy co-catalyst on TiO2
nanotube arrays in photocatalysis H2 generation. Pt and Cu nanofilms were deposited on highly
ordered TiO2 nanotubes and subsequently were converted to nanoparticles by the effective
solid-state dewetting. The highly ordered TiO2 nanotubes not only provide very efficient
semiconductor geometry due to a high degree of order but also the ideal hexagonally patterned
scaffold for controllable metal deposition. The activity of PtCu bimetallic co-catalyst with
regard to Pt and Cu monometallic co-catalyst on TiO2 catalyst was studied for producing
1
hydrogen from water in the presence of methanol as a hole scavenger. The PtCu co-catalyst
showed a remarkably higher photocatalytic activity compared to single Pt and Cu. This
enhancement is ascribed to electronic interactions between Pt and Cu which leads to an increase
in electron density on Pt, improving charge transfer for hydrogen evolution reactions.
The last part of the study covers hydrothermally grown single crystal anatase TiO2
nanosheets on FTO. Hydrothermal conditions such as synthesis time, the concentration of Ti
and F precursors, and HCl were systematically tuned in order to synthesize TiO2 nanosheets
with a wide range of morphologies and faceting. Then the effect of morphology and faceting
on the electrochemical activity of nanosheets is further discussed. We find that the
photoelectrochemical performance of the nanosheets layer can be strongly enhanced when their
morphology and faceting are optimized. The performance enhancement under optimized
condition is ascribed to simultaneously increasing the active surface area and percentage of
highly active (001) facets.
I believe the results of this thesis provide considerable insight into the importance of not
only using TiO2 in specific nanostructure forms but also appropriate modification approaches
in order to obtain efficient solar energy conversion systems.
2
Kurzzusammenfassung
Unter den verschiedenen Halbleitern werden die meisten Studien zur Umwandlung von
Sonnenenergie mit Titandioxid (TiO2) durchgeführt. TiO2 ist ein Halbleiter mit hoher
Photoaktivität, der kostengünstig, ungiftig und bei chemischen und photochemischen
Reaktionen stabil ist. TiO2 wird hauptsächlich in Form von Nanostrukturen wie
nulldimensionalen (z. B. Nanopartikel), eindimensionalen (z. B. Nanoröhren und
Nanostäbchen) und zweidimensionalen (z. B. Nanoblätter) verwendet. Nanostrukturiertes TiO2
kann
einzigartige
Vorteile
wie
eine
große
Oberfläche
und
gerichtete
Elektronentransfereigenschaften für photovoltaische und photokatalytische Reaktionen bieten.
In dieser Arbeit konzentrieren wir uns auf die Synthese von nanostrukturiertem TiO2 und die
Anwendung neuartiger Ansätze zur Modifizierung seiner Eigenschaften für den Einsatz in der
Photokatalyse, der H2-Entwicklung und in photovoltaischen Solarzellen.
Zunächst entwickeln wir die Grenzfläche zwischen der Elektronentransportschicht (ETL)
und dem Lichtabsorber in Perowskit-Solarzellen (PSCs). Hydrothermal synthetisierte TiO2Nanostäbchen werden als ETL in Perowskit-Solarzellen auf Methylammonium-Bleiiodid-Basis
(MAPI)
verwendet.
Ladungsträgerbeweglichkeit
TiO2-Nanostäbchen
und
eine
zeigen
hohe
eine
ausgezeichnete
Elektronenlebensdauer,
zwei
Schlüsseleigenschaften einer effizienten Elektronentransportschicht. Zum ersten Mal wird die
Auswirkung einer dualen Modifikation der Oberfläche von TiO2-Nanostäbchen (NRs) mit
TiCl4-Behandlung und PC61BM-Monolayer-Abscheidung auf die Gesamtzellenleistung
untersucht. Bei dualen Modifikationen werden synergistische Effekte beobachtet, die zu einer
signifikanten Verbesserung der Zelleffizienz und einer Verringerung der Hysterese der
Solarzelle führen. Die Leistungssteigerung wird auf die effiziente Elektronenmobilität an der
Grenzfläche zwischen dem TiO2 ETL und dem Perowskit-Film aufgrund der Passivierung von
Defekten zurückgeführt.
Zweitens wird die Auswirkung der kristallographischen Phasenstruktur der TiO2Dünnschicht als ETL auf die Zelleffizienz von MAPI-basierten Perowskit-Solarzellen
untersucht. Anatas- und Rutil-TiO2 wurden unter kontrollierten Bedingungen mittels
Magnetron-Sputtertechnik abgeschieden. Die Ergebnisse zeigen eine vergleichbare Leistung
der Solarzellen, unabhängig ob die ETL auf Anatas- oder Rutil-TiO2 basierte.
1
Im dritten Teil untersuchten wir die Wirksamkeit von PtCu-Legierungen als CoKatalysatoren auf TiO2-Nanoröhrchen-Arrays bei der photokatalytischen H2-Erzeugung. Ptund Cu-Nanofilme wurden auf hoch geordneten TiO2-Nanoröhrchen abgeschieden und
anschließend durch eine effektive Solid State Dewatering in Nanopartikel umgewandelt. Die
hoch geordneten TiO2-Nanoröhrchen bieten aufgrund ihres hohen Ordnungsgrades nicht nur
eine sehr effiziente Halbleitergeometrie, sondern auch ein ideales hexagonal gemustertes Gerüst
für die kontrollierbare Metallabscheidung.
Die Aktivität des bimetallischen PtCu-Co-
Katalysators im Vergleich zum monometallischen Pt- und Cu-Co-Katalysator auf TiO2Katalysator wurde für die Herstellung von Wasserstoff aus Wasser in Gegenwart von Methanol
als Lochfänger untersucht. Der PtCu-Cokatalysator zeigte eine bemerkenswert höhere
photokatalytische Aktivität im Vergleich zu einzelnen Pt und Cu Beschichtungen. Diese
Steigerung wird auf elektronische Wechselwirkungen zwischen Pt und Cu zurückgeführt, die
zu einer Erhöhung der Elektronendichte auf Pt führen und so die Ladungstransferreaktion für
die Bildung von Wasserstoff verbessern.
Der letzte Teil der Studie befasst sich mit hydrothermal gewachsenen einkristallinen
Anatas-TiO2-Nanoblättern auf FTO. Die hydrothermalen Bedingungen wie die Synthesezeit,
die Konzentration von Ti und F-Vorläufern und HCl wurden systematisch abgestimmt, um
TiO2-Nanoblätter mit einer breiten Palette von Morphologien und Facetten zu synthetisieren.
Anschließend wird die Auswirkung der Morphologie und der Facettierung auf die
elektrochemische Aktivität der Nanoblätter diskutiert. Wir stellen fest, dass die
photoelektrochemische Leistung der Nanobleche stark verbessert werden kann, wenn ihre
Morphologie und Facettierung optimiert werden. Die Leistungssteigerung unter optimierten
Bedingungen wird auf die gleichzeitige Erhöhung der aktiven Oberfläche und des Anteils
hochaktiver (001)-Facetten zurückgeführt.
Ich glaube, dass die Ergebnisse dieser Arbeit nicht nur einen erheblichen Einblick in die
Bedeutung von spezifischen Nanostrukturformen für die Anwendung von TiO2 haben, sondern
auch geeigneter Modifikationsansätze geben, um effiziente Solarenergieumwandlungssysteme
zu erhalten.
2
1. Introduction and motivation
1. Introduction and motivation
Air pollution and climate change as a result of fossil fuels have become a critical issue. An
approach for reducing pollution and providing energy which is both green and clean has been
growing over the past few decades. Eco-friendly solar power is widely seen as a suitable
alternative to fossil fuels due to its low cost and unlimited availability. [1–3] Photocatalysts and
solar cells have been invented to both use solar energy and convert it to chemical or electrical
energy.
Titanium dioxide (TiO2) and its potential have been intensively investigated and it is
considered one of the most widely used semiconductors in solar energy conversion. TiO2 has
valuable properties such as being cost-effective, stable, non-toxic, and highly resistant to
corrosion.[4,5] When TiO2 is under irradiation using the sufficient photon energy, the
photoexcited electron can be moved from a valence band to a higher energy band (conduction
band) and electron-hole pairs are created.[6–8] Photoexcited electrons (e-) in the conduction band
and photoexcited electron vacancies (holes (h+)) in the valence band create a flow of charge
which can be utilized in photovoltaic cells like solar cells or as an oxidizing agent in
photocatalysis reactions.
In most applications, TiO2 with nanostructure morphologies (when at least one dimension
is < 100 nm) is preferred compared to bulk TiO2. Nanoscale TiO2 materials have different
charge transport properties and higher surface to volume ratio which increase the number of
active sites. As a result, the activity of the materials in their nanoscale formats can be increased.
In the last few years, one-dimensional TiO2 (1D TiO2) such as nanorods, nanowires and
nanotubes and two-dimensional TiO2 (2D TiO2) such as nanoflakes and nanofilms have been
widely investigated due to their superior electronic properties.[9–11]
TiO2 has been extensively used in different types of solar cells. Solar cells are considered
to be one of the most effective and low-cost means of collecting solar energy and converting it
to electrical energy. Recently, perovskite solar cells (PSCs) have attracted great attention owing
to their unique optical and electronic properties. TiO2 in different nanostructure forms is mainly
employed as one of the important functional parts in a PSC, the electron transport layer (ETL).
However, TiO2 ETL properties and its interface with the light absorber layer play an essential
role in the overall efficiency of the solar cell. The energy barrier, due to the presence of defects
and trap states on the ETL surface leading to charge accumulation at interface of perovskite
3
1. Introduction and motivation
with ETL, not only decreases the charge transportation efficiency but can also greatly affect the
hysteresis in PSCs.[12,13] Therefore, it is necessary to engineer the interface of TiO2 ETL with
the perovskite light absorber. Multitreatment of nanostructured TiO2 as ETL in PSCs compared
to single treatment is considered a noble approach for effective surface modification due to the
synergistic effects.
Another common way to utilize solar energy is photocatalytic reactions as solar-tochemical energy reactions by means of a photocatalyst. A considerable amount of literature has
been dedicated to producing high-performance photocatalytic materials.[14–18] Titania’s
photocatalytic applications have been widely investigated for organic compound decomposition
because of the capability of TiO2 as an oxidizing agent for organic and inorganic materials
through redox reactions.[19–21] Nanostructured TiO2 is preferred due to its unique features such
as improved light absorption, enhanced charge separation and transport, preferentially exposed
crystal facets and large surface area. However, the photocatalytic activity of TiO2 under opencircuit conditions is mostly constrained by the recombination of electron-hole pairs and thus,
TiO2 alone cannot be efficient in photocatalytic reactions. Therefore, many studies moved
towards novel strategies for employing effective co-catalysts that can accelerate electron
transport and improve photocatalytic efficiency.
The aim of this work
TiO2 nanostructure has been extensively utilized for energy conversion applications.
However, TiO2 still needs to overcome the intrinsic drawback due to its wide band gap (> 3 eV)
and fast recombination of photoinduced electrons and holes. In perovskite solar cells,
optimizing the morphological and structural features of titania as ETL and its interface with the
perovskite layer plays a key role in the efficiency of the devices. This work introduces, a specific
strategy for modifying the perovskite/TiO2 ETL interface and its impact on the performance of
the cells. Moreover, the effect of the crystal phase structure of TiO2 ETL on the overall
efficiency of PSCs is investigated.
In photocatalytic reactions for hydrogen evolution, decoration of TiO2 with a noble or nonnoble catalyst in order to decrease recombination and improve photocatalytic efficiency is of
vital importance. The aim of part of this study was to find out the effect of the contribution from
the combination of noble and non-noble co-catalyst nanoparticles (NPs) and a typical deposition
approach on the enhancement of photocatalysis hydrogen evolution efficiency.
4
2. Background and state of the art
2. Background and state of the art
The goal of this chapter is to provide an overview of titanium dioxide and its properties. The
synthesis processes and benefits of using different forms of nanostructured TiO2 such as
nanotubes, nanorods, nanosheets and nanofilms in solar energy conversion are discussed. In
addition, the limitations of using nanostructured TiO2 in solar energy conversion devices such
as photovoltaic solar cells and photocatalytic cells are investigated and the strategies for
overcoming these limitations are introduced.
2.1 Nanostructured TiO2
2.1.1 Titanium dioxides
Titanium belongs to group IV of elements and, like most of the metals, is able to form
different oxides. Titanium dioxide (TiO2) or titania is the naturally appearing compound as a
result of the reaction of titanium with oxygen. In the late 1800s, titanium dioxide was produced
in the laboratory and then in the early 20th century, after discovering its properties as a white
pigment, it was produced in large quantities.[22] Titania has several distinguishing characteristics
that make it appropriate for a broad range of applications. It appears naturally as a solid and is
insoluble in water. It has a high melting point and boiling point of 1,843 ºC and 2,972 ºC,
respectively. Remarkably, it has a supremely high refractive index which makes it a very bright
material.[23,24] On top of all this, titanium dioxide has a unique photovoltaic and photocatalytic
activity under UV light, which is the focus of this thesis.
Titania can be formed in three crystalline phases: rutile (tetragonal), anatase (tetragonal)
and brookite (orthorhombic). Rutile is considered the most thermodynamically stable
crystallographic phase of TiO2. The anatase and brookite which are metastable phases of TiO2
can be converted to the stable rutile phase when heated to a temperature of over 600–800
°C.[25,26] It is important to note that some parameters such as impurities, morphology and
residual stress can be changed the transition temperature. Anatase is the most impressive phase
of TiO2 for numerous applications because of its intrinsic properties, such as the higher activity
of anatase compared to other TiO2 polymorphs. Table 2.1 shows the chemical, physical and
optical properties of different crystallographic phases of TiO2.
5
2. Background and state of the art
TiO2 has generally been discovered to be an n-type and indirect semiconductor. Rutile and
anatase have bandgaps of 3.0 and 3.2 eV, respectively.[27,28] The electronic properties of titania
depend on the existence of oxygen vacancies which act as structural defects. These oxygen
vacancies behave as electron donors and thus the substance consists of extra electrons which
lead to an increase in electrical conductivity.[23]
Figure 2.1 Three crystallographic structures of titanium dioxide. Adapted with permission
from ref.[29]
Table 2.1 General properties of anatase, rutile and brookite.[6,30,31]
Properties
Anatase
Rutile
Brookite
Crystal system
Tetragonal
Tetragonal
Orthorhombic
Density (g/cm3)
3.89
4.52
4.18
55.52
55.06
56.2
Semiconductor type
n
n
n
Band gap energy (eV)
3.2
3
3.3
Refractive index
2.56
2.94
2.58
Standard heat capacity Cp
(J/KgK)
6
2. Background and state of the art
The unique optical and physical properties of TiO2 make it the most studied semiconductor
for solar energy conversion application. However, design and innovation of titania from bulk
to nanostructure has also drawn considerable attention due to structural features e.g.
morphology, size, and porosity that greatly influence the physical properties of titania.
2.1.2 Different TiO2 nanostructure forms
A new sort of material is nanostructured materials which offer one of the promising
potentials for enhancing performance in a variety of applications. It is referred to as
"nanomaterial" when a material contains at the minimum one dimension smaller than 100
nm.[32] By decreasing the size of materials to nanometers, their chemical and physical properties
change. Moreover, nanostructures with different morphologies exhibit various properties. Many
outstanding studies on nanomaterials synthesis and their properties have been published.[6,33–36]
A particular property of the semiconductor nanomaterials is the improvement of electrons and
holes transport properties because of the quantum confinement effects and the geometry and
size of the materials. Furthermore, by decreasing the size of the material, the surface-to-volume
ratio (specific surface area) increases dramatically. In general, a high surface area improves the
interaction of the materials and the environment and as a result, the activity of semiconductor
material is largely affected by its size and morphology.[36,37]
Nanostructured TiO2 has been extensively used in photocatalytic applications,[38,39] solar
cells,[40,41] gas sensors[42] and nanocomposite biomaterials.[43] Its special optical properties result
in enhanced photovoltaic performance and its bifunctional mechanism leads to improved
poisoning resistance on catalyst and increase photocatalytic activity.
There are various techniques to synthesize titania nanostructures which are categorized
into five general groups; deposition methods, sol-gel, oxidation methods hydro/solvothermal
methods and chemical methods.[44] Among the different types of TiO2 nanostructured, onedimensional (1D) nanostructured like nanotubes (NTs) and nanorods (NRs) have been widely
studied due to their performance enhancement in photovoltaics and photocatalysis applications.
In general, a 1D nanostructure with size ranging from 1 to 100 nm possesses a high aspect ratio.
Thus, 1D TiO2 nanoscale materials can not only have all the typical properties of 0D TiO2 (e.g.
nanoparticles) but also show a large active surface area. These features favor mobility of
photogenerated carriers through the axial direction. Moreover, they are normally produced by
simple techniques.
7
2. Background and state of the art
Limitations. However, some critical issues can limit use of 1D nanostructured TiO2 in
different applications. Titania is a wide bandgap semiconductor (ruile: ≈ 3.0 eV, anatase: ≈ 3.2
eV), therefore, only 3-5% of total solar radiation can be absorbed by TiO2. In addition, a fast
recombination rate of photoexcited electrons and holes causes a reduction in solar conversion
efficiency. Several approaches such as doping metal and non-metal elements,[35,45] constructing
heterojunctions[45,46] and metal ion-implanted[47,48] have been reported as a means of
overcoming the issues mentioned. Therefore, absorption of light by TiO2 can be widened from
UV light to part of the visible light and electron-hole recombination resistance can be increased
at the same time. This thesis aims to introduce typical approaches for modification of different
TiO2 nanostructures for solar energy conversion by perovskite solar cells and photocatalysis
cells.
Nanotubes array by anodization
Since the discovery of carbon nanotubes by Iijima in 1991, this combination of unique
molecular geometry and impressive properties has become a milestone in nanotechnology.[49]
Subsequently, the formation of nanotubular arrangements by various transition metal oxides
(TiO2 is the most explored transition metal oxide) was demonstrated.[37] Over the past decade,
many research studies have focused on the formation of nanotubular geometry. Excellent
electronic properties such as quantum confinement effect or high charge mobility, large specific
surface area, and high mechanical strength make nanotubes a unique 1D nanostructure.[50,51]
Nanotubes may be produced by different techniques including electrochemical methods,
sol-gel methods, hydro/solvothermal approaches and template-assisted means.[52–55]
Anodization is a simple electrochemical process which is used to form an oxide layer in an
electrochemical setup when a specific voltage is applied. Anodization has become one of the
significant techniques for producing nanostructures since the innovative work of Masuda et
al.[56] in developing self-organized porous alumina. A key advantage of anodizing is that
nanotubes geometry e.g. tube length, tube diameter and tube wall thickness can be simply
controlled by the electrochemical parameters.[57] Figure 2.2 shows the anodization setup which
consists of a working electrode (anode), a counter electrode (cathode) and appropriate
electrolyte. In the anodizing process of Ti, applying sufficient voltage to the working electrode
(Ti) can drive the following reactions:[37]
8
2. Background and state of the art
Anodic reactions:
Cathodic reactions:
Ti → Ti4+ + 4e
(2.1)
Ti + 2H O → TiO + 4H + 4e
(2.2)
Ti4+ + solv → Tisolv
(2.3)
4H O + 4e
(2.4)
→ 2H + 4OH
Figure 2.2 (a) Schematic of electrochemical anodization setup, (b) nanotubes produced by
anodization method. Figure b is adapted with permission from ref. [58]
Based on the electrolyte and the certain anodization parameters, the reactions can occur
with three possibilities which are shown in Figure 2.3a: (i) the metal is steadily dissolved in
electrolyte, thus no oxide forms on the surface of the metal (electropolishing happens), (ii)
reaction of metal ions (Ti4+) with oxygen ions (O2-) leads to formation of the metal compact
oxide film which is entirely insoluble in the electrolyte, (iii) the oxide layer is partly soluble in
the electrolyte and competition between oxide formation and dissolution takes place which
causes the formation of a nanostructure film after reaching a steady state condition. The tubes
grow and become longer over time until the dissolution (etching) becomes evident and tube
growth at the bottom and dissolution at the top reach the steady-state condition.
9
2. Background and state of the art
Figure 2.3 (a) Current-time characteristics at different stages, (b) schematic of oxide
formation in the existence of F- ion. Inspired by ref. [26]
Normally, the formation of oxide film is strongly controlled by the migration rate of the
ions (Ti4+, O2-) at both metal/oxide and oxide/electrolyte interfaces (Figure 2.3b).[26,59] The
electric field (E) can be determined from the applied voltage (∆V) and thickness of the oxide
film (d).
E = ∆V/d
(2.5)
While the applied voltage is constant, the electric field is inversely related to the thickness of
the oxide layer and decreases steadily with an increase in the thickness of the oxide film. In the
case of insoluble oxide layer, the migration of the ions is limited which leads to the creation of
the compact oxide layer. When fluoride ions (F-) are present in the electrolyte, however, soluble
[TiF6]2– complex ions are formed, and the oxide film is etched according to the following
reaction:[26,37,59]
Ti4+ + 6F
→ [TiF ]
TiO + 6F
→ [TiF ]
(2.6)
+ 2H O
(2.7)
Once the equilibrium between the formation of oxide and dissolution of oxide is occurred,
the TiO2 layer with nanoporous or nanotubular morphology can grow continuously under
specific conditions. The morphology of the oxide film is influenced by the solubility of
oxidation products and the stability of the oxide film.[37] Fluoride ions play an essential role in
the formation of TiO2 nanotubes. In the presence of F- ions with a critical concentration in the
10
2. Background and state of the art
electrolyte, the oxide undergoes breakdown resulting in pore formation. Pore formation is an
initial step in the growth of nanotubes.
Figure 2.4 shows extremely ordered TiO2 nanotubes that were produced in electrolytes
containing a high concentration of HF and o-H3PO4. These types of tubes, also known as "Utubes", reveal the highest level of self-ordering among all nanotubes. U-tubes not only provides
very efficient semiconductor geometry due to a high degree of order but also the ideal
hexagonally patterned scaffold for controllable metal deposition. In addition, some
disadvantages of tubes that were grown in organic electrolytes such as carbon contamination
layer and closed tube month are not observed in this form of nanotubes. Therefore, in this study,
this type of TiO2 nanocavity is selected as a template catalyst for the deposition of metal
nanoparticles as a co-catalyst for photocatalytic hydrogen generation.
Figure 2.4 SEM image of extremely ordered TiO2 nanotube arrays. Adapted with
permission from ref. [60]
Nanorods array by hydrothermal synthesis
Hydrothermal synthesis is a process to crystallize substances from electrolyte with high
temperature and pressure. The hydrothermal technique is extensively used to synthesize a
variety of nanostructured materials due to low cost, high efficiency, simplicity and great control
of the chemical and morphology of synthesized products.[61,62] Single- and poly-crystalline
nanostructured materials can be synthesized in an autoclave by using the hydrothermal
technique. The first use of the hydrothermal method for producing TiO2 nanotubular materials
was reported by Kasuga et al. in 1998.[63] Since then, different types of TiO2 nanostructures
such as nanotubes, nanorods, nanosheets and nanobelts have been synthesized by the
hydrothermal method. This technique is favored over other methods to synthesize TiO2
nanorods (NRs).
11
2. Background and state of the art
Based on the reactants utilized for hydrothermal process of 1D TiO2, the synthesis
technique can be divided into alkali-hydrothermal and acid-hydrothermal. In alkalihydrothermal approach, TiO2 nanoparticles and sodium hydroxide solution are normally used
as reactants although in acid-hydrothermal approach, the reactants are a mixture of titanium
salts and hydrochloric acid.[63]
In general, the production of TiO2 NRs employing a hydrothermal process is conducted in
a mixture of hydrochloric acid (HCl) and deionized water (DI) by adding Ti precursor (usually
Ti (IV) butoxide (Ti(OCH2CH2CH2CH3)4 or [Ti(RO)4])) as a reactant. In the hydrothermal
aqueous solution, hydrolysis and dehydration under acidic conditions can occur by means of
the following chemical reactions:[64]
Hydrolysis:
Ti (OR)₂ + 4H₂O → Ti(OH)₄ + 4ROH
(2.8)
Dehydration:
Ti(OH)₄ + Ti(OH)₄ → 2TiO₂ + 4H₂O
(2.9)
Ti(OH)₄ + Ti(OR)₄ → 2TiO₂ + 4ROH
(2.10)
Principally, for crystal orientation, acidic concentration (OHˉ/H+ ratio) in the synthesis
electrolyte is critical.[64]
Figure 2.5 (a) Schematic of hydrothermal setup, (b) TiO2 nanorods produced by
hydrothermal method.
For many applications, nanorods are an interesting nanostructured compound to be
investigated due to their physical properties. They provide a straightforward pathway for charge
transportation which is an important parameter in the solar energy conversion efficiency.[41,65,66]
It has been reported that TiO2 nanorods show better charge carrier mobility than TiO2
nanoparticles due to having fewer grain boundaries.[64] The grain boundaries can act as charge
carrier traps and reduce electrons lifetime and charge transfer efficiency.
12
2. Background and state of the art
Moreover, TiO2 nanorods show greater electron transfer properties than other 1D TiO2
nanostructures. The electron lifetime of hydrothermally grown TiO2 nanorods is higher than
TiO2 nanotubes that are hydrothermally synthesized which leads to photoelectrochemical
efficiency around six times higher of nanorods compared to nanotubes.[64]
As a result, TiO2 nanorods have shown a great potential for use as an electron transport
layer in solar conversion devices like solar cells due to their excellent electron transport ability,
high electron lifetime and open structure for infiltration of the light absorber materials.[66–69]
Therefore, part of this thesis focuses on producing hydrothermally grown TiO2 nanorods for
use as an electron transport layer in perovskite solar cell. Furthermore, we investigated an
effective approach for TiO2 NRs ETL surface engineering to improve functionality of TiO2 as
ETL and improve device performance.
Nanosheets array by hydrothermal
It was shown by Yang et al.[70] in 2008 that single crystalline TiO2 nanosheet can be
hydrothermally synthesized in the presence of fluorine ions (F-). In general, for TiO2 with
anatase crystallographic phase, (101) facets are the most thermodynamically stable and thus
anatase TiO2 is dominated by (101) facets. However, in a solution containing fluoride ions, the
surface energy of the more reactive (001) facets decreases to the value lower than (101) facets
resulting in the formation of anatase single crystal with high ratio of (001) facets.[70–72] When
an optimum amount of F- ions is present in the synthesis solution, anatase single crystals with
tetragonal structure are produced. The formation of nanosheets in hydrothermal setup can be
described by the nucleation and growth mechanism.[72] For anatase single crystals consisting of
mainly (101) and (001) facets, because of the intrinsic electronic junction, photoexcited
electrons transfer from the (101) facets while holes transfer from the (001) facets, when the
nanosheet is exposed to an aqueous environment. This electronic junction is caused by a
difference in the surface energy of (001) and (101) facets.[73] Figure 2.6 shows the schematic of
electronic junction in single crystal TiO2 nanosheet. Therefore, unique optical and electronic
properties of anatase TiO2 nanosheet layers as a result of crystallographic surface energy
differences make these two-dimensional single crystalline materials an interesting candidate for
photocatalysis applications.
13
2. Background and state of the art
Figure 2.6 (a) Electrons and holes transfer in (101) and (001) facets, (b) electronic
junction of (101)-(001). Inspired by ref.[74]
However, controlling the morphology and the percentage of (001) facets are the two key
parameters that affect surface and electronic features of TiO2 anatase and its potential
applications. As part of this study, anatase single crystal TiO2 nanosheets were hydrothermally
produced and the influence of morphology and crystal facet ratio on photoelectrochemical
performance is investigated. It is shown how morphology and faceting optimization of TiO2
nanosheets layer can significantly improve its photoelectrochemical efficiency.
TiO2 thin films
Nanostructured titania thin films were synthesized using various methods including dip coating,
spray pyrolysis, spray coating, thermal evaporation, E-beam evaporation, magnetron sputtering,
oxidation and so on. Sputtering is one of these techniques to deposit dense and homogenous
layers. In general, sputtering is defined as the process of ejecting solid material particles from
a surface as a result of plasma or gas bombardment.[75]
Despite most of the techniques which require additional steps like thermal treatment to obtain
stable and crystalline titania for efficient photocatalytic function, the as-sputtered TiO2 thin film
can be fully crystalline. In fact, the sputtering conditions can be adjusted to not only control the
crystallographic phase but also the morphology and structure properties such as thickness,
surface roughness and crystal size of the deposited layers.
Additional advantages of the sputtering process are excellent adhesion of deposited films,
precise control on the film thickness and homogeneity. [31] Direct current magnetron sputtering
for deposition of thin films of metal oxide like titanium oxide is very common as an industrial
14
2. Background and state of the art
process for large-scale deposition. By employing this technique, high-quality films of TiO2 can
be produced even when the substrate temperature is low.
Thermal annealing
There are three distinct crystallographic phases of TiO2; anatase, rutile and brookite.
Anatase and rutile possess tetragonal crystal structure and brookite possesses orthorhombic
crystal structure. In general, thermal treatment can transform the structure of TiO2 from an
amorphous phase into a crystalline phase with different characteristics.[76] An Aanatase phase
is obtained when the synthesizing process of titania is below 600°C. Then it converts to anataserutile by heating at 800°C and further pure rutile above 1000°C. Figure 2.7 shows the phase
transition diagram of TiO2.
Figure 2.7 Phase transition diagram of TiO2. Inspired by ref.[77]
In addition to annealing temperature, other parameters such as starting materials’
composition and deposition method play a determinative role in the final crystal structure of
TiO2.[78]
Deweting of thin films
Deposited metal nanoparticles on metal oxide semiconductors are attracting attention due
to their different applications i.e. as sensors, catalysts, and photonics. There are different
techniques for depositing nanoparticles on the surface of the metal oxide such as
electrodeposition, photodeposition, sputtering and evaporation.[59]
15
2. Background and state of the art
Dewetting is one of the ideal methods to form metal nanoparticles with controlled
morphology and size from a deposited thin layer on metal oxide. Dewetting of the thin metal
films explains the conversion of a thin layer into a series of droplets or particles with favorable
energy when heated. Typically, dewetting of metal film can occur below the melting point of
the metal film and begins at temperatures between 1/3 and 2/3 of the melting point of the metal.
Thus, during dewetting the thin film stays in the solid state. In general, dewetting’s driving
force is the minimization of the surface energy of both the substrate and thin metal film. When
the thickness of the layer increases, its surface energy decreases and the driving force of
dewetting also decreases.[79] Figure 2.8 shows a schematic of dewetting of deposited thin film
to nanoparticles.
Figure 2.8 Schematic illustration of dewetting of deposited thin film to nanoparticles. Inspired
by ref.[79]
Localized defects (e.g., impurities, dislocations, holes, grain boundaries) on metal film and
substrate are the suitable places to initiate dewetting. Using a pre-patterned substrate (like selfordered TiO2 nanotubes) enables a desirable and controlled pattern to be achieved by dewetting.
Thermodynamics
The minimization of energy as the driving force of dewetting can be explained by YoungLaplace equation as follows:[79]
γs = γi cosθs + γf cosθv
(2.11)
where γs is the energy of the interface between substrate and vapor, γi is the energy of the
interface between metal film and substrate, γf is the energy of the interface between metal film
and vapor and θs and θv are the contact angles (Figure 2.9). The metal layer is stable and wets
the substrate if the θs and θv are zero. In cases where the contact angle is not zero, the metal
16
2. Background and state of the art
layer is not stable and in order to minimize its surface energy when heated, it breaks into islands
which is referred to dewetting.
Figure 2.9 Schematic of the equilibrium shape of an island on a substrate. Inspired by ref.[80]
Phenomenology of dewetting
In general, the dewetting process is divided into three different stages: (i) initiation of holes
(hole formation), (ii) growth of hole (hole growth), (iii) formation of islands. The dewetting
process begins with the formation of holes at the surface defects. Metal films leave the
dewetting area under the capillary energy. When the holes are initiated on a metal surface, the
capillary energy retracts the holes’ edge and the holes will develop. The rate of hole
development is controlled by the rate of its edge retraction. At the final stage of dewetting,
unstable cylindrical rods form sphere particles by decreasing their free energy. This is generally
known as Rayleigh instability.[79] The instability of cylinders versus perturbation is reported by
Nichols and Mullins. They have shown that the cylinder rods are unstable when the perturbation
wavelength is greater than the rod circumference.[81] Figure 2.10 shows three stages of the
deweting process.
Figure 2.10 Schematic of the three stages of dewetting process. Inspired by ref.[79]
Simplicity and controllability of solid state dewetting motivated us to use this approach for
decorating oxide surfaces with metal nanoparticles in order to produce an efficient photocatalyst
17
2. Background and state of the art
system to improve the photocatalytic activity of the nanostructured TiO2 semiconductor. In the
following chapter, the details of performance improvement are discussed.
2.2 Energy conversion with nanostructured TiO2
2.2.1 Solar energy conversion to electricity
Nowadays, environmental pollution and energy shortages have forced societies to
substitute fossil fuels with clean and renewable sources of energy. The sun is the main source
of renewable energy on the earth. Solar energy can be converted into other forms of energy
such as electricity, heat and chemical fuel.[82] Light harvesting technology such as photovoltaic
(PV) devices[83] which convert the solar photons to electricity, concentrating solar thermal
power (CSP)[84] which converts the solar power to heat, and photocatalysts[85] which convert
the solar power to the useful chemicals have been considered main sources for solar energy
conversion. However, at present global demand for energy is still overwhelmingly supplied
from solar energy compared to fossil fuels. The low consumption of solar energy in spite of its
high potential is due to its cost and conversion efficiency.
Solar-based devices are used to convert solar energy to electricity in order to power
different devices. Photovoltaic cells (e.g., solar cells) are utilized for the converting process.
They are available in various types and ratings according to the application. The solar cells are
typically fabricating from materials that illustrate the photovoltaic effect. Silicon-based solar
cells, which are the most efficient and commercial type of solar cells, deliver an efficiency of
about 26%. Thin-film solar cells which mainly consist of semiconductor materials such as
CdTe, GIGS and GaAs also show promising performance.[86] However, over the years many
efforts have been made to develop new materials in photovoltaic cells in order to improve their
performance and stability. Progress in materials and manufacturing techniques has played a key
role in this development. Still, there are some challenges to overcome before photovoltaics
could supply efficient, clean, and abundant energy. In this thesis, one of these challenges for
the efficiency enhancement of new generation of photovoltaic cells, perovskite solar cells, is
introduced and attempts to solve this issue are discussed.
Development of perovskite solar cells
Perovskite is a group of minerals with the crystal structure of calcium titanate (CaTiO3).
This was discovered for the first time in 1839 by German mineralogist Gustav Rose. In general,
perovskite materials own the formula of ABX3, where A and B are metal cations and X is a
18
2. Background and state of the art
halogen anion with the tetragonal or cubic crystal structure. Figure 2.11 shows the cubic crystal
structure of perovskite.
Figure 2.11 Crystal structure of perovskite. Adapted with permission from ref.[87]
The early studies on semiconductor perovskite compounds were focused on perovskite
consisting of inorganic cations and halides. Later, organic cations were incorporated in
perovskite structure leading to the hybrid halide perovskite (CH3NH3PbX3) compound with the
same structure of perovskite which showed promising physical properties including a high
absorption coefficient, long carrier diffusion length, high tolerance for defects and low
cost.[86,88,89] The first perovskite solar cells with organic-inorganic metal halide formula were
reported by Miyasaka et al. in 2009.[90] At the first stage, perovskites were used as a sensitizer
in photoelectrochemical cells with a power conversion efficiency (PCE) of 3.8%. Later, much
research was devoted to investigating different types of perovskite solar cells and improving
their performance which led to the rapid progress in a new generation of photovoltaic
technologies in a very short time. Today, the efficiency of perovskite solar cells (PSCs) have
reached a new record PCE of 25.6%.[91]
The perovskite solar cell typically contains of a FTO electrode, electron transport layer
(ETL), perovskite light absorber, hole transport layer (HTL) and a top electrode. They are
mostly classified into two groups: planar heterostructure and mesoporous structure.[92–94] Figure
2.12 shows the schematic of planar heterostructure and mesoscopic structure of PSCs. The
mesoporous structure consists of a mesostructured metal oxide as an ETL, the perovskite light
absorber penetrates into the metal oxide scaffold and subsequently results in light-receiving
area improvement. In planar structure, due to the two separate interfaces between the perovskite
light absorber and electron transport layer and hole transport layer, the photoinduced electrons
and holes can be extracted quickly by ETL and HTL respectively.[92] The planar structure can
19
2. Background and state of the art
be divided into two types; n-i-p (regular) or p-i-n (inverted)[95] (Figure 2.12 a). Semiconductors
metal oxides such as TiO2, SnO2, ZnO, WO3, Al2O3 have frequently used ETLs materials in
PSCs. TiO2 is the most common metal oxide that has been used in both mesoporous and planar
architecture as a scaffold for transferring photogenerated electrons.
Figure 2.12 Typical structure of PSC (a) planar, (b) mesoporous. Inspired by ref.[96]
The basic working principle of PSCs involved the absorption of light by perovskite light
absorber materials and created electron-hole pairs in a perovskite layer (Figure 2.13). The
photogenerated carriers move to the interface contacts and electrons transfer to ETL and holes
transfer to HTL (green arrow).
Limitations. However, during charge transportation, some undesirable processes can occur
such as a recombination of photo-induced electrons and holes, recombination because of direct
contact of ETL and HTL, and back carrier transfer to the interface with both ETL and HTL
(black arrow). Overcoming charge extraction obstacle at the interfaces leads to charge carriers
transporting through the interfaces and finally being extracted by the electrodes.
20
2. Background and state of the art
Figure 2.13 Schematic of working principle of PSC. Inspired by ref.[86]
The carrier lifetime or recombination lifetime is the average required time for extracting
the carrier before recombination. The carrier lifetime is one of the key parameters in the overall
performance of perovskite solar cells. The charge carrier decay can be described as
follows:[86,97]
d𝑛 /d𝑡 = - k3n3 - k2n2 - k1n
(2.12)
Where n is the photoinduced carrier density, K1, K2 and K3 are the rate constant related to
monomolecular
recombination
(recombination
associated
with
traps),
bimolecular
recombination (radiative) and Auger recombination respectively. The whole charge-carrier
recombination rate is defined by these three distinct recombination mechanisms. The trapassisted recombination and Auger recombination are considered as non-radiative
recombination. The trap-assisted recombination largely depends on the density of the trap states
and energy depths, whereas the Auger recombination is largely dependent on electronic band
structure.
21
2. Background and state of the art
Figure 2.14 (a) Schematic of charge recombination mechanisms (b) schematic of interfaceassociated recombination losses, trap states in perovskite layer (type I), trap states in perovskite
layer + unsuitable band alignment (type II), recombination due to electron back transfer (type
III), trap states in electron transport layer (type IV). Inspired by refs.[86,97]
Selecting the ETL and HTL with suitable band alignments with the conduction band and
valence band of perovskite light absorber is necessary to decrease the charge recombination and
increase charge extraction in PSCs and consequently achieve high performance. In addition,
ETL/perovskite and HTL/perovskite interfaces play an essential role in improving charge
extraction and decreasing charge recombination. However, overcoming the efficiency losses at
both ETL/perovskite and HTL/perovskite interfaces has been a concern of many researchers
and still needs to be studied. In the following sections, engineering the TiO2 ETL/perovskite
interface and the influence on cell performance is discussed.
TiO2 as an electron transport layer in PSCs
The electron transport layer (ETL) or hole blocking layer as one of the major elements in
PSCs plays a key role in efficient electron extraction and transportation and suppressing charge
recombination. Several ETL have been discovered, mainly made of metal oxides such as TiO2,
SnO2, ZnO, Al2O3, ZrO2. TiO2 is the most frequent ETL material. To date, TiO2-based ETL
with compact, mesoporous, or nanostructured scaffolds (e.g nanotubes, nanowires and
22
2. Background and state of the art
nanorods, nanosheets) has been reported by different research groups.[98–102] In general,
nanostructured ETLs have been found to be more efficient for charge collection and
transportation. A nanoarchitecture with higher and superior contact with the perovskite layer
provides a short electron transport path and therefore fast electron injection. In other words,
TiO2 in nanostructure forms not only provides a large surface area in contact with the light
absorber layer but also improves optical absorption or charge transport efficiency.[103]
TiO2 Nanorods array as ETL
A one-dimensional TiO2 nanorods array is an ideal option for ETL in PSCs by providing a
direct path for electron transporting and a suitable scaffold for the infiltration of the perovskite
light absorber.[69] Another advantage of single crystalline nanorods in comparison with
polycrystalline nanostructured metal oxides such as nanotubes and nanowires is that single
crystalline nanorods possess fewer grain boundaries and thus fewer defects. These defects act
as traps and block the transfer of electrons.[101] Therefore, TiO2 nanorod arrays have been widely
employed as ETL in perovskite solar cells.[67,104–106]
Limitations. However, TiO2 ETLs suffer from high surface defect states that lead to
recombination at the interface of ETL and perovskite light absorber resulting in performance
losses. The interface recombination is still one of the main obstacles to achieve high performing
device and needs to be alleviated by an effective strategy. In addition, trap states at the TiO2
ETL/perovskite layer interface increase device hysteresis. Therefore, an effective approach for
interface modification of TiO2 NR arrays with perovskite light absorber is essential in order to
decrease charge recombination and to improve charge extraction and collection efficiency.
Interface engineering of ETL/perovskite light absorber
Typical PSCs possess six major interfaces: (1) transparent conductive oxide (FTO/ITO)/
ETL interface; (2) ETL/perovskite light absorber interface; (3) perovskite grain boundaries; (4)
the perovskite light absorber/HTL interface; (5) HTL/top electrode interface; and (6) top
electrode/atmosphere interface.[107] The interface between ETL and perovskite light absorber is
one of the areas this thesis focuses on and optimizing these interface properties is the key factor
in the efficiency of the devices.[13] Well engineered interfaces of the layers in order to obtain
correct energy level matching between layers is a critical parameter for efficient PSCs.
Photogenerated electrons are transferred to ETL across the ETL/perovskite interface and
photogenerated holes are transferred to HTL across the HTL/perovskite interface. Electron-hole
recombination can take place at these two interfaces due to the interfacial traps or poor charge
23
2. Background and state of the art
extraction ability of the ETL or HTL.[108] The accumulation of charge at ETL/perovskite or
HTL/perovskite has a considerable impact on the hysteresis of the device. Therefore,
engineering the interfaces of perovskite with transport materials is essential in order to obtain a
high-performance device.
In the following, some strategies for ETL/perovskite interface modification are discussed.
Energy-level alignment at the ETL/perovskite interface is essential for charge mobility
dynamics and charge recombination and as a result the overall cell performance. The energy
barrier at the interface of the ETL with the perovskite layer is one of the reasons for inefficient
charge extraction and increasing electron-hole recombination. Improving the extraction of
electrons by optimizing the band offsets at the TiO2 ETL/perovskite has been reported by many
groups.[109–111] To achieve this, a functional approach can either be designing ETL with a
suitable conduction band or introducing an intermediate layer between ETL and perovskite
layer such as self-assembled monolayers (SAMs) and PCBM.[112,113] Tao et al. demonstrated
that deposition of PCBM intermediate layer on TiO2 thin layer leads to a more efficient charge
extraction and improvement in cell efficiency from 14.4% to 17.9%.[114] Guo et al. employed
PC61BM, C60-SAM and a combination (PC61BM+C60-SAM) to modify the interface of the TiO2
compact layer, and they found that these interlayers can improve interface contact and increase
cell performance as well as device stability.[115]An alternative technique for manipulating the
energy alignment at the ETL/perovskite interface is doping ETL or perovskite materials.[116–118]
Another reason for preventing efficient charge transfer is the trap states at perovskite, ETL,
HTL and their interfaces. In general, the recombination process is dominant when the trapping
rate is greater than the injection rate. Passivation of defect states and traps of perovskite layer
and interface with ETL and HTL is a powerful approach to improve charge transfer across the
interfaces. Passivation of defects at TiO2 ETL/perovskite interface can also be conducted by
inserting a thin surface modifier like phenyl-C61-butyric acid methyl ester (PCBM) or
poly(vinylcarbazole) (PVCz) between TiO2 ETL and the perovskite light absorber. It has been
widely reported that introducing a PCBM layer into PSC devices has a significant effect on
reducing or eliminating hysteresis by decreasing ionic migration. Okada et al. reported
PVCz/PCBM modification for passivation of TiO2 mesoporous layer. They introduced
poly(vinylcarbazole) (PVCz) as a scaffold of PCBM for TiO2/perovskite interface passivation.
For the device modified with PVCz/PCBM, electron transport to the ETL is increased and
charge recombination at TiO2/perovskite interface is decreased. [119]
24
2. Background and state of the art
Figure 2.15 shows the schematic of passivation of traps at TiO2 ETL/perovskite interface
and energy diagram of PSC with an intermediate PCBM layer between TiO2 ETL and
perovskite film.
Figure 2.15 (a) Schematic of passivation of traps at TiO2 ETL/perovskite interface by
PCBM, (b) energy diagram of PSC. Inspired by ref.[112]
Additionally, deposition of another TiO2 film to cover the primary surface such as chemical
deposition using different precursors e.g. TiCl4,[113,120] or deposition of ultrathin TiO2 layer
using a technique like atomic layer deposition[66] can passivate the surface defects of TiO2 film.
The role of TiCl4 nanoparticles treatment passivates the defects by filling gaps and traps in the
TiO2 origin layer. Moreover, this treatment forms a conformal thin layer not only on the TiO2
layer but also at the free space of FTO/TiO2 ETL interface which increases the adhesion of the
TiO2 layer on FTO, enhances the morphology of the TiO2 film and acts as a hole blocking
layer.[121–123] Sun et al. applied TiCl4 treatment to modify the TiO2 layer in PSC. They showed
the role of TiCl4 nanoparticles in the passivation of surface defects as well as in formation of
nanostructure on a TiO2 surface to prevent direct contact of perovskite with FTO and improve
TiO2/perovskite contact.[120] Ma et al. investigated the effect of dual modification of the TiO2
compact layer with TiCl4 and PC60BM on the perovskite device performance. According to
25
2. Background and state of the art
their results, a combination of PC60BM and TiCl4 improves the efficiency from 10.8% for a
control device to 16.4% for a modified device.[113] Figure 2.16 shows the blocking effect of
TiCl4 treatment.
Figure 2.16 Schematic illustrations of TiO2-FTO junctions without and with TiCl4
treatment. Inspired by ref.[124]
Some studies have also investigated modification approaches for TiO2 nanorod array as
ETL in perovskite devices. Mali et al. reported passivation effect of the ultrathin TiO2 layer
deposited by ALD on hydrothermally grown TiO2 NR layers. Their results showed that ALD
passivation greatly affected device performance parameters such as VOC and JSC.[125] Li et al.
worked on modification of TiO2 NR arrays with SnO2-Sb2O3 nanoparticles by the spin coating
technique. The modification led to an increase in photocurrent density and fill factor due to
higher electron mobility of the modified solar cells compared to the unmodified solar cells.[105]
Improvement of efficiency and stability of perovskite solar cell by chemical bath deposition of
CdS shell layers on TiO2 NR arrays is demonstrated by Liu et al. The performance enhancement
is ascribed to recombination suppression because of oxygen vacancies passivation at the TiO2
surface and prevent their contact with perovskite layer. Moreover, the charge transfer improved
due to the formation of type-II structure of CdS@TiO2. [126] An increased performance of
perovskite solar cells by decorating TiO2 NR arrays with CuInS2 quantum dots is reported by
Gao et al. The enhancement is ascribed to the role of CuInS2 in light harvesting and improving
electron extraction from perovskite light absorber.[127] Zhang et al. reported deposition of
ultrathin anatase TiO2 film on TiO2 nanorods ETL using a solution technique in order to create
type-II band alignment and facilitate charge mobility at the TiO2 ETL/ perovskite interfaces.
Then, a self-assembled monolayer (SAM) was deposited on the surface of TiO2 nanorods
modified with ultrathin anatase TiO2 film to improve chemical combination of TiO2 ETL and
perovskite layer for further interface contact enhancement.[128]
26
2. Background and state of the art
However, a few works have reported multitreatment of TiO2 NRs ETL/perovskite interface
and the synergistic effects that result from this multitreatment. It is expected that compared to
a single treatment of ETL, dual or multitreatment provides synergistic effects that lead to better
modification of ETL and further efficiency enhancement of PSCs. The synergistic effects of
TiO2 surface modifications facilitate charge mobility across TiO2 ETL/perovskite interface. We
aimed to modify TiO2 NRs ETL/perovskite interface with dual treatments of TiCl4 treatment
combined with PC61BM monolayer. It has been shown that TiCl4 treatment improves the surface
contact of TiO2 NRs ETL with the perovskite layer, therefore increasing charge transfer
efficiency. At the same time, the PC61BM monolayer passivates trap states at TiO2 NRs
ETL/perovskite interface, improve the cell performance. Under dual ETL modification, cell
efficiency increased remarkably from 14.2% to 19.5% and device hysteresis considerably
suppressed. Therefore, our results demonstrate the synergistic effects as a result of proper dual
modification of TiO2 ETL which lead to cell efficiency improvement and device hysteresis
reduction.
Nanocrystalline TiO2 thin film as ETL
The TiO2 thin films produced by different methods are effective electron transport layers
(blocking layers) in planar heterostructure type of perovskite solar cells. The properties of the
TiO2 thin layer, such as thickness, transparency and surface roughness, play important roles in
cell performance. The optimal thickness of the ETL depends on the method that is used to
prepare the ETL. However, ETLs should be deposited uniformly, pinhole free and provide
blocking properties. The effect of crystallographic phase of TiO2 ETL on the efficiency of
perovskite solar cells has already been reported. Zhu et al. have shown the increase of the device
performance when anatase TiO2 is used as ETL compared to rutile TiO2.[129] Yella et al. reported
better performance of nanocrystalline rutile TiO2 in comparison with planar anatase TiO2 in
perovskite solar cells. Better performance of nanocrystalline rutile TiO2 is attributed to the large
interface area of nanocrystalline rutile TiO2 with perovskite which improves electron extraction
at the interfaces.[130] In another work by Wang et al., TiO2 anatase ETL and TiO2 rutile ETL
were prepared by spin coating and chemical bath deposition respectively. Their results showed
better performance of the device that is fabricated with Rutile TiO2 than anatase TiO2 due to
the higher conductivity of rutile ETL and better interface of rutile TiO2 that was deposited by
chemical bath deposition with perovskite light absorber.[131]
27
2. Background and state of the art
However, because various techniques were employed to synthesize anatase and rutile TiO2,
it may not result in ETLs with comparable morphological and structural properties. An
important criterion for proper comparison of functionality of different crystal phases TiO2 in
PSCs is that ETLs possess comparable morphological and structural properties. In this thesis,
we have shown that for a logical comparison of TiO2 ETL of various crystal phases and the
effect of ETL crystal phases on the cell performance, differences in other properties of the layers
such as surface morphology, thickness, crystallite size and roughness should be considered. We
wanted to point out that a proper comparison requires that ETLs (i) produce using the same
technique, and (ii) have comparable morphological or micro-structural properties. Then, we
reported that the efficiency of the device is not greatly affected by the crystal phase structure of
the TiO2 thin ETL.
2.2.2 Solar energy conversion to valuable chemicals
Solar energy can be converted into chemical substances by green planets via natural
photosynthesis or via artificial photosynthesis systems that have been designed by people. A
very low amount of solar energy is consumed by human activities despite the availability and
abundance of solar light. Natural photosynthesis to produce biomass is the largest application
of solar energy.[82] Fuels such as methane (CH4), hydrogen (H2) or alcohol can be produced
from solar energy and stored as an alternative to fossil fuels.
There are two methods of solar energy conversion to solar fuels, direct and indirect
processes. In the direct process, solar light is converted to chemical fuels without employing
extra energy conversions whereas, in an indirect process, solar energy first converts to another
type of energy like electricity and then this type of energy then produces solar fuels. In
comparison to the indirect process, the direct process is more efficient since it eliminates
intermediate steps.[132]
Artificial photosynthesis is a direct process and its aim is to design and develop catalysts
that are capable of directly oxidizing (splitting) water without using another form of
intermediate energy.[132] Therefore, many research focused on developing different catalysts to
catalyze the oxidation/reduction reactions in artificial photosynthesis.
Hydrogen is the most frequently studied solar fuel since it can be produced by oxidizing
water and reducing carbon dioxide in an electrochemical process.[133] Hydrogen is considered
as the main energy resource of the future and as a replacement for fossil fuels. This work aimed
to develop an efficient catalyst system to produce hydrogen at a high rate in photocatalytic H2
evolution from methanol-water mixed solutions.
28
2. Background and state of the art
Photocatalysis
Photocatalysis is the process of speeding up a photoreaction by using a catalyst. It is
considered a promising method for producing clean energy since this method is
environmentally protection, sustainable and economically favorable. In general, the
photocatalysis process consists of two main reactions: oxidation by photoinduced holes and
reduction by photoinduced electrons.
Photocatalysis was first mentioned in 1911 by German chemist Alexander Eibner in the
context of the effect of zinc oxide illumination on inorganic and organic pigments on the
bleaching of Prussian blue. In 1938 the photosensitizer function of TiO2 for dye bleaching in
the presence of oxygen was discovered by Kitchener and Goodeve.[134] They found that oxygen
species can be formed on a TiO2 surface and cause blotching of organic chemicals when TiO2
absorbs ultraviolet light. This achievement marks the first time the fundamental characteristics
of heterogeneous photocatalysis.[31] Later, Fujishima and Honda[135] published an article in
Nature on the photoelectrolysis of water to produce hydrogen and oxygen in an electrochemical
cell consisting of TiO2 and Pt electrodes. Figure 2.17 shows the schematic of their
electrochemical setup for water splitting. Since their ground-breaking work, photocatalytic
water splitting for producing H2 using TiO2 has been widely studied. In their setup, the UV light
is absorbed by TiO2 electrode and electrons transfer through the anode (TiO2 electrode) to the
cathode (Pt electrode) and reduce water to produce hydrogen.
Figure 2.17 Schematic of electrochemical cell of Fujishima and Honda. Adapted with
permission from ref.[135]
Among various photocatalyst materials, TiO2 has proved to be a special candidate because
of its advantages such as the alignment of the conduction band position of TiO2 with regard to
29
2. Background and state of the art
electrochemical potential of hydrogen production, low cost, availability, high resistance to
corrosion and photocorrosion. Nanostructured TiO2 such as nanotubes, nanorods and nanowires
have demonstrated great potential in photo energy conversion. Among them, TiO2 nanotubes
have attracted much attention due to their excellent charge transport properties and orthogonal
electron−hole pairs separation. In addition, highly ordered TiO2 structures are an ideal platform
for further deposition of co-catalyst particles. This means that a highly ordered TiO2 nanotubes
scaffold provides not only an ideal geometry for photocatalytic reactions, but also selforganized nanocavities for highly controlled deposition of co-catalysts. Therefore, the
photocatalytic properties of a self-organized TiO2 nanotube array and he use of this type of
nanostructured scaffold for energy conversion have been the basis for much research.[21,136–138]
Limitations. However, the photocatalytic application of TiO2 under visible light is limited
due to its large band gap. It needs UV light for photoactivation and only a low amount of the
solar spectrum is ultraviolet light. Moreover, in an electroless process, TiO2 alone has a poor
performance for photocatalytic reactions due to fast recombination of electron-hole pairs
leading to a decline in photo efficiency. According to the literature, in the presence of
appropriate co-catalysts, the photocatalytic activity of a TiO2 semiconductor can be largely
improved. Hence, using suitable co-catalysts is a straightforward approach to overcome the
limitations of a TiO2 catalyst in photocatalytic reactions to obtain a reasonable H2 production
rate. Achieving an efficient photocatalysis process requires an understanding of the parameters
that control the photocatalysis kinetics.
Mechanism of photocatalysis
In semiconductors where photogenerated electrons are promoted from the valence band to
the conduction band, a chemical reaction can take place. When a photon with an energy similar
to or greater than the bandgap of the semiconductor like TiO2 is irradiated, the electrons from
the valence band of the catalyst are excited to the conduction band and create holes in the
valence band. The photogenerated electrons and holes can transfer to the photocatalyst surface
and then interact with electron donor-acceptor complexes. It is feasible for the electrons and
holes to recombine either at the surface and interface trap sites or in bulk. The recombination
process results in releasing energy as heat.[139]
TiO₂ + ℎ𝜈 → e- + h+
(2.13)
e- + h+ → heat
(2.14)
30
2. Background and state of the art
This electrons-holes recombination process is not favorable for an efficient photocatalysis
reaction. For this reason, increasing the lifetime of photogenerated charges and enhancing the
separation of electron-hole pairs could play a key role in improving the overall efficiency of the
photocatalysis process. Figure 2.18 shows the principle of photocatalysis process.
Figure 2.18 Schematic diagram of photocatalysis process. Inspired by ref.[139]
In general, there are two types of photocatalysis reaction: heterogeneous and homogenous
photocatalysis. Homogeneous photocatalysis occurs when photocatalysts and reactants exist in
the same phase. In heterogeneous catalysis, the catalysts and reactants are in different phases.
Heterogeneous photocatalysis includes a wide range of reactions such as dehydrogenation,
hydrogen evolution, oxidations, water detoxification, removing gaseous pollutants and metal
deposition. Transition metal oxides and semiconductors like TiO2, which have special
properties, are the most frequently used heterogeneous photocatalysts.[31]
Photocatalytic hydrogen production
Photocatalytic H2 production from water is a promising approach to produce clean energy
as a substitute for fossil fuels. Extensive progress on improvement of efficiency has been made
in order to produce hydrogen on a large-scale. Nowadays, H2 is the most important industrial
gas and can be stored as useable energy. It has different applications in petroleum refining, the
electronic industry, as well as in aerospace, food industry and other applications.[140]
31
2. Background and state of the art
The H2 evolution reaction (HER) can be divided into two parts: oxidation for oxygen
evolution and reduction of water for hydrogen generation, which is demonstrated as
follows:[140][141]
H2O → 2H2 + O2
∆E0 = 1.23
(2.15)
H2O → 4H+ + 4e- + O2
E0 = +1.23 V vs. NHE, pH = 0
(2.16)
4H+ + 4e- → 2H2
∆E0 = 0 V vs. NHE, pH = 0
(2.17)
Figure 2.19 shows the photocatalytic H2 evolution process. In general, photoelectrolysis
consists of three parts: (i) photon absorption with energy higher than the bandgap of
semiconductor, which leads to the formation of electron-hole pairs, (ii) carrier separation and
migration to the surface of semiconductor photocatalyst, and (iii) reaction of charge carriers and
water resulting in O2 or H2 evolution. The kinetics of these three parts and thermodynamic
balance determine the efficiency of the photocatalysis.[140,142]
As illustrated in Figure 2.19, in H2 evolution in the presence of photocatalyst, the
conduction band (CB) position must be more negative than the reduction potential of H2O/H2,
and the valence band (VB) position must be more positive than the oxidation potential of
H2O/O2.[143,144]
Obviously, to initiate photoelectrochemical water splitting by solar light, the
semiconductor must absorb a considerable amount of visible light. A transition metal oxide like
TiO2 provides not only a suitable band edge potential for hydrogen evolution reaction and
oxygen evolution reaction but also possesses good chemical and photochemical stability.
32
2. Background and state of the art
Figure 2.19 Schematic diagram of photocatalysis H2 evolution. Inspired by ref.[140]
Undesirable electron-hole recombination processes are the major inhibitor for the highly
efficient photocatalysis HER and thus the semiconductor surface needs to be modified to obtain
reasonable amount of hydrogen. Many techniques were carried out in order to improve the
separation of electrons and holes sand decrease recombination, such as doping,[145,146] sacrificial
agents[147,148]
and
heterojunction
photocatalysts.[149,150]
Among
these
techniques,
heterojunctions were shown to be the most effective approach for improving electron-hole
separation and consequently achieving better photocatalysis performance.
Heterojunction photocatalyst
To enhance the photocatalytic efficiency, the strategy that has normally been used is the
process of forming a semiconductor heterojunction by combining a semiconductor with other
cocatalysts such as semiconductors, noble or non-noble metals and so on. Typically, when two
components with different Fermi levels are coupled, the carrier can flow at the interface of the
components until their Fermi levels align at the same level. This synergistic effect can increase
the separation of electrons and holes.[151]. A heterojunction is described as the interface of two
distinct semiconductors that have uneven band structure, which can cause band alignments.[152]
In general, heterojunction photocatalysts can be divided into four categories: (i) semiconductorsemiconductor heterojunction; (ii) semiconductor-metal heterojunction; (iii) semiconductorgraphene heterojunction; (iv) multicomponent heterojunction.[153] Figure 2.20 shows the charge
carrier transfer on a single photocatalyst and on a heterojunction photocatalyst.
Figure 2.20 Schematic diagram of photogenerated electron and hole transfer on a single
photocatalyst and a heterojunction photocatalyst. Inspired by ref.[151]
33
2. Background and state of the art
Noble metals
Noble metals such as platinum (Pt), palladium (Pd), ruthenium (Ru), silver (Ag) have been
extensively used as a co-catalyst in photocatalysis reactions. Due to the lower energy level of
the Noble metals’ Fermi level compared to the conduction band of TiO2, photogenerated
electrons can transfer and be caught by the metals.
The ability of a noble metal as co-catalysts for the photocatalytic H2 evolution is determined
by the noble metal’s work function Φ (the mini mum thermodynamic work to remove an
electron from the Fermi level to vacuum). The difference between the work function of the
noble metal and TiO2 leads to the formation of the electronic potential barrier which is called
Schottky barrier. The formation of the Schottky barrier promotes electron migration to noble
metal and subsequently enhances H2 generation rate (Figure 2.21). Thus, in photocatalytic H2
generation, incorporating noble metals as co-catalysts facilitates electron movement at the cocatalyst/TiO2 interface by forming a junction to TiO2. In addition, Pt is able to enhance the
hydrogen recombination rate (H+ → H2). Pt is the most commonly used co-catalyst for
photocatalytic H2 evolution because of its largest work function (ΦPt = 5.93 eV, , ΦAu = 5.31
eV, ΦAg = 4.74 eV, ΦPd = 4.6 eV).[57,59,154] Yoo et al. decorated TiO2 nanotubes with Pt
nanoparticles and used this structure as a photocatalyst for H2 evolution reaction.[57] Nguyen et
al. produced a photocatalyst based on suspended Pt nanoparticles on the surface of highly
ordered TiO2 nanotubes to enhance photocatalytic H2 production performance.[38] The
photocatalytic activity of TiO2 nanotubes selectively decorated with Pt nanoparticles was
reported by Liu et al.[155] Many other studies have set out to discover the role of the most
promising co-catalyst, Pt, in photocatalytic H2 production improvement.[156–158]
However, a resource shortage and the high price of noble metals pushed the investigation
towards the non-noble and cost-effective co-catalysts. Thus, the necessity for alternative
candidates such as non-noble metals in the catalysis application has been widely investigated.
34
2. Background and state of the art
Figure 2.21 Schematic diagram of the co-catalyst effect. Inspired by ref.[159]
Non-noble metals
Non-noble metals are abundant, cost-effective and some of them have properties similar to
noble metals. Moreover, they show a high work function and good conductivity (Table 2.2).
Therefore, the comparable efficiency of non-noble metals in photocatalysis application in place
of noble metals has aroused the interest of research groups.
Among different transition metals, copper (Cu) is a promising candidate due to high
conductivity, low cost, plasmonic properties and favorable activity in photocatalysis [160]. Like
noble metals, Cu has a large work function (4.65 eV) and it is able to capture electrons from the
conduction band of a semiconductor like TiO2 and improve the charge mobility at the Cu/TiO2
interface and accelerate the reduction reaction of the proton.
In general, when Cu nanoparticles are loaded on the semiconductor surface, they are
covered by thin oxide shells which form when the nanoparticles are exposed to air. The oxide
shells are typically Cu2O or CuO and can be photoreduced to Cu in the photoelectrochemical
process.[161,162] The Cu in the metallic state then enhances the photocatalytic activity.
35
2. Background and state of the art
Table 2.2 The work function and conductivity of metals.[161]
Work
Metal
function
(eV)
Electrical conductivity
(×107 ) /S m-1
Pt
5.65
0.94
Au
5.1
4.17
Ag
4.26
6.31
Ni
5.15
1.46
Cu
4.65
5.7
Co
5.0
1.51
Fe
4.5
1.03
Al
4.28
3.77
Bi
4.22
0.10
Ti
4.33
0.24
Cd
4.08
1.46
The combination of transition metals like Cu with noble metals such as Pt, Ag, Au to form
bimetallic or multimetallic co-catalysts has been widely investigated since they show notably
higher activity compared to monometallic co-catalysts. When a second metal is combined with
Cu, it not only reduces the oxidation of Cu and increases its stability but also the synergistic
effect between Cu and the second metal results in photocatalytic activity improvement.[60,163]
However, the proper selection of noble and non-noble metals that can have the most synergistic
effects for increasing photocatalytic hydrogen generation as well as an appropriate approach to
combine them and produce an effective co-catalysts plays an important role in the performance
of the final catalyst system. A combination of a noble metal Pt with a non-noble metal Cu to
form an efficient photocatalyst has already been studied. Chandran and Dharmalingam reported
the production of PtCu core–shell nanoparticles by successive and co-reduction method.[164]
Bernareggi showed synthesis of Cu/TiO2 and Cu-Pt/TiO2 photocatalysts for H2 generation by
flame spray pyrolysis.[165] In another work by Bernareggi et al. TiO2 powders coated with Cu
and Pt clusters produced by DC magnetron sputtering for photocatalytic H2 evolution.[166] A
two step photodeposition of bimetal Pt−Cu on TiO2 for increasing photocatalytic water splitting
was demonstrated by Liu et al.[1] Dozzi et al. reported a two step chemical deposition of Cu
nanoclusters and Pt nanoparticles on TiO2 powder for photocatalytic H2 production.[167]
36
2. Background and state of the art
However, due to the nature of these techniques like (photo-) reduction, the metal nanoparticles
are deposited in an inhomogeneous manner on the TiO2 surface.
There is no report on a dewetting-alloying approach to form bimetallic PtCu nanoparticles
as co-catalyst directly on TiO2 nanotube surface for photocatalytic hydrogen production.
Dewetting-alloying is a straightforward technique to convert single or multiple metallic
nanofilms into nanoparticles. In the case of multiple nanofilms, the proper dewetting
temperature results in alloying of metals. The size, density, and self-ordering degree of dewetted
nanoparticles can be simply controlled by adjusting the thickness of deposited primary
nanofilms.
Here, we introduced a simple and novel technique based on dewetting-alloying for the
deposition of PtCu alloy nanoparticles on the surface of highly ordered TiO2 nanotubes. To the
best of our knowledge, it was the first time this approach was used to form ordered PtCu
bimetallic nanoparticles on TiO2 nanotube arrays to co-catalyze a TiO2 nanotube array with
PtCu bimetallic nanoparticles. Our results indicate that PtCu nanoparticles with optimal
composition can remarkably improve photocatalytic hydrogen generation compared to
monometallic Pt or Cu co-catalyst. The HER enhancement is ascribed to an increased charge
transfer to reduce H+. In addition, our results show that Cu in the bimetallic PtCu co-catalyst
increase poisoning resistance of Pt by weakening the binding of intermediate species (e.g. CO)
to Pt sites during methanol oxidation process.
Sacrificial agents
Sacrificial agents (hole scavenger/electron donor) are substances that catch the holes from
the valence band and take part in oxidation reactions quickly. Introducing a sacrificial agent as
an electron donor to the photocatalysis process results in utilization of photogenerated holes
and decreases electron-hole recombination. In addition, when a hole scavenger is used in
photocatalytic H2 evolution, the reactions between the hole scavenger and photogenerated holes
proceed one-half reaction of the photoelectrochemical water splitting. Therefore, in presence of
a sacrificial agent the production of H2 can be increased.[168] Among various compounds such
as isopropanol, ethanol, methanol, sodium sulfide/sulfite, triethanolamine, lactic acid and so on,
methanol is one of the most common hole scavengers for photocatalytic H2 generation.
Therefore, in this thesis, we used methanol as a sacrificial agent for photocatalytic H2 generation
processes. The photoelectrochemical reactions during H2 generation from water in the presence
of methanol are summarized as follows:[169]
37
2. Background and state of the art
H2O (l) + h + → •OH + H+
(2.18)
CH3OH (l) + •OH → •CH2OH + H2O (1)
(2.18)
•
CH2OH → HCHO (l) + H++ e –
(2.19)
2H+ + 2e− → H2
(2.20)
HCHO (l) + H2O (l) → HCOOH (l) + H2
(2.21)
HCOOH (l) → CO2 (g) + H2 (g)
(2.22)
CH3OH (l) + H2O (l) → CO2 (g) + 3H2
(2.23)
Overall reaction:
In the presence of a sacrificial agent, the current doubling effect may occurring in the
photoelectrochemical setup results in doubling the photocurrent. Current doubling refers to the
increasing number of electrons in the conduction band of photocatalyst due to the injection of
electrons from unstable radicals which are the product of photoelectrochemical oxidation of
hole scavengers.
38
3. Scientific objectives
3. Scientific objectives
The key objective of this thesis was to develop novel and effective approaches for
modification of nanostructured TiO2 and investigate the effect of modification on solar energy
conversion efficiency. The scientific hypotheses are described as follows:
1. How effective the dual modification of TiO2 nanorod arrays ETL/ perovskite layer
interface with PC61BM monolayer and TiCl4 nanoparticle is on overall cell
performance? What is the effect of dual modification on the device photovoltaic
parameters with reference to single modifications?
2. Is the crystallographic phase composition of TiO2 thin film as ETL in perovskite solar
cell a key factor in the overall cell performance? What parameters should be considered
for a proper comparison of different TiO2 phases as ETLs in perovskite solar cells?
3. Can synergistic effects of Pt and Cu in PtCu bimetallic co-catalyst deposited on TiO2
nanotube arrays be effective in improving photocatalytic H2 production efficiency
compared to monometallic Pt and Cu co-catalyst? Is solid state dewetting an appropriate
approach to form ordered PtCu alloy nanoparticles on TiO2 nanotubes array?
4. How do hydrothermal conditions tune morphology and facet ratio of the TiO2
nanosheets grown on FTO substrate? To what extent does morphology and facet ratio
of the TiO2 nanosheets layer affect the photoelectrochemical performance?
39
4. Thesis overview
4. Thesis overview
This thesis contains a background and state of the art chapter, and four published research
studies. The background and state of the art chapter covers the most notable achievements in
the application of nanostructured TiO2 for solar energy conversion. Moreover, the limitations
of using nanostructured TiO2 in solar energy conversion devices were investigated and the
strategies for overcoming the limitations were introduced. Following is the list of published
research studies.
1. Publication I. This study was presented the improvement in the performance of
perovskite solar cells by dual modification of the interface between TiO2 nanorods
array as ETL and perovskite light absorber. We investigated the role of dual
modification of TiO2 ETL by a PC61BM monolayer and a TiCl4 treatment and their
effects on the device photovoltaic efficiency (Appendix A).
2. Publication II. The effect of the crystallographic phase structure of TiO2 thin films as
ETL in the overall cell performance of the perovskite solar cell was investigated. The
TiO2 layers were deposited by magnetron sputtering technique and the sputtering
conditions were adjusted to precisely control the TiO2 crystallographic phase. In
addition, factors that can affect proper comparison of different crystal phase structures
of TiO2 as ETL were discussed (Appendix B).
3. Publication III. The improvement of photocatalytic H2 evolution from methanol-water
solutions by PtCu alloy nanoparticles as a co-catalyst on TiO2 nanotubes was studied.
The co-catalysts were deposited on TiO2 nanotube arrays and were transformed to the
nanoparticles by solid-state dewetting. We were investigated the synergistic activity
enhancement of PtCu bimetallic co-catalyst compared to the Pt or Cu monometallic cocatalyst (Appendix C).
4. Publication IV. This study was focuses on hydrothermally grown single crystal anatase
TiO2 nanosheets on a FTO substrate. The hydrothermal conditions were tuned to
control not only the morphology but also the facet ratio of nanosheets. Then we were
40
4. Thesis overview
investigated the effect of growth parameters on the photoelectrochemical performance
of synthesized nanosheets (Appendix D).
41
5. Author contributions
5. Author contributions
The author contributions to the published manuscripts are summarized in the following. In
addition, the confirmation of contributions made by co-authors in the publications (Section 10
(3) sentence 2 FPromO Tech) are issued with this thesis.
1. Publication I. F.S is the first author and conceives of the work idea. F.S carried out the
experiments. M.A was in charge of project planning. F.S, M.A, Y.H, S.H, N.L
contributed to data interpretation. F.S drafted the manuscript and designed the figures.
M.A contributed to writing and editing. P.S and C.B supervised the project and
commented on the manuscript.
2. Publication II. F.S and M.A conceptualized the project. F.S performed the main
experiments. S.H helped in sample preparation. S.HO conducted Raman spectroscopy.
K.Z carried out AFM. F.S, M.A, N.L and C.B contributed to data interpretation. F.S
provided the first draft designed the figures. C.B, N.L and S.HO contributed to writing
and editing and provided critical comments. C.B and P.S supervised the project.
3. Publication III. M.A and F.S conceptualized and generated the idea. F.S performed
anodization, solid-state dewetting, photocatalytic measurements and UV-Vis
measurements. E.W performed XPS. P.G and A.M conducted XANES. F.S wrote the
first draft and designed figures. M.A and F.S contributed to data interpretation and
editing of the manuscript. P.S supervised the project.
4. Publication IV. The main idea of the study is conceived by F.S. F.S carried out the
main experiments (Hydrothermal treatment, electrochemical experiments, Impedance
measurements). N.D conducted TEM. The manuscript is written by F.S and P.S. All
authors discussed the results. P.S supervised the project.
42
6. Summary
6. Summary
Within the present work, different techniques such as hydrothermally synthesizing TiO2
nanorods, TiO2 nanotubes by anodization, hydrothermally grown TiO2 nanosheets, sputtering
of TiO2 thin films, TiO2 surface modification by TiCl4 and PC61BM monolayer and TiO2
decoration with dewetted PtCu nanoalloy particles were studied in order to use nanostructured
TiO2 as ETL in perovskite solar cells and as catalyst in photocatalysis H2 evolution.
Firstly, TiO2 nanorods were hydrothermally grown on the surface of FTO as ETL in MAPIbased perovskite solar cells. One-dimensional TiO2 nanomaterials like nanorods have been
extensively utilized as nanostructured ETL because they supply the directional electron
transporting path and a large surface area. These advantages result in the improvement of
ETL/perovskite interfacial contact and increase the charge diffusion. However, another key
parameter to enhance device performance is still the modification of the ETL/perovskite light
absorber interface. Here, we employ dual modification of TiO2 nanorods by a TiCl4 treatment
and a PC61BM monolayer in order to passivate traps and defect states of the TiO2 ETL surface.
The results show that dual modification of TiO2 ETL interface with perovskite
considerably increases the efficiency of the cell from 14.2% to 19.5% and decreases hysteresis.
The performance improvement can be ascribed to the more efficient charge movement across
the interface of ETL with a perovskite layer as a result of passivation of trap and defect states.
Next, the effect of crystallographic phase structure of TiO2 as an ETL in perovskite solar
cells is investigated. TiO2 thin films were synthesized with the magnetron sputtering method
under specific conditions to control the crystallographic phase of the final layers. Anatase and
rutile TiO2 thin film with a similar morphology and structure were deposited on the FTO
substrate. We explored the structural and optical properties of both layers in order to obtain
proper and systematic comparison. Both layers show almost similar morphology, roughness,
work function and optical properties.
Our results, demonstrate that the crystallographic phase structure of TiO2 ETLs cannot
greatly affect the overall efficiency of the perovskite solar cells. The devices fabricated with
anatase and rutile TiO2 ETLs show an efficiency of 18.4% and 17.7% respectively which are
quite comparable. We intend to point out that for a reliable comparison between the
43
6. Summary
performances of TiO2, other properties of the layers should be similar which can be obtained
by controlling the deposition process.
In another approach, the performance of PtCu alloy nanoparticles as a co-catalyst on TiO2
nanotubes for H2 production is evaluated. To produce TiO2 photocatalyst decorated with PtCu
bilayer, Pt and Cu nanofilms with different thicknesses were deposited on TiO2 substrate, and
then the application of a proper thermal treatment leads to the formation of PtCu bimetallic
nanoparticles via solid-state dewetting. Physicochemical differences between Pt, Cu
monometallic and PtCu bimetallic co-catalysts were studied by XPS and XAS. PtCu alloying
during thermal treatment (solid-state dewetting) is confirmed by XPS and XAS.
We prepared a series of TiO2 samples decorated with Pt, Cu and PtCu nanoparticles. The
photocatalytic activity of different samples was investigated for H2 generation from water in
the presence of methanol as a hole scavenger. The results indicate that the activity of a sample
decorated with PtCu co-catalyst is ~ 50 times and ~ 4 times higher than the activity of TiO2
decorated with Cu and Pt respectively. This improvement can be attributed to the synergistic
effect of Pt with Cu. Alloying Pt with Cu leads to an increase in electron density on Pt due to
the higher electronegativity of Pt than Cu. Increasing the density of electrons improves the
electron transfer for H2 evolution. Moreover, incorporation of Pt with Cu may increase its
tolerance against poisoning from the intermediate products (e.g. CO) of methanol oxidation
which can be another reason for photocatalytic activity enhancement.
In the last part, single crystalline TiO2 nanosheets were hydrothermally grown on a FTO
substrate. We systematically tune the hydrothermal conditions such as synthesis time,
concentration of Ti and F precursors and HCl/water ratio. A wide range of nanosheets with
various morphologies and faceting (surface area (001)/surface area (101)) were produced. In
the next step, the effect of morphology and faceting on the photoelectrochemical performance
of the nanosheet films was studied. We have demonstrated that when the morphology and
faceting of the nanosheet films are optimized, the photoelectrochemical efficiency increases
significantly. The remarkable incident photon conversion efficiency (IPCE) of > 80% is
obtained for the optimized conditions. Performance enhancement originates from larger active
surface area and high ratio of (001) crystallographic facets.
44
7. Outlook
7. Outlook
This thesis provides deep insight into the importance not only of using TiO2 in specific
nanostructure forms but also appropriate modification approaches in order to obtain efficient
solar energy conversion systems. In terms of perspective, in perovskite solar cells,
multitreatment approach of nanostructured TiO2 ETL with PC61BM monolayer and TiCl4
treatment combined with more efficient light absorbers like mixed cation/mixed anion
perovskite materials has been expected to further enhance performance.
In photocatalytic H2 generation systems, bimetallic and trimetallic alloys co-catalyst on
TiO2 NTs and their effects on the improvement of photocatalytic H2 production activity would
be interesting. In trimetallic catalyst systems, by further reducing the number of noble metals
and replacing them with non-noble metals, the costs of catalyst systems can be further
decreased. In addition, thermal dewetting is a simple approach that can be used to produce
bimetallic or trimetallic alloy nanoparticles from deposited layers.
This thesis presents fundamental information that facilitates further application of TiO2
nanosheet electrodes. For instance, TiO2 nanosheet arrays with optimized morphology and
faceting can be used as an ideal host platform for single-atom decoration for photocatalytic H2
production. This structure with well-defined crystal orientation provides a superior platform for
the deposition of various catalysts.
45
Reference
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
J. Liu, M. Liu, X. Yang, H. Chen, S. F. Liu, J. Yan, ACS Sustain. Chem. Eng. 2020, 8,
6055.
K. C. Christoforidis, P. Fornasiero, ChemCatChem 2017, 9, 1523.
J. H. Heo, S. H. Im, J. H. Noh, T. N. Mandal, C. Lim, J. A. Chang, Y. H. Lee, H. Kim,
A. Sarkar, K. Nazeeruddin, M. Gra, 2013, 7, DOI 10.1038/NPHOTON.2013.80.
H. C. Liang, X. Z. Li, J. Nowotny, Solid State Phenom. 2010, 162, 295.
O. Ola, M. M. Maroto-Valer, J. Photochem. Photobiol. C Photochem. Rev. 2015, 24,
16.
X. Chen, S. S. Mao, Chem. Rev. 2007, 107, 2891.
K. Lv, B. Cheng, J. Yu, G. Liu, Phys. Chem. Chem. Phys. 2012, 14, 5349.
Y. Nam, J. H. Lim, K. C. Ko, J. Y. Lee, J. Mater. Chem. A 2019, 7, 13833.
W. A. Abbas, I. H. Abdullah, B. A. Ali, N. Ahmed, A. M. Mohamed, M. Y. Rezk, N.
Ismail, M. A. Mohamed, N. K. Allam, Nanoscale Adv. 2019, 1, 2801.
C. A. Grimes, n.d.
W. Q. Fang, J. Z. Zhou, J. Liu, Z. G. Chen, C. Yang, C. H. Sun, G. R. Qian, J. Zou, S.
Z. Qiao, H. G. Yang, Chem. - A Eur. J. 2011, 17, 1423.
Y. Hou, W. Chen, D. Baran, T. Stubhan, N. A. Luechinger, B. Hartmeier, M. Richter,
J. Min, S. Chen, C. O. R. Quiroz, N. Li, H. Zhang, T. Heumueller, G. J. Matt, A. Osvet,
K. Forberich, Z. Zhang, Y. Li, B. Winter, P. Schweizer, E. Spiecker, C. J. Brabec, Adv.
Mater. 2016, 28, 5112.
F. Shahvaranfard, M. Altomare, Y. Hou, S. Hejazi, W. Meng, B. Osuagwu, N. Li, C. J.
Brabec, P. Schmuki, Adv. Funct. Mater. 2020, 30, 1909738.
E. Bonmatí, A. Casanovas, I. Angurell, J. Llorca, Top. Catal. 2015, 58, 77.
D. Spanu, A. Minguzzi, S. Recchia, F. Shahvardanfard, O. Tomanec, R. Zbořil, P.
Schmuki, P. Ghigna, M. Altomare, ACS Catal. 2020, acscatal.0c01373.
L. Sinatra, A. P. Lagrow, W. Peng, A. R. Kirmani, A. Amassian, H. Idriss, O. M. Bakr,
J. Catal. 2015, 322, 109.
M. KOTOBUKI, A. WATANABE, H. UCHIDA, H. YAMASHITA, M.
WATANABE, J. Catal. 2005, 236, 262.
X. Ye, X. He, Y. Lei, J. Tang, Y. Yu, H. Shi, K. Wang, Chem. Commun. 2019, 55,
2321.
A. L. Linsebigler, G. Lu, J. T. Yates, Chem. Rev. 1995, 735.
S. Chuangchote, J. Jitputti, T. Sagawa, S. Yoshikawa, ACS Appl. Mater. Interfaces
2009, 1, 1140.
M. Altomare, N. T. Nguyen, S. Hejazi, P. Schmuki, Adv. Funct. Mater. 2018, 28,
1704259.
Dongfang Yang, Titanium Dioxide - Material for a Sustainable Environment, InTech,
2018.
D. Yang, IntechOpen 2018.
M. Janus, IntechOpen 2017.
S. J. Smith, R. Stevens, S. Liu, G. Li, A. Navrotsky, J. Boerio-Goates, B. F. Woodfield,
Am. Mineral. 2009, 94, 236.
46
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
P. Roy, S. Berger, P. Schmuki, Angew. Chemie Int. Ed. 2011, 50, 2904.
J. Pascual, J. Camassel, H. Mathieu, Phys. Rev. B 1978, 18, 5606.
H. Tang, K. Prasad, R. Sanjinès, P. E. Schmid, F. Lévy, J. Appl. Phys. 1994, 75, 2042.
J. E. S. Haggerty, L. T. Schelhas, D. A. Kitchaev, J. S. Mangum, L. M. Garten, W. Sun,
K. H. Stone, J. D. Perkins, M. F. Toney, G. Ceder, D. S. Ginley, B. P. Gorman, J. Tate,
Sci. Rep. 2017, 7, 15232.
D. S. Seo, H. Kim, 2003, 2, 571.
Seyedsina Hejazi, Friedrich-Alexander-Universität Erlangen-nürnb. 2020.
K. S. K. Jitendra N.Tiwari, Rajanish N.Tiwari, Prog. Mater. Sci. 2012, 57, 724.
D. . Tryk, A. Fujishima, K. Honda, Electrochim. Acta 2000, 45, 2363.
A. Hagfeldt, M. Graetzel, Chem. Rev. 1995, 95, 49.
M. A. El-Sayed, Acc. Chem. Res. 2004, 37, 326.
Z. Weng, H. Guo, X. Liu, S. Wu, K. W. K. Yeung, P. K. Chu, RSC Adv. 2013, 3,
24758.
K. Lee, A. Mazare, P. Schmuki, Chem. Rev. 2014, 114, 9385.
N. T. Nguyen, J. Yoo, M. Altomare, P. Schmuki, Chem. Commun. 2014, 50, 9653.
S. Hejazi, S. Mohajernia, Y. Wu, P. Andryskova, G. Zoppellaro, I. Hwang, O.
Tomanec, R. Zboril, P. Schmuki, Electrochem. commun. 2019, 98, 82.
P. Roy, D. Kim, I. Paramasivam, P. Schmuki, Electrochem. commun. 2009, 11, 1001.
S. Wu, C. Chen, J. Wang, J. Xiao, T. Peng, ACS Appl. Energy Mater. 2018, 1, 1649.
H. Tang, K. Prasad, R. Sanjinés, F. Lévy, Sensors Actuators B Chem. 1995, 26, 71.
Y. Zhao, Y. Wang, G. Xiao, H. Su, Chinese J. Chem. Eng. 2019, 27, 1416.
P. Nyamukamba, O. Okoh, H. Mungondori, R. Taziwa, S. Zinya, in Titan. Dioxide Mater. a Sustain. Environ., InTech, 2018.
A. Khlyustova, N. Sirotkin, T. Kusova, A. Kraev, V. Titov, A. Agafonov, Mater. Adv.
2020, 1, 1193.
L. A. Al-Hajji, A. A. Ismail, A. Bumajdad, M. Alsaidi, S. A. Ahmed, F. Almutawa, A.
Al-Hazza, Ceram. Int. 2020, 46, 20155.
A. L. Stepanov, Rev. Adv. Mater. Sci. 2012, 30, 150.
H. Yamashita, M. Harada, J. Misaka, M. Takeuchi, Y. Ichihashi, F. Goto, M. Ishida, T.
Sasaki, M. Anpo, J. Synchrotron Radiat. 2001, 8, 569.
S. Lijima, Nature 1991, 354, 56.
C. M. Lieber, Solid State Commun. 1998, 107, 607.
P. Roy, S. Berger, P. Schmuki, Angew. Chemie Int. Ed. 2011, 50, 2904.
B. Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F. Kim, H. Yan,
2003, 353.
P. Hoyer, Langmuir 1996, 12, 1411.
T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Langmuir 1998, 14, 3160.
D. V. Bavykin, J. M. Friedrich, F. C. Walsh, Adv. Mater. 2006, 18, 2807.
H. Masuda, K. Fukuda, Science (80-. ). 1995, 268, 1466.
J. Yoo, M. Altomare, M. Mokhtar, A. Alshehri, S. A. Al-Thabaiti, A. Mazare, P.
Schmuki, J. Phys. Chem. C 2016, 120, 15884.
S. Hejazi, M. Altomare, P. Schmuki, Zeitschrift für Phys. Chemie 2020, 234, 615.
Nhat Truong Nguyen, Friedrich-Alexander-Universität Erlangen-nürnb. 2017.
47
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
F. Shahvaranfard, P. Ghigna, A. Minguzzi, E. Wierzbicka, P. Schmuki, M. Altomare,
ACS Appl. Mater. Interfaces 2020, 12, 38211.
Y. Wang, L. Tan, L. Wang, J. Nanomater. 2011, 2011, DOI 10.1155/2011/529874.
M. Malekshahi Byranvand, A. Nemati Kharat, L. Fatholahi, Z. Malekshahi Beiranvand,
J. Nanostructures 2013, 3, 1.
M. Ge, C. Cao, J. Huang, S. Li, Z. Chen, K. Q. Zhang, S. S. Al-Deyab, Y. Lai, J.
Mater. Chem. A 2016, 4, 6772.
A. Prathan, J. Sanglao, T. Wang, C. Bhoomanee, P. Ruankham, A. Gardchareon, D.
Wongratanaphisan, Sci. Rep. 2020, 10, 8065.
C. Liu, R. Zhu, A. Ng, Z. Ren, S. H. Cheung, L. Du, S. K. So, J. A. Zapien, A. B.
Djurišić, D. Lee Phillips, C. Surya, J. Mater. Chem. A 2017, 5, 15970.
B. Cao, H. Liu, L. Yang, X. Li, H. Liu, P. Dong, X. Mai, C. Hou, N. Wang, J. Zhang, J.
Fan, Q. Gao, Z. Guo, ACS Appl. Mater. Interfaces 2019, 11, 33770.
L. Yang, X. Wang, X. Mai, T. Wang, C. Wang, X. Li, V. Murugadoss, Q. Shao, S.
Angaiah, Z. Guo, J. Colloid Interface Sci. 2019, 534, 459.
A. Fakharuddin, F. Di Giacomo, I. Ahmed, Q. Wali, T. M. Brown, R. Jose, J. Power
Sources 2015, 283, 61.
R. Li, H. Zhang, R. Chai, M. Zhang, M. Guo, Ceram. Int. 2019, 45, 12353.
H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith, H. M. Cheng, G. Q. Lu,
Nature 2008, 453, 638.
Y. Dai, C. M. Cobley, J. Zeng, Y. Sun, Y. Xia, Nano Lett. 2009, 9, 2455.
S. Feng, J. Yang, H. Zhu, M. Liu, J. Zhang, J. Wu, J. Wan, J. Am. Ceram. Soc. 2011,
94, 310.
X. Hu, S. Lu, J. Tian, N. Wei, X. Song, X. Wang, H. Cui, Appl. Catal. B Environ.
2019, 241, 329.
A. Meng, J. Zhang, D. Xu, B. Cheng, J. Yu, Appl. Catal. B Environ. 2016, 198, 286.
R. Behrisch, Springer 1981.
M. Catauro, E. Tranquillo, G. Dal Poggetto, M. Pasquali, A. Dell’Era, S. V. Ciprioti,
Materials (Basel). 2018, 11, DOI 10.3390/ma11122364.
X. Nie, S. Zhuo, G. Maeng, K. Sohlberg, Int. J. Photoenergy 2009, 2009, 1.
R. Mechiakh, N. Ben Sedrine, R. Chtourou, R. Bensaha, Appl. Surf. Sci. 2010, 257,
670.
C. V. Thompson, Annu. Rev. Mater. Res. 2012, 42, 399.
D. J. Srolovitz, M. G. Goldiner, JOM 1995, 47, 31.
W. M. Robertson, J. Appl. Phys. 1971, 42, 463.
G. W. Crabtree, N. S. Lewis, Phys. Today 2007, 60, 37.
A. A. Bayod-Rújula, in Sol. Hydrog. Prod. Process. Syst. Technol., Elsevier, 2019, pp.
237–295.
G. Xiao, G. Zheng, M. Qiu, Q. Li, D. Li, M. Ni, Appl. Energy 2017, 208, 1318.
X. Wang, D. Jing, M. Ni, Sci. Bull. 2017, 62, 597.
Y. Hou, Friedrich-Alexander-University Erlangen-nürnb. 2017.
A. Gheno, S. Vedraine, B. Ratier, J. Bouclé, Metals (Basel). 2016, 6, 21.
C. Chen, Y. Cheng, Q. Dai, H. Song, Nat. Publ. Gr. 2015, 1.
K. Wojciechowski, S. D. Stranks, A. Abate, G. Sadoughi, A. Sadhanala, N. Kopidakis,
48
G. Rumbles, C. Li, R. H. Friend, A. K.-Y. Jen, H. J. Snaith, ACS Nano 2014, 8, 12701.
[90] M. T, K. A, T. K, S. Y, J. Am. Chem. Soc. 2009, 131, 6050.
[91] EMILIANO BELLINI, PV Mag. 2021.
[92] D. Zhou, T. Zhou, Y. Tian, X. Zhu, Y. Tu, J. Nanomater. 2018, 2018, 1.
[93] T.-B. Song, Q. Chen, H. Zhou, C. Jiang, H.-H. Wang, Y. (Michael) Yang, Y. Liu, J.
You, Y. Yang, J. Mater. Chem. A 2015, 3, 9032.
[94] W. Wu, L. Wang, Adv. Funct. Mater. 2018, 28, 1804356.
[95] S. S. Mali, C. K. Hong, Nanoscale 2016, 8, 10528.
[96] I. Hussain, H. P. Tran, J. Jaksik, J. Moore, N. Islam, M. J. Uddin, Emergent Mater.
2018, 1, 133.
[97] D. Luo, R. Su, W. Zhang, Q. Gong, R. Zhu, Nat. Rev. Mater. 2020, 5, 44.
[98] H.-H. Wang, Q. Chen, H. Zhou, L. Song, Z. S. Louis, N. De Marco, Y. Fang, P. Sun,
T.-B. Song, H. Chen, Y. Yang, J. Mater. Chem. A 2015, 3, 9108.
[99] R. Salazar, M. Altomare, K. Lee, J. Tripathy, R. Kirchgeorg, N. T. Nguyen, M.
Mokhtar, A. Alshehri, S. A. Al-Thabaiti, P. Schmuki, ChemElectroChem 2015, 2, 824.
[100] C. Chen, Y. Cheng, Q. Dai, H. Song, Sci. Rep. 2016, 5, 17684.
[101] A. A. Navab, A. Nemati, A. A. Navab, H. M. M. Abad, in AIP Conf. Proc., 2018, p.
020015.
[102] Y. Wu, X. Yang, H. Chen, K. Zhang, C. Qin, J. Liu, W. Peng, A. Islam, E. Bi, F. Ye,
M. Yin, P. Zhang, L. Han, 2014.
[103] K. Mahmood, S. Sarwar, M. T. Mehran, RSC Adv. 2017, 7, 17044.
[104] L. Li, C. Shi, X. Deng, Y. Wang, L. Ni, Micro Nano Lett. 2018, 13, 131.
[105] Y. Li, Q. Zhang, L. Niu, J. Liu, X. Zhou, Thin Solid Films 2017, 621, 6.
[106] U. K. Thakur, A. M. Askar, R. Kisslinger, B. D. Wiltshire, P. Kar, K. Shankar,
Nanotechnology 2017, 28, 274001.
[107] P. Zhao, B. J. Kim, H. S. Jung, Mater. Today Energy 2018, 7, 267.
[108] S. Shao, M. A. Loi, Adv. Mater. Interfaces 2020, 7, 1901469.
[109] Y. Hou, W. Chen, D. Baran, T. Stubhan, N. A. Luechinger, B. Hartmeier, M. Richter,
J. Min, S. Chen, C. Omar, R. Quiroz, N. Li, H. Zhang, T. Heumueller, G. J. Matt, A.
Osvet, K. Forberich, Z. Zhang, Y. Li, B. Winter, P. Schweizer, E. Spiecker, C. J.
Brabec, 2016, DOI 10.1002/adma.201504168.
[110] Q. Xue, Y. Bai, M. Liu, R. Xia, Z. Hu, Z. Chen, X.-F. Jiang, F. Huang, S. Yang, Y.
Matsuo, H.-L. Yip, Y. Cao, Adv. Energy Mater. 2017, 7, 1602333.
[111] Z. Zhou, S. Pang, Z. Liu, H. Xu, G. Cui, J. Mater. Chem. A 2015, 3, 19205.
[112] S. Y. Kim, S. J. Cho, S. E. Byeon, X. He, H. J. Yoon, Adv. Energy Mater. 2020, 10,
2002606.
[113] J. Ma, J. Chang, Z. Lin, X. Guo, L. Zhou, Z. Liu, H. Xi, D. Chen, C. Zhang, Y. Hao, J.
Phys. Chem. C 2018, 122, 1044.
[114] C. Tao, S. Neutzner, L. Colella, S. Marras, A. R. Srimath Kandada, M. Gandini, M. De
Bastiani, G. Pace, L. Manna, M. Caironi, C. Bertarelli, A. Petrozza, Energy Environ.
Sci. 2015, 8, 2365.
[115] X. Guo, B. Zhang, Z. Lin, J. Ma, J. Su, W. Zhu, C. Zhang, J. Zhang, J. Chang, Y. Hao,
Org. Electron. 2018, 62, 459.
[116] Z. Xu, X. Yin, Y. Guo, Y. Pu, M. He, J. Mater. Chem. C 2018, 6, 4746.
49
[117] Z. Xu, S. H. Teo, L. Gao, Z. Guo, Y. Kamata, S. Hayase, T. Ma, Org. Electron. 2019,
73, 62.
[118] P. Sakthivel, S. Foo, M. Thambidurai, P. C. Harikesh, N. Mathews, R. Yuvakkumar, G.
Ravi, C. Dang, J. Power Sources 2020, 471, 228443.
[119] W. Okada, T. Suga, K. Oyaizu, H. Segawa, H. Nishide, ACS Appl. Energy Mater.
2019, 2, 2848.
[120] W. Sun, K.-L. Choy, M. Wang, Molecules 2019, 24, 3466.
[121] V. N. Vineeth, G. E. Unni, P. Srikrishnarka, S. Nandan, A. S. Nair, Mater. Today Proc.
2020, 33, 1351.
[122] N. F. M. Sharif, S. Shafie, M. Z. A. Ab. Kadir, W. Z. W. Hasan, M. N. Mustafa, B.
Samaila, Results Phys. 2019, 15, 102725.
[123] H. K. Adli, T. Harada, S. Nakanishi, S. Ikeda, Phys. Chem. Chem. Phys. 2017, 19,
26898.
[124] S. Yang, Y. Hou, B. Zhang, X. H. Yang, W. Q. Fang, H. J. Zhao, H. G. Yang, J. Mater.
Chem. A 2013, 1, 1374.
[125] S. S. Mali, C. S. Shim, H. K. Park, J. Heo, P. S. Patil, C. K. Hong, Chem. Mater. 2015,
27, 1541.
[126] W. Liu, L. Chu, N. Liu, Y. Cheng, F. Wu, Y. Li, Y. Pu, J. Zhang, X. Li, W. Huang,
Adv. Mater. Interfaces 2019, 6, 1801976.
[127] F. Gao, H. Dai, H. Pan, Y. Chen, J. Wang, Z. Chen, J. Colloid Interface Sci. 2018, 513,
693.
[128] R. Zhang, W. Liu, R. Hu, Y. Ma, Y. Sun, J. Zhang, Y. Pu, J. Yang, L. Chu, X. Li, Sol.
Energy Mater. Sol. Cells 2019, 191, 183.
[129] L. Zhu, Q. Lu, L. Lv, Y. Wang, Y. Hu, Z. Deng, Z. Lou, Y. Hou, F. Teng, RSC Adv.
2017, 7, 20084.
[130] A. Yella, L.-P. Heiniger, P. Gao, M. K. Nazeeruddin, M. Grätzel, Nano Lett. 2014, 14,
2591.
[131] Y. Wang, J. Wan, J. Ding, J. S. Hu, D. Wang, Angew. Chemie - Int. Ed. 2019, 58, 9414.
[132] S. Styring, Faraday Discuss. 2012, 155, 357.
[133] O. Siddiqui, I. Dincer, Energy Convers. Manag. 2020, 208, 112590.
[134] N. Serpone, A. V. Emeline, S. Horikoshi, V. N. Kuznetsov, V. K. Ryabchuk,
Photochem. Photobiol. Sci. 2012, 11, 1121.
[135] A. FUJISHIMA, K. HONDA, Nature 1972, 238, 37.
[136] D. Spanu, S. Recchia, S. Mohajernia, O. Tomanec, Š. Kment, R. Zboril, P. Schmuki,
M. Altomare, ACS Catal. 2018, 8, 5298.
[137] N. T. Nguyen, M. Altomare, J. E. Yoo, N. Taccardi, P. Schmuki, Adv. Energy Mater.
2016, 6, 1501926.
[138] J. E. Yoo, K. Lee, M. Altomare, E. Selli, P. Schmuki, Angew. Chemie 2013, 125, 7662.
[139] J. Zhang, B. Tian, L. Wang, M. Xing, J. Lei, 2018, pp. 1–15.
[140] Y. Zhang, Y.-J. Heo, J.-W. Lee, J.-H. Lee, J. BajGai, K.-J. Lee, S.-J. Park, Catalysts
2018, 8, 655.
[141] A. Galińska, J. Walendziewski, Energy & Fuels 2005, 19, 1143.
[142] A. Hauch, S. D. Ebbesen, S. H. Jensen, M. Mogensen, J. Mater. Chem. 2008, 18, 2331.
[143] B. Kte’pi, in Green Technol. An A-to-Z Guid., SAGE Publications, Inc., 2455 Teller
50
Road, Thousand Oaks California 91320 United States, 2012.
[144] S. Chen, L. W. Wang, Chem. Mater. 2012, 24, 3659.
[145] S. M. El-Refaei, P. A. Russo, N. Pinna, ACS Appl. Mater. Interfaces 2021, 13, 22077.
[146] W. Choi, G. Hu, K. Kwak, M. Kim, D. E. Jiang, J. P. Choi, D. Lee, ACS Appl. Mater.
Interfaces 2018, 10, 44645.
[147] M. Wang, S. Shen, L. Li, Z. Tang, J. Yang, J. Mater. Sci. 2017, 52, 5155.
[148] A. C. Williams, N. Camp, in Six-Membered Hetarenes with One Chalcogen, Georg
Thieme Verlag KG, 2003, p. 1.
[149] L. Wu, Q. Li, C. Yang, X. Ma, Z. Zhang, X. Cui, J. Mater. Chem. A 2018, 6, 20947.
[150] G. Zhao, K. Rui, S. X. Dou, W. Sun, Adv. Funct. Mater. 2018, 28, 1803291.
[151] Q. Su, Y. Li, R. Hu, F. Song, S. Liu, C. Guo, S. Zhu, W. Liu, J. Pan, Adv. Sustain. Syst.
2020, 4, 1.
[152] J. Low, J. Yu, M. Jaroniec, S. Wageh, A. A. Al-Ghamdi, Adv. Mater. 2017, 29, 1.
[153] H. Wang, L. Zhang, Z. Chen, J. Hu, S. Li, Z. Wang, J. Liu, X. Wang, Chem. Soc. Rev.
2014, 43, 5234.
[154] G. L. Chiarello, M. H. Aguirre, E. Selli, J. Catal. 2010, 273, 182.
[155] Y. Liu, D. Su, Y. Zhang, L. Wang, G. Yang, F. Shen, S. Deng, X. Zhang, S. Zhang, J.
Mater. Res. 2017, 32, 757.
[156] L. Ji, D. Spanu, N. Denisov, S. Recchia, P. Schmuki, M. Altomare, Chem. – An Asian
J. 2019, asia. 201901545.
[157] T. Chen, Z. Feng, G. Wu, J. Shi, G. Ma, P. Ying, C. Li, J. Phys. Chem. C 2007, 111,
8005.
[158] Z. Shu, Y. Cai, J. Ji, C. Tang, S. Yu, W. Zou, L. Dong, Catalysts 2020, 10, 1047.
[159] T. Tatsuma, H. Nishi, T. Ishida, Chem. Sci. 2017, 8, 3325.
[160] S. Xu, J. Ng, X. Zhang, H. Bai, D. D. Sun, Int. J. Hydrogen Energy 2010, 35, 5254.
[161] S.-S. Yi, X.-B. Zhang, B.-R. Wulan, J.-M. Yan, Q. Jiang, Energy Environ. Sci. 2018,
11, 3128.
[162] D. Spanu, A. Minguzzi, S. Recchia, F. Shahvardanfard, O. Tomanec, R. Zbořil, P.
Schmuki, P. Ghigna, M. Altomare, Under Consid. n.d.
[163] D. Spanu, S. Recchia, S. Mohajernia, O. Tomanec, Š. Kment, n.d.
[164] R. Chandran, S. Dharmalingam, J. Nanosci. Nanotechnol. 2014, 14, 6957.
[165] M. Bernareggi, M. Dozzi, L. Bettini, A. Ferretti, G. Chiarello, E. Selli, Catalysts 2017,
7, 301.
[166] M. Bernareggi, G. L. Chiarello, G. West, M. Ratova, A. M. Ferretti, P. Kelly, E. Selli,
Catal. Today 2019, 326, 15.
[167] M. V. Dozzi, G. L. Chiarello, M. Pedroni, S. Livraghi, E. Giamello, E. Selli, Appl.
Catal. B Environ. 2017, 209, 417.
[168] Q. Li, X. Li, J. Yu, Surface and Interface Modification Strategies of CdS-Based
Photocatalysts, 2020.
[169] V. Kumaravel, M. D. Imam, A. Badreldin, R. K. Chava, J. Y. Do, M. Kang, A. AbdelWahab, Catalysts 2019, 9, DOI 10.3390/catal9030276.
[170] E. Kalamaras, P. Lianos, J. Electroanal. Chem. 2015, 751, 37.
51
Abbreviations and symbols
0D
zero-dimensional
1D
one-dimensional
2D
two-dimensional
3D
three-dimensional
AFM
Atomic force microscopy
Ag
Silver
Au
Gold
CB
Conduction band
Cu
Copper
Eg
Bandgap
EQE
External quantum efficiency
ETL
Electron transport layer
FF
Fill factor
FL
Fermi level
FTO
Fluorine doped tin oxide
HER
Hydrogen evolution reaction
HR-TEM
High-resolution transmission electron microscopy
HTL
Hole transport layer
IPCE
Incident photon-to-current efficiency
MnOx
Molybdenum trioxide
mpp
Maximum power point
NP
Nanoparticle
NR
Nanorod
NT
Nanotube
PCE
Power conversion efficiency
PCBM
[6,6]-Phenyl-C61-Butyric-acid-Methyl ester
52
PDCBT
poly[5,5′-bis(2-butyloctyl)-(2,2′-bithiophene)-4,4′-dicarboxylate-alt-5,5′2,2′-bithiophene]
PL
Photoluminescence
PSC
Perovskite solar cell
Pt
Platinum
RMS
Root mean square
SAM
Self-assembled monolayer
SEM
Scanning electron microscopy
TEM
Transmission electron microscopy
Ti
Titanium
VB
Valence band
XAS
X-ray absorption spectroscopy
XPS
X-ray photoelectron spectroscopy
XRD
X-ray diffraction
E
Electric field
h+
Electron hole
e-
Electron
JSC
Short circuit current
T
Temperature
t
Time
V
Voltage
VOC
Open circuit voltage
Z
Impedance
θ
X-ray scattering angle
λ
Wavelength
Φ
Work function
53
Appendix A
Engineering of the electron transport layer/perovskite interface in solar
cells designed on TiO2 rutile nanorods
Fahimeh Shahvaranfard, Marco Altomare, Yi Hou, Seyedsina Hejazi, Wei Meng, Benedict
Osuagwu, Ning Li, Christoph J. Brabec, Patrik Schmuki
Adv. Funct. Mater. 2020, 30, 1909738
54
Full Paper
www.afm-journal.de
Engineering of the Electron Transport Layer/Perovskite
Interface in Solar Cells Designed on TiO2 Rutile Nanorods
Fahimeh Shahvaranfard, Marco Altomare,* Yi Hou, Seyedsina Hejazi, Wei Meng,
Benedict Osuagwu, Ning Li, Christoph J. Brabec, and Patrik Schmuki*
photovoltaic technologies. Cell efficiencies have reached beyond 25% in a short
period of time, making perovskite solar
cells (PSCs) one of the fast-advancing solar
technology to date.[6,7]
PSCs are mainly classified into three
types of architectures depending on the
morphology of the beneath interface layer,
e.g. planar, mesoporous, or nanostructured.[8–10] In all configurations, both the
electron transport layer (ETL), typically
a TiO2 or SnO2 scaffold, and its interface with the perovskite light absorber
play a crucial role for the overall cell
performance.[11]
1D TiO2 nanomaterials such as
nanorods (NRs), nanowires, and nanotubes have meanwhile been widely applied
to the fabrication of nanostructured ETLs
as they can provide advantageous features,
such as improved interfacial contact with
the perovskite and enhanced charge diffusion, due to their large available surface area and directional charge transport
paths. Both these features can be exploited
to limit the charge recombination in the perovskite and at the
ETL/perovskite interface.[8,9,12–15]
Layers of TiO2 nanotubes, produce, for example, by anodic
oxidation of Ti metal or by selective core-etching of hydrothermally-grown NRs, have been explored as ETL for perovskite
cells though with limited success, i.e., typical efficiency values
are <15%.[15–18] A main reason for the limited performance can
The engineering of the electron transport layer (ETL)/light absorber interface
is explored in perovskite solar cells. Single-crystalline TiO2 nanorod (NR)
arrays are used as ETL and methylammonium lead iodide (MAPI) as
light absorber. A dual ETL surface modification is investigated, namely
by a TiCl4 treatment combined with a subsequent PC61BM monolayer
deposition, and the effects on the device photovoltaic performance were
evaluated with respect to single modifications. Under optimized conditions,
for the combined treatment synergistic effects are observed that lead
to remarkable enhancements in cell efficiency, from 14.2% to 19.5%, and to
suppression of hysteresis. The devices show JSC, VOC, and fill factor as high
as 23.2 mA cm−2, 1.1 V, and 77%, respectively. These results are ascribed
to a more efficient charge transfer across the ETL/perovskite interface,
which originates from the passivation of defects and trap states at the ETL
surface. To the best of our knowledge, this is the highest cell performance
ever reported for TiO2 NR-based solar cells fabricated with conventional
MAPI light absorber. Perspective wise, this ETL surface functionalization
approach combined with more recently developed and better performing
light absorbers, such as mixed cation/anion hybrid perovskite materials, is
expected to provide further performance enhancements.
1. Introduction
In the past few years, organic–inorganic hybrid halide perovskite
materials have come into focus due to their unique optoelectronic properties such as strong visible light absorbance and high
carrier mobility.[1–5] The implementation of this class of materials
into solar cells has led to a new generation of rapidly developing
F. Shahvaranfard, Dr. M. Altomare, S. Hejazi, B. Osuagwu,
Prof. P. Schmuki
Institute for Surface Science and Corrosion WW4-LKO
Department of Materials Science and Engineering
University of Erlangen-Nuremberg
Martensstrasse 7, 91058 Erlangen, Germany
E-mail: marco.altomare@fau.de; schmuki@ww.uni-erlangen.de
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adfm.201909738.
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim. This is an open access article under the terms of the
Creative Commons Attribution License, which permits use, distribution
and reproduction in any medium, provided the original work is properly
cited.
DOI: 10.1002/adfm.201909738
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (1 of 9)
Dr. Y. Hou, W. Meng, Dr. N. Li, Prof. C. J. Brabec
Institute of Materials for Electronics and Energy Technology (i-MEET)
Department of Materials Science and Engineering
University of Erlangen-Nuremberg
Martensstrasse 7, 91058 Erlangen, Germany
Dr. N. Li, Prof. C. J. Brabec
Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (HI ERN)
Immerwahrstr. 2, 91058 Erlangen, Germany
Dr. N. Li
National Engineering Research Center for Advanced
Polymer Processing Technology
Zhengzhou University
450002 Zhengzhou, China
Prof. P. Schmuki
Chemistry Department
Faculty of Sciences
King Abdulaziz University
80203 Jeddah, Kingdom of Saudi Arabia
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
be the incomplete filling of pores with the perovskite. This can
affect the light absorbance, the charge transport and interfacial
transfer, and thus the overall cell efficiency.
In this context, rutile TiO2 NRs provide better properties, i.e.,
an open structure for an effective infiltration of the perovskite
and a single crystalline nature. The latter can, for example in
contrast to anodically grown polycrystalline TiO2 nanotubes,
provide enhanced electron transfer properties[10,12] and thus, in
principle, better cell performances.
Nevertheless, aside from the ETL morphology, key to the
device efficiency is also to optimize the ETL/perovskite interface properties, e.g. by passivation of the TiO2 surface defects
and trap states.[4,19,20] To this end, titanium tetrachloride (TiCl4)
treatment, has been reported as a most efficient approach to
improve TiO2 electron transport ability: it forms at the treated
surface a nm-thick layer of TiO2 anatase nanoparticles (NPs,
with size in the 5–10 nm range) that typically favors a more
efficient charge extraction across the ETL/perovskite interface,
thereby limiting photon losses ascribed to charge recombination in the light absorber.[3,21,22]
An alternative approach is to introduce an additive between
the ETL and perovskite layer. Phenyl-C61-butyric acid methyl
ester (PC61BM), a thin organic surface modifier, was recently
implemented in perovskite cells’ technology as an efficient
interlayer to passivate surface defects in both the metal oxide
ETL and perovskite light absorber. This can substantially reduce
the device hysteresis and improve the cell efficiency.[23,24]
Herein, we discuss the synergistic effects in cell performance enhancement achieved by surface modification of TiO2
NR ETLs by TiCl4 treatment combined with PC61BM monolayer
deposition. Such an engineering of the TiO2 ETL/perovskite
interface leads to a remarkable increase of the cell efficiency up
to 19.5% for our champion device (17.9% average PCE). Our
results suggest that on the one hand, the TiCl4 surface treatment allows for improving the surface contact of the ETL with
the perovskite and thus increases the cell performance due to a
more efficient charge transfer. On the other hand, the PC61BM
monolayer, by passivating surface defects at the TiO2 ETL/
perovskite interface, enhances the efficiency even further and
leads to a nearly complete suppression of the device hysteresis.
2. Results and Discussion
To optimize the PSCs, we engineered the ETL/perovskite interface using hydrothermally grown TiO2 NRs modified by TiCl4
and/or PC61BM treatments. The TiO2 NR arrays were decorated
with different loadings of TiO2 anatase NPs; for this the NRs
were treated in TiCl4 solutions of different concentrations, i.e.,
50, 100, and 200 × 10−3 m. After TiCl4 treatment and annealing,
the ETL were used to fabricate the PSCs as outlined in the
Experimental Section. Temperature and time of the TiCl4 treatment were kept constant for all the samples.
Figure 1 and Figure S1 in the Supporting Information show
cross-sectional and top view scanning electron microscope
(SEM) images, respectively, of vertically aligned TiO2 NRs
grown on fluorine-doped tin oxide (FTO) slides, before and
after decoration (surface modification) with different NP loadings. Figure 1a shows the bare NRs (as grown). The length of
the NRs is in average 550–600 nm. It is clear from the SEM
images b–d) that when increasing the concentration of the TiCl4
solution from 50 to 200 × 10−3 m, the TiO2 NP loading consistently increases as well as the average TiO2 NP size. The TiO2
NP forms at the NR surface a conformal layer, the thickness
Figure 1. Cross-sectional SEM images of TiO2 NRs decorated by TiCl4 treatment, using different TiCl4 solution concentrations: a) pristine NRs;
b) 50 × 10−3 m; c) 100 × 10−3 m, and d) 200 × 10−3 m.
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (2 of 9)
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
Figure 3. Cross-sectional SEM image of a typical PSC investigated in the
present work.
Figure 2. XRD patterns of a) pristine NRs, b) TiCl4-treated NRs, c) NRs/
MAPbI3, and d) TiCl4-treated NRs/PC61BM/MAPbI3 on FTO.
of which scales with the TiCl4 solution concentration (this is
well in line with previous work[25]). The NP seem to form also
at the free FTO surface (i.e., FTO surface not coated by NRs).
The X-ray diffraction (XRD) patterns of pure and TiCl4treated NRs are compiled in Figure 2. In general, NRs’ features such as diameter, length, and density could be varied by
changing growth parameters, such as growth time and temperature, initial reactant concentration, acidity, or additives. Nevertheless, the growth mechanism remains based on the epitaxial
relation between the FTO substrate and the growing rutile
TiO2 phase. Precisely, the small lattice mismatch between FTO
and rutile TiO2 phases plays a key role in driving the nucleation and epitaxial growth towards single-crystalline rutile TiO2
NRs.[26,27]
The XRD data in Figure 2 show that, upon NR growth, peaks
appear that agree well with the tetragonal rutile phase. However,
compared to common powder diffraction patterns (randomly
oriented rutile TiO2 crystals), the intensity of the (101) diffraction
peak (at 36.0°) is relatively high, while, e.g., the peak ascribed to
the (110) reflection (at 27.36°, typically a main reflection for a
polycrystalline rutile specimen) is absent. This indicates that the
NRs are highly oriented with respect to the substrate, supporting
the epitaxial growth on FTO. Previous work,[26,27] where similar
experimental conditions were used to grow TiO2 NRs, proved
by high-resolution transmission electron microscopy (HR-TEM)
analysis the single-crystalline nature of such rutile TiO2 NRs.
No information on the crystallographic features of the decorated NPs could be obtained by XRD, probably owing to the
low loading (below XRD detection limits). However, based on
the adopted experimental conditions and previous literature,
the deposited NPs are supposed to be composed of anatase
phase.[25,28,29]
In the case of dual modification of the ETL, the PSCs were
constructed with the following architecture: FTO/TiO2 NRs/
TiO2 NPs/PC61BM/MAPbI3/PDCBT/MoOx/Ag. An SEM
image of the cross-sectional cut of a typical device is provided
in Figure 3. We also fabricated reference cells either without
PC61BM or without TiCl4 treatment.
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (3 of 9)
The device performance of the PSCs was assessed by
measuring their J–V curves under AM 1.5G irradiation
(100 mW cm−2); data are compiled in Figure 4a. The results
for the average photovoltaic parameters of each cell are summarized in Table 1. According to the results, the pristine TiO2
NR-based device shows the lowest performance with a power
conversion efficiency (PCE) of 12.2%, a short-circuit current
density (JSC) of 21.1 mA cm−2, an open-circuit voltage (VOC) of
0.85 V, and a fill factor (FF) of 69%. The relatively poor performance of this device has to be ascribed to a large density of trap
states at the surface of the unmodified TiO2 NRs.[30]
The TiCl4 and PC61BM treatments are expected to influence
the charge extraction across the TiO2 ETL/perovskite interface.
Figure 4a shows the J–V curves for devices (champion cells)
fabricated by different ETL treatments, while in Figure 4b–e
we have compiled a statistical evaluation of the photovoltaic
parameters for devices based on differently treated ETLs. The
photovoltaic parameters of these cells are also summarized in
Table 1.
We first explored the effect of TiCl4 and PC61BM treatments
separately. For the devices fabricated from ETLs treated with
only TiCl4 (100 × 10−3 m), the cell performance increases in
average from 12.2% (reference cell with pristine NRs) to 13.7%.
This is caused by a slight increase of the JSC and, more importantly, by a significant improvement of the VOC (from 0.85 V
for the reference device to 0.94 V). Here the deposited TiO2 NP
layer plays a key role, i.e., it provides a charge blocking effect
by forming a barrier layer at the FTO/perovskite interface,[31]
thereby aiding in suppressing detrimental charge recombination. It has been also reported that TiCl4 treatment can also
effectively reduce the density of surface defects in metal oxide
ETLs.[11,22,25]
To prove the synergistic effects arising from decorating the
NRs by the TiCl4 treatment, we fabricated control devices without
NRs and by a direct TiCl4 treatment of the FTO substrates. With
this treatment, the FTO slides result covered in a homogeneous
fashion by TiO2 NPs, the size of which is in average comparable to that of NPs decorated on the NRs (see the SEM of a
typical TiCl4-treated FTO slide in Figure S2 in the Supporting
Information). As illustrated in Figure S3 (and Table S1)
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
Figure 4. a) Current density–voltage characteristics of the best devices fabricated by different ETL modifications; b–e) statistics of the photovoltaic
parameters for devices constructed from A) pristine NRs; B) NRs treated with 100 × 10−3 m TiCl4; C) NR coated with PC61BM; D) NR/50 × 10−3 m
TiCl4/PC61BM; E) NR/100 × 10−3 m TiCl4/PC61BM; and F) NR/200 × 10−3 m TiCl4/PC61BM.
in the Supporting Information, the devices produced from TiO2
NP-decorated FTO show a clearly worse performance (PCE of the
champion cell: 10.6%) compared to cells fabricated from pristine
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (4 of 9)
TiO2 NRs (14.2%). This confirms not only the benefit of using
the NR arrays as nanostructured ETL but also the synergistic
effects arising from their surface modification with TiO2 NPs.
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
Table 1. Summary of the photovoltaic parameters of devices fabricated from differently treated ETLs.
ETL construction
JSC [mA cm−2]
VOC [V]
FF [%]
PCEavg [%]
PCEmax [%]
TiO2 NRs
21.1±0.83
0.85±0.04
69±2.5
12.2±1.04
14.2
TiO2 NRs/100 × 10−3 m TiCl4
21.6±0.82
0.94±0.06
69±2.8
13.7±1.4
16.3
TiO2 NRs/PC61BM
21.5±0.83
1±0.04
70±3.9
14.9±1.4
17.2
TiO2 NRs/50 × 10−3 m TiCl4/PC61BM
22.1±0.81
1.07±0.03
72±4.7
16.8±1.5
19
TiO2 NRs/100 × 10−3 m TiCl4/PC61BM
22.8±0.56
1.09±0.02
74±3.4
17.9±0.95
19.5
TiO2 NRs/200 × 10−3 m TiCl4/PC61BM
22.1±0.58
1.09±0.02
72±3.5
17±1.02
18.2
For devices constructed from ETLs treated with PC61BM only,
the PCE reached 14.9% and the VOC increased to 1 V. PC61BM
binds to the hydroxyl groups at the TiO2 surface through the
carboxylic acid anchoring group, while the fullerene moiety
faces the perovskite side. As reported by Wojciechowski et al.,
the VOC and PCE enhancement can be ascribed to an interfacial alteration of the TiO2/perovskite junction, i.e., PC61BM
passivates or inhibits the formation of trap states at the interface (sub-bandgap states originated by under-coordinated
surface Ti(IV) ions), on the TiO2 surface through the carboxylic acid anchoring group, and on the perovskite side by the
fullerene moiety.[32–35] This in turn prevents interfacial electrostatic barriers and provides swift charge (electron) mobility
across the ETL/perovskite interface.[33]
It is worth pointing out that for both single modifications
(either TiCl4 treatment or PC61BM deposition) the cell performance enhancement should be attributed mainly to an evident
increase of the VOC, while the JSC improves only to a minor
extent and the FF remains virtually unchanged.
We then investigated the dual modification of TiO2 NR ETLs,
i.e., by TiCl4 treatment combined to a subsequent PC61BM
deposition. Upon dual modification, we observed a maximized
improvement of the device performance. Such an improvement shows a clear dependence on the TiCl4 concentration and
is particularly pronounced when treating the TiO2 NRs with a
100 × 10−3 m TiCl4 solution. Under optimized conditions, the
cells delivered superior photovoltaic parameters with a Voc as
high as 1.09 V along with improved Jsc of 22.8 mA cm−2 and
FF of 74%, resulting in an average PCE of 17.9%. The champion device achieved a JSC of 23.2 mA cm−2, a VOC of 1.1 V, an
FF of 77%, and a PCE of 19.5%. To the best of our knowledge,
this is the highest cell performance reported for TiO2 rutile NRbased devices fabricated with methylammonium lead triiodide
(MAPbI3) light absorber (see Table S2 in the Supporting Information for a summary of the highest cell performance values
reported in the literature). Perspective wise, we expect further
enhancements in cell efficiency by replacing MAPI with more
recently developed and better performing light absorbers, such
as mixed cation/anion hybrid perovskite materials.[36]
Aside from the beneficial effect of the PC61BM interlayer
and the blocking effect of NPs (outlined above), an additional
reason for the performance enhancement can be associated
with the more negative conduction band (CB) edge of anatase
NPs with respect to the CB minimum of the rutile NRs: as
demonstrated by Yang et al.,[31] such energetic situation favors
efficient transfer of photo-generated electrons from MAPbI3 to
the CB of anatase (NPs) and then onward to the CB of rutile
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (5 of 9)
NRs. In addition, the TiCl4 treatment increases the roughness of the NR, providing a nanostructured interfacial contact
between the ETL and the perovskite which can thereby sustain
a more efficient charge separation and transfer.[37] Nevertheless,
a further increase of NPs loading, i.e., a thicker TiO2 NP layer,
was found to cause a decline in cell efficiency (Figure 4 and
Table 1). The charge transport across the NP layer may occur
via “random walk”, through the NP and across the NP boundaries. The thicker the NP layer, the larger the number of diffusion paths from a statistical point of view, and hence the less
efficient the charge transport. This decreases the charge carrier
collection efficiency,[21,25] as suggested by the dramatic drop of
JSC and FF (this aspect is discussed also below along with the
results of impedance spectroscopy, IS).
A summary of the effect of the TiCl4 and PC61BM treatments
is outlined in Figure 4b–e. The characterization of 20 devices
fabricated for each condition proves the consistency of the
statistical distribution and tendency of the performance parameters. Overall, one can conclude that each parameter is maximized by combining the TiCl4 and PC61BM treatments, and a
most optimized TiCl4 concentration is 100 × 10−3 m.
The VOC, as mentioned above, is significantly affected by
the ETL/perovskite interface modification. The average value
increases from 0.85 V for the reference device to 1.09 V for the
dual modification. This result may be ascribed to the reduction of the electronic disorder at the ETL/perovskite interface.[3] In general, the “electronic disorder” can be interpreted
as a measure of the distribution of the electronic density of
states (DOS). The electronic states at the ETL/perovskite interface can feature a broad distribution, i.e., can be subject to a
large energy disorder originating from structural and chemical
inhomogeneity, low crystallinity, random molecular orientation, interactions with neighboring molecules, and presence of
impurities, among others. Higher disorder, e.g., in the titania
ETL can increase the number of trap sites in the device structure, which eventually leads to electron–hole recombination
and short charge carrier lifetimes. Here, the TiCl4 treatment is
proven to passivate the titania surfaces, lowering the DOS and
reducing the disorder level, which in turn lowers the interfacial
charge transfer resistance, enhances the open circuit voltage
and thus maximizes the device performance.[3,38]
The increase of current density for devices based on
dual modification is supported by a clear enhancement of
the device external quantum efficiency (EQE), as shown in
Figure 5a. The JSC values for the device fabricated from pristine TiO2 NRs and for the cell produced by dual modification
(TiCl4 and PC61BM treatments) are 19.5 and 21.1 mA cm−2,
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
Figure 5. a) EQE spectra of PSCs constructed from differently treated TiO2 NRs; b) hysteresis curves of perovskite devices based on pristine TiO2 NRs
and TiO2 NRs modified by an optimized TiCl4 and PC61BM treatment (F and R indicate the forward and reverse scans, respectively).
respectively (these data are consistent with the values derived
from the J–V curves). Besides, the FF enhancement (from
69% for the reference device to 74% measured by ETL dual
modification) can be ascribed to a decrease of the charge
transport resistance and to a better filling of the perovskite in
the ETL structure.[3]
Figure S4a in the Supporting Information shows the current
density delivered by the best performing device: a stable current
density of ≈21.8 mA cm−2 is tracked under steady-state conditions; this result is in agreement with the J–V curves and EQE
data. We evaluated the stability for a device fabricated with the
following cell architecture: FTO/TiO2 NRs/(100 × 10−3 m) TiO2
NPs/PC61BM/MAPbI3/PDCBT/MoOx/Ag. The photovoltaic
performance was assessed for the as-fabricated device as well
as after storing the device for 7 and 14 d. The results, provided
in Figure S4b in the Supporting Information, show that JSC
drops to some extent over time, suggesting that the perovskite
is undergoing partial degradation as also reported in previous
work.[39] On the contrary, the VOC remains almost unaltered.
In this context, we anticipate the use of more stable mixed
cation/anion perovskite materials[10] to reduce efficiency losses
ascribed to perovskite decomposition.
Figure 5b shows the effect of the combined TiCl4 treatment
and PC61BM modification on cell hysteresis with respect to a
reference device (constructed from pristine NRs). Severe loss
in cell performance were measured by sequential forward and
reverse bias scans for the device fabricated with unmodified
NRs, while the hysteresis is considerably suppressed for the
device modified with both TiCl4 and PC61BM. The influence of
the interface engineering on the open-circuit voltage is evident:
for pristine NR, there is a considerable difference between the
VOC measured in forward and reverse scans, while after the
dual surface modification the difference becomes negligible.
Previous work demonstrated that trap states at the TiO2 ETL
surface (i.e., at the ETL/perovskite interface) are the major
cause for hysteresis.[33,40,41] Our results suggest that the dual
ETL modification leads to an effective passivation of such
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (6 of 9)
trap states, thereby improving the charge mobility across the
ETL/perovskite interface.
UV–vis diffuse reflectance and steady-state photoluminescence (PL) measurements were conducted to further study
the light absorbance feature and charge transfer properties
of the perovskite/ETL architectures based on various surface engineering methods. The data are compiled in Figure 6
(the optical features of differently treated ETLs are reported as
reference data in Figure S5 in the Supporting Information). The
UV–vis spectra (Figure 6a) show a higher absorbance for the
perovskite deposited on ETLs modified by TiCl4 and PC61BM
treatments, particularly for NRs treated under optimized conditions (100 × 10−3 m TiCl4 solution). This can be ascribed to
enhanced internal light scattering effects induced by the NR
surface roughness, which seems to increase the overall photon
harvesting by the perovskite. At the same time, the SEM images
in Figure S6 in the Supporting Information and the grain size
distribution statistics shown in Figure S7 in the Supporting
Information confirm that the optical properties of devices fabricated from differently treated ETLs cannot be ascribed to the
perovskite crystallographic features. The grain size statistics of
MAPbI3 layers grown on NRs appears in fact not to be affected
by different ETL surface treatments. Moreover, the XRD patterns in Figure 2 reveal that the crystallographic properties of
the perovskite grown on unmodified and modified ETLs do not
feature remarkable differences.
The PL spectra of perovskite layers formed on differently
treated ETLs are compiled in Figure 6b. Steady-state PL can
provide a measure of the quality of the perovskite film (e.g.,
to assess the effects of defect passivation) or of the quenching
efficiency (e.g., by charge extraction). The results point to a
more efficient charge extraction for the modified ETL/perovskite interface compared to perovskite layers grown on pristine (unmodified) NRs. The highest PL (i.e., worst quenching
yield) is measured for the perovskite layer grown on pristine
NRs. After TiCl4 or PC61BM modification of the ETL, the PL
intensity substantially decreased. Precisely, the most effective
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
Figure 6. a) Visible light absorption and b) PL spectra of MAPbI3 layers grown on differently treated TiO2 NRs; c,d) Results of impedance spectroscopy:
c) Nyquist plots measured under light illumination with an applied voltage of 0.8 V along with the relative equivalents circuit used for fitting;
and d) enlarged view of data in (c).
PL quenching is reached by a TiCl4 treatment with TiCl4 concentration of 100 × 10−3 m combined with PC61BM modification; the results are well in line with the data discussed above
and further prove that TiO2 NRs modified by a combined TiCl4
and PC61BM treatment can lead to a most efficient extraction of
photo-excited electrons from the light absorber.[41]
Solid-state IS was performed to determine the charge recombination properties of reference, PC61BM- and/or TiCl4-treated
devices.[42] The results are reported in Figure 6. The IS measurements were carried out under light illumination, in a highfrequency range (1 MHz to 1 Hz), and at an applied bias of
0.8 V. The semicircle in the Nyquist plots (see in Figure 6c a
magnified view of the spectra of different devices) is assigned
to the parallel combination of the recombination resistance and
the geometric capacitance of the devices. For data analysis, a
constant phase element (Q) instead of an ideal super-capacitor
element was used, in order to improve the quality of the fitting.
The equivalent electric parameters were extracted accordingly
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (7 of 9)
and are compiled in Table S3 in the Supporting Information.
From a diagnostic point of view, in previous work, it has been
reported that a high recombination resistance (i.e., low recombination of generated charges) results typically in highly efficient PSCs.[10] In line with the photovoltaic parameters, we
observed high recombination resistance for cells fabricated
from ETLs modified by dual TiCl4 and PC61BM treatment. This
once again confirms the positive effects of the dual surface
treatment in suppressing detrimental charge recombination
phenomena. According to the impedance results, the TiO2 NPs
(formed by TiCl4 treatment) seem to have a considerable effect
in decreasing the charge recombination rate. This is more likely
caused by the TiO2 NP blocking layer formed at the free FTO
surface, which prevents leakage current that could originate
from a direct perovskite/FTO contact. Here, we should mention
that the growth of TiO2 NRs can be ascribed to heterogeneous
nucleation (of TiO2 seeds on the FTO substrate) driven by lower
degrees of super-saturation.[27] The NR coverage (or NR density
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
at the FTO surface) can be determined by the competition
between seed nucleation and NR growth rates, which in turn is
affected by the experimental conditions (e.g., Ti precursor and
HCl concentrations). The experimental conditions here adopted
lead to rutile TiO2 NRs with an open structure, i.e., the NR density leaves empty space for the perovskite filling. While this on
the one side is beneficial for an effective infiltration and crystallization of the perovskite within the ETL, the free FTO surface
left behind negatively affects the cell functionality (perovskite/
FTO contact) and has therefore to be coated (blocking layer)
via an optimized TiCl4 treatment. Besides, by increasing the
concentration of the TiCl4 solution, the modified ETLs show
a gradually higher charge recombination resistance. This indicates that the TiCl4 modification is beneficial as it forms at the
NR surface a TiO2 NP buffer interface that enhances the charge
transfer.
3. Conclusion
We systematically explored the effects of multiple ETL surface
treatments on the efficiency of MAPbI3 PSCs. Our results
showed that also in the case of mesoporous nanostructured
ETLs the open-circuit voltage, and therefore the overall cell
efficiency, can be largely improved by an adequate engineering of the interface between the ETL and the perovskite
absorber. We observed that a combined TiCl4 and PC61BM
treatment is a pre-requisite to fabricate efficient PSCs from
ETLs consisting of arrays of single crystalline TiO2 NRs. The
dual treatment led to a champion efficiency of 19.5% and
enabled a nearly complete suppression of the device hysteresis. The performance enhancement originates from a more
efficient charge transfer across the ETL/perovskite interface,
which is caused by the passivation of defects and trap states at
the TiO2 ETL surface. This ETL surface engineering approach
can potentially provide a further cell performance enhancement if combined to better performing, mixed cation/anion
hybrid perovskite materials.
followed by heating at 110 °C for 5 min. For the preparation of the
perovskite solution, 710 mg of PbI2 (Lumtec) and 240 mg of CH3NH3I
(Lumtec) were dissolved in a mixture of DMF (dimethylformamide;
Sigma–Aldrich) and DMSO (dimethyl sulfoxide; Sigma–Aldrich) at
40 °C. The perovskite precursor was deposited on the PC61BM layer by
spin-coating at a speed of 4000 rpm for 15 s followed by blowing off the
leftover solvent with a N2 stream. The samples were then annealed on
a hot plate at 110 °C for 10 min. The hole transport material solution,
consisting of a 10 mg mL−1 solution of poly[5,5′-bis(2-butyloctyl)(2,2′-bithiophene)-4,4′-dicarboxylate-alt-5,5′-2,2′-bithiophene] (PDCBT,
1-Material) in chlorobenzene, was spin-coated on the perovskite layer at
2000 rpm for 30 s and annealed at 70 °C for 3 min. Finally, a 15-nm-thick
MoOx (Alfa Aesar) and 120-nm-thick Ag films were evaporated on
PDCBT by using a thermal evaporation system. A mask was used to
pattern the deposited MoOx and Ag films.
Characterization: Current density–voltage (J–V) and steady-state
current density (JMPP) characterization of the devices was carried
out under simulated solar light illumination (AM1.5G) with a light
intensity of 100 mW cm−2 (calibrated with a Si-cell). The EQE spectra
were recorded using an Enli Technology (Taiwan) EQE measurement
system (QE-R). Prior to measurement, the light intensity at each
wavelength was calibrated by using a standard single-crystal Si
photovoltaic cell.
The morphology and crystalline structure of the nanostructured TiO2
ETLs and perovskite layers were characterized by a field-emission SEM
(Fe-SEM, S4800, Hitachi) and by XRD (X’pert Philips MPD diffractometer,
using a Panalytical X’celerator detector and graphite monochromized Cu
Kα radiation, λ = 1.54056 A Å) respectively. The perovskite grain size
distribution was determined by using the software Image J.
UV–vis light absorption spectra were recorded using an Avantes
spectroscopy system (AvaLight-DH-S-BAL Balanced Power and AvaSpecULS2048L StarLine Versatile Fiber-optic Spectrometer) equipped with
an integrating sphere (a BaSO4 standard white board was used as
reference).
Solid-state IS measurements were carried out under 0.1 sun
illumination, at 0.8V and in a frequency range from 1 MHz to 1 Hz, by
using of an Agilent HP 4192A impedance analyzer. Steady-state PL was
measured using an Argon-ion 488-nm laser excitation source, a Horiba
monochromator, and a Silicone detector.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
4. Experimental Section
Preparation of Rutile TiO2 Nanorod ETL: Laser patterned FTO
substrates (7 Ω m−2, Solaronix) were cleaned by ultrasonication in
acetone first, then in isopropanol, and were then dried in a stream
of nitrogen. TiO2 NRs were grown directly on the FTO surfaces by
a hydrothermal process.[43] Typically, 1 mL of titanium isopropoxide
(Sigma–Aldrich) was added to an equal volume (50 mL: 50 mL) of a
mixture of DI water and HCl (37%, Sigma–Aldrich) in a 250 mL Teflonlined autoclave. This mixture was stirred for 15 min. Then, the pre-cleaned
FTO was immersed in the solution facing down. The hydrothermal
synthesis was performed at 180°C for 2.5 h. Finally, the as-prepared NRs
were annealed at 450 °C for 1 h in air furnace. For the TiCl4 treatment,
experimental conditions reported elsewhere were adopted.[44] briefly,
the TiO2 NRs were immersed in aqueous TiCl4 solutions of different
concentrations (50, 100, and 200 × 10−3 m) at 70 °C for 30 min. The
samples were then heat-treated at 450 °C for 10 min to crystallize the
deposited NPs into anatase phase.
Device Fabrication: The as-prepared FTO supported NR layers (ETL)
were moved to a nitrogen-filled glovebox and processed as substrates
to fabricate the PSCs. PC61BM (99.5%, Solenne) solution (10 mg mL−1
in chlorobenzene, Sigma–Aldrich) was spin-coated on the substrate
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (8 of 9)
Acknowledgements
The authors would like to acknowledge the ERC, the DFG, and the DFG
cluster of excellence “Engineering of Advanced Materials” for financial
support. Gihoon Cha (Department of Materials Science and Engineering
WW4-LKO, University of Erlangen-Nuremberg, Germany) and Dr. Nhat
Truong Nguyen (current affiliation: GAO Materials Chemistry Research
Group, Department of Chemistry, University of Toronto, Canada) are
acknowledged for their help with XRD and optical measurements,
respectively. N.L. gratefully acknowledges the financial support from the
DFG research grant: BR 4031/13-1. C.J.B. gratefully acknowledges the
financial support through the “Aufbruch Bayern” initiative of the state
of Bavaria (EnCN and SFF), the Bavarian Initiative “Solar Technologies
go Hybrid” (SolTech), and the projects SFB 953 (DFG, project no.
182849149) and DFG INST 90/917-1 FUGG.
Conflict of Interest
The authors declare no conflict of interest.
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com
www.afm-journal.de
Keywords
defect passivation, PC61BM, perovskite solar cells, TiCl4, TiO2 nanorods
Received: November 21, 2019
Revised: December 18, 2019
Published online: January 9, 2020
[1] C. Chen, Y. Cheng, Q. Dai, H. Song, Sci. Rep. 2016, 5, 17684.
[2] N. J. Jeon, J. H. Noh, W. S. Yang, Y. C. Kim, S. Ryu, J. Seo, S. Il Seok,
Nature 2015, 517, 476.
[3] M. Abdi-Jalebi, M. I. Dar, A. Sadhanala, S. P. Senanayak,
F. Giordano, S. M. Zakeeruddin, M. Grätzel, R. H. Friend,
J. Phys. Chem. Lett. 2016, 7, 3264.
[4] J. Min, Z. Zhang, Y. Hou, C. O. Ramirez Quiroz, T. Przybilla,
C. Bronnbauer, F. Guo, K. Forberich, H. Azimi, T. Ameri, E. Spiecker,
Y. Li, C. J. Brabec, Chem. Mater. 2015, 27, 227.
[5] F. Guo, H. Azimi, Y. Hou, T. Przybilla, M. Hu, C. Bronnbauer,
S. Langner, E. Spiecker, K. Forberich, C. J. Brabec, Nanoscale 2015,
7, 1642.
[6] S. Ameen, M. Nazim, M. S. Akhtar, M. K. Nazeeruddin, H. Shin,
Nanoscale 2017, 9, 17544.
[7] Y. Hou, W. Chen, D. Baran, T. Stubhan, N. A. Luechinger,
B. Hartmeier, M. Richter, J. Min, S. Chen, C. O. R. Quiroz, N. Li,
H. Zhang, T. Heumueller, G. J. Matt, A. Osvet, K. Forberich,
Z. Zhang, Y. Li, B. Winter, P. Schweizer, E. Spiecker, C. J. Brabec,
Adv. Mater. 2016, 28, 5112.
[8] S. Wu, C. Chen, J. Wang, J. Xiao, T. Peng, ACS Appl. Energy Mater.
2018, 1, 1649.
[9] W. Wu, L. Wang, Adv. Funct. Mater. 2018, 28, 1804356.
[10] L. Yang, X. Wang, X. Mai, T. Wang, C. Wang, X. Li, V. Murugadoss,
Q. Shao, S. Angaiah, Z. Guo, J. Colloid Interface Sci. 2019, 534, 459.
[11] J. Ma, J. Chang, Z. Lin, X. Guo, L. Zhou, Z. Liu, H. Xi, D. Chen,
C. Zhang, Y. Hao, J. Phys. Chem. C 2018, 122, 1044.
[12] H.-S. Kim, J.-W. Lee, N. Yantara, P. P. Boix, S. A. Kulkarni,
S. Mhaisalkar, M. Grätzel, N.-G. Park, Nano Lett. 2013, 13, 2412.
[13] H. Sun, K. Deng, Y. Zhu, M. Liao, J. Xiong, Y. Li, L. Li, Adv. Mater.
2018, 30, 1801935.
[14] Z. Lan, X. Xu, X. Zhang, J. Tang, L. Zhang, X. He, J. Wu, J. Mater.
Chem. C 2018, 6, 334.
[15] R. Salazar, M. Altomare, K. Lee, J. Tripathy, R. Kirchgeorg,
N. T. Nguyen, M. Mokhtar, A. Alshehri, S. A. Al-Thabaiti,
P. Schmuki, ChemElectroChem 2015, 2, 824.
[16] X. Gao, J. Li, J. Baker, Y. Hou, D. Guan, J. Chen, C. Yuan,
Chem. Commun. 2014, 50, 6368.
[17] A. Nawaz, K. K. Wong, C. Ebenhoch, E. Zimmermann, Z. Zheng,
M. N. Akram, J. Kalb, K. Wang, A. Fakharuddin, L. Schmidt-Mende,
Sol. Energy 2018, 170, 541.
[18] P. Qin, M. Paulose, M. I. Dar, T. Moehl, N. Arora, P. Gao,
O. K. Varghese, M. Grätzel, M. K. Nazeeruddin, Small 2015, 11,
5533.
Adv. Funct. Mater. 2020, 30, 1909738
1909738 (9 of 9)
[19] G. Choe, J. Kang, I. Ryu, S. W. Song, H. M. Kim, S. Yim, Sol. Energy
2017, 155, 1148.
[20] L. Cojocaru, S. Uchida, Y. Sanehira, J. Nakazaki, T. Kubo, H. Segawa,
Chem. Lett. 2015, 44, 674.
[21] A. Yella, L. Heiniger, P. Gao, M. K. Nazeeruddin, M. Grätzel,
Nano Lett. 2014, 14, 2591.
[22] Z. Liu, Q. Chen, Z. Hong, H. Zhou, X. Xu, N. De Marco, P. Sun,
Z. Zhao, Y. Cheng, Y. Yang, ACS Appl. Mater. Interfaces 2016, 8,
11076.
[23] Z. Zhou, S. Pang, Z. Liu, H. Xu, G. Cui, J. Mater. Chem. A 2015, 3,
19205.
[24] J. Kim, G. Kim, T. K. Kim, S. Kwon, H. Back, J. Lee, S. H. Lee,
H. Kang, K. Lee, J. Mater. Chem. A 2014, 2, 17291.
[25] H. K. Adli, T. Harada, S. Nakanishi, S. Ikeda, Phys. Chem. Chem.
Phys. 2017, 19, 26898.
[26] B. Liu, E. S. Aydil, J. Am. Chem. Soc. 2009, 131, 3985.
[27] E. Hosono, S. Fujihara, K. Kakiuchi, H. Imai, J. Am. Chem. Soc.
2004, 126, 7790.
[28] N. T. Nguyen, S. Ozkan, I. Hwang, A. Mazare, P. Schmuki,
Nanoscale 2016, 8, 16868.
[29] S. Feng, A. Runa, L. Liu, J. Wang, P. Su, T. Liu, S. Su, G. Zhu, W. Fu,
H. Yang, J. Mater. Sci. Mater. Electron. 2018, 29, 16903.
[30] C. Liu, R. Zhu, A. Ng, Z. Ren, S. H. Cheung, L. Du, S. K. So,
J. A. Zapien, A. B. Djurišic′ , D. Lee Phillips, C. Surya, J. Mater. Chem.
A 2017, 5, 15970.
[31] J. Yang, W. Liao, J. Wu, ACS Appl. Mater. Interfaces 2013, 5, 7425.
[32] F. Cai, J. Cai, L. Yang, W. Li, R. S. Gurney, H. Yi, A. Iraqi, D. Liu,
T. Wang, Nano Energy 2018, 45, 28.
[33] K. Wojciechowski, S. D. Stranks, A. Abate, G. Sadoughi,
A. Sadhanala, N. Kopidakis, G. Rumbles, C. Li, R. H. Friend,
A. K.-Y. Jen, H. J. Snaith, ACS Nano 2014, 8, 12701.
[34] C. Tao, S. Neutzner, L. Colella, S. Marras, A. R. Srimath Kandada,
M. Gandini, M. De Bastiani, G. Pace, L. Manna, M. Caironi,
C. Bertarelli, A. Petrozza, Energy Environ. Sci. 2015, 8, 2365.
[35] Q. Wang, Y. Shao, Q. Dong, Z. Xiao, Y. Yuan, J. Huang, Energy
Environ. Sci. 2014, 7, 2359.
[36] B. Cao, H. Liu, L. Yang, X. Li, H. Liu, P. Dong, X. Mai, C. Hou,
N. Wang, J. Zhang, J. Fan, Q. Gao, Z. Guo, ACS Appl. Mater.
Interfaces 2019, 11, 33770.
[37] Y. Zhao, A. M. Nardes, K. Zhu, Faraday Discuss. 2014, 176, 301.
[38] Y. Shao, Y. Yuan, J. Huang, Nat. Energy 2016, 1, 15001.
[39] X. Li, S.-M. Dai, P. Zhu, L.-L. Deng, S.-Y. Xie, Q. Cui, H. Chen,
N. Wang, H. Lin, ACS Appl. Mater. Interfaces 2016, 8, 21358.
[40] B. Chen, M. Yang, X. Zheng, C. Wu, W. Li, Y. Yan, J. Bisquert, G. GarciaBelmonte, K. Zhu, S. Priya, J. Phys. Chem. Lett. 2015, 6, 4693.
[41] W. Zhou, J. Zhen, Q. Liu, Z. Fang, D. Li, P. Zhou, T. Chen, S. Yang,
J. Mater. Chem. A 2017, 5, 1724.
[42] H. Kim, I. Mora-Sero, V. Gonzalez-Pedro, F. Fabregat-Santiago,
E. J. Juarez-Perez, N. Park, J. Bisquert, Nat. Commun. 2013, 4, 2242.
[43] F. Gao, H. Dai, H. Pan, Y. Chen, J. Wang, Z. Chen, J. Colloid Interface
Sci. 2018, 513, 693.
[44] P. Roy, D. Kim, I. Paramasivam, P. Schmuki, Electrochem. Commun.
2009, 11, 1001.
© 2020 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Copyright WILEY-VCH Verlag GmbH & Co. KGaA, 69469 Weinheim, Germany, 2020.
Supporting Information
for Adv. Funct. Mater., DOI: 10.1002/adfm.201909738
Engineering of the Electron Transport Layer/Perovskite
Interface in Solar Cells Designed on TiO2 Rutile Nanorods
Fahimeh Shahvaranfard, Marco Altomare,* Yi Hou,
Seyedsina Hejazi, Wei Meng, Benedict Osuagwu, Ning Li,
Christoph J. Brabec, and Patrik Schmuki*
Supporting Information
Engineering of the electron transport layer/perovskite interface in solar cells designed
on TiO2 rutile nanorods
Fahimeh Shahvaranfard, Marco Altomare*, Yi Hou, Seyedsina Hejazi, Wei Meng, Benedict
Osuagwu, Ning Li, Christoph J. Brabec, Patrik Schmuki*
F. Shahvaranfard, Dr. M. Altomare, S. Hejazi, B. Osuagwu, Prof. P. Schmuki
Institute for Surface Science and Corrosion WW4-LKO, Department of Materials Science and
Engineering, University of Erlangen-Nuremberg, Martensstrasse 7, 91058 Erlangen, Germany
E-mail:
marco.altomare@fau.de
schmuki@ww.uni-erlangen.de
Dr. Y. Hou, W. Meng, Dr. N. Li, Prof. C.J. Brabec
Institute of Materials for Electronics and Energy Technology (i-MEET), Department of
Materials Science and Engineering, University of Erlangen-Nuremberg, Martensstrasse 7,
91058 Erlangen, Germany
Prof. P. Schmuki,
Chemistry Department, Faculty of Sciences, King Abdulaziz University, 80203 Jeddah,
Kingdom of Saudi Arabia
Dr. N. Li, Prof. C.J. Brabec
Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (HI ERN), Immerwahrstr. 2,
91058, Erlangen, Germany
Dr. N. Li
National Engineering Research Center for Advanced Polymer Processing Technology,
Zhengzhou University, 450002, Zhengzhou, China
1
Figure S1. Top view SEM images of TiO2 nanorods decorated with different TiO2
nanoparticle loadings by TiCl4 treatment with different TiCl4 solution concentrations: (a)
pristine NRs; (b) 50 mM; (c) 100 mM; (d) 200 mM.
Figure S2. Top view SEM image of an FTO slide treated in a 100 mM TiCl4 solution.
2
Figure S3. Current density – voltage characteristics of the best device fabricated from a
TiCl4-treated FTO slide (100 mM TiCl4 solution).
Figure S4. A) Steady-State current density of the best-performing device at 0.8 V; and b)
photovoltaic performance measured for the as-fabricated device and after storing the device
for 7 and 14 days.
3
Figure S5. UV-Vis light absorption spectra of differently treated TiO2 NRs.
4
Figure S6. Top view SEM images of perovskite layers grown on different substrates: (a)
pristine FTO; (b) FTO / NRs; (c) FTO / NRs / 100 mM TiCl4; (d) FTO /NRs / PC61BM; and
(e) FTO /NRs / 100 mM TiCl4 / PC61BM.
5
Figure S7. Grain size distribution statistics of perovskite layers grown on different substrates:
(a) pristine FTO; (b) FTO / NRs; (c) FTO / NRs / 100 mM TiCl4; (d) FTO /NRs / PC61BM;
and (e) FTO /NRs / 100 mM TiCl4 / PC61BM.
6
Table S1. Summary of the photovoltaic parameters of the best device fabricated from a TiCl4treated FTO slide (100 mM TiCl4 solution).
ETL construction
TiCl4-treated FTO
JSC
(mA cm-2)
21.5
VOC
(V)
0.95
FF
(%)
52
PCE
(%)
10.6
Table S2. Summary of the device performance data for perovskite solar cells based on
different light absorbers and constructed on ETLs consisting of arrays of 1D TiO 2
nanostructures.
Light absorber
ETL
Efficiency
Year
Ref.
CH3NH3PbI3
TiO2 NR array
9.4%
2013
[1]
CH3NH3PbI3
TiO2 nanowire
14.71%
2015
[2]
CH3NH3PbI3
TiO2 NR array
18.22%
2016
[3]
CH3NH3PbI3
TiO2 NR array
11.7%
2018
[4]
CH3NH3PbI3
TiO2 NR array
10.15%
2018
[5]
CH3NH3PbI3
TiO2 NR array
16.57%
2019
[6]
CH3NH3PbI3
TiO2/CdS coreshell NR array
17.71%
2019
[7]
CH3NH3PbI3
TiO2 NR array
19.5%
-
CH3NH3PbI3-xClx
TiO2 NR array
12.2%
2015
[8]
Mixture of PbI2, CH3NH3I
(MAI) and CH3NH3Cl
(MACl)
TiO2 NR array
19.02%
2017
[9]
MAPbI3-xClx
TiO2 NR array
14.3%
2018
[10]
TiO2 NR array
18.76%
2018
[11]
TiO2 NR array
19.33%
2019
[12]
TiO2 NR array
20.28%
2019
[13]
Csx(MA0.17FA0.83)(100-x)
Pb(I0.83Br0.17)3
Cs0.1(FA0.83MA0.17)0.9Pb
(I0.83Br0.17)3
K0.05Cs0.05(MA0.17FA0.83)0.9
Pb(I0.83Br0.17)3
7
This
work
Table S3. Impedance spectroscopy parameters of solar cells based on different ETLs.
ETL construction
TiO2 NRs
TiO2 NRs / 100 mM TiCl4
TiO2 NRs / PC61BM
TiO2 NRs / 50 mM TiCl4 / PC61BM
TiO2 NRs / 100 mM TiCl4 / PC61BM
TiO2 NRs / 200 mM TiCl4 / PC61BM
8
Rs (Ω)
Rrec(kΩ)
22
23
64
25
29
22
0.34
5.42
0.67
4.36
6.36
9.13
CPE
Q (nF)
a
51
0.89
32.5
0.97
23
0.94
34
0.91
30
0.92
32
0.91
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
H.-S. Kim, J.-W. Lee, N. Yantara, P. P. Boix, S. A. Kulkarni, S. Mhaisalkar, M.
Grätzel, N.-G. Park, Nano Lett. 2013, 13, 2412.
J. Im, J. Luo, M. Franckevičius, N. Pellet, P. Gao, T. Moehl, S. M. Zakeeruddin, M. K.
Nazeeruddin, M. Grätzel, N. Park, Nano Lett. 2015, 15, 2120.
X. Li, S.-M. Dai, P. Zhu, L.-L. Deng, S.-Y. Xie, Q. Cui, H. Chen, N. Wang, H. Lin,
ACS Appl. Mater. Interfaces 2016, 8, 21358.
F. Gao, H. Dai, H. Pan, Y. Chen, J. Wang, Z. Chen, J. Colloid Interface Sci. 2018, 513,
693.
W. Liu, L. Chu, R. Hu, R. Zhang, Y. Ma, Y. Pu, J. Zhang, J. Yang, X. Li, W. Huang,
Sol. Energy 2018, 166, 42.
R. Li, H. Zhang, R. Chai, M. Zhang, M. Guo, Ceram. Int. 2019, 45, 12353.
W. Liu, L. Chu, N. Liu, Y. Cheng, F. Wu, Y. Li, Y. Pu, J. Zhang, X. Li, W. Huang,
Adv. Mater. Interfaces 2019, 6, 1801976.
A. Fakharuddin, F. Di Giacomo, I. Ahmed, Q. Wali, T. M. Brown, R. Jose, J. Power
Sources 2015, 283, 61.
C. Liu, R. Zhu, A. Ng, Z. Ren, S. H. Cheung, L. Du, S. K. So, J. A. Zapien, A. B.
Djurišić, D. Lee Phillips, C. Surya, J. Mater. Chem. A 2017, 5, 15970.
S. Wu, C. Chen, J. Wang, J. Xiao, T. Peng, ACS Appl. Energy Mater. 2018, 1, 1649.
Z. Lan, X. Xu, X. Zhang, J. Tang, L. Zhang, X. He, J. Wu, J. Mater. Chem. C 2018, 6,
334.
L. Yang, X. Wang, X. Mai, T. Wang, C. Wang, X. Li, V. Murugadoss, Q. Shao, S.
Angaiah, Z. Guo, J. Colloid Interface Sci. 2019, 534, 459.
B. Cao, H. Liu, L. Yang, X. Li, H. Liu, P. Dong, X. Mai, C. Hou, N. Wang, J. Zhang, J.
Fan, Q. Gao, Z. Guo, ACS Appl. Mater. Interfaces 2019, 11, 33770.
9
Appendix B
Comparison of the sputtered TiO2 anatase and rutile thin films as electron
transporting layers in perovskite solar cells
Fahimeh Shahvaranfard, Ning Li, Saman Hosseinpour, Seyedsina Hejazi, Kaicheng Zhang,
Marco Altomare, Patrik Schmuki, Christoph J. Brabec
Nano Select, 2021;1–8
74
Received: 5 October 2021
Accepted: 7 October 2021
DOI: 10.1002/nano.202100306
RESEARCH ARTICLE
Comparison of the sputtered TiO2 anatase and rutile thin
films as electron transporting layers in perovskite solar cells
Fahimeh Shahvaranfard1
Kaicheng Zhang1
Ning Li1,2
Marco Altomare4
Saman Hosseinpour3
Patrik Schmuki4
Seyedsina Hejazi4
Christoph J. Brabec1,2
1 Department of Materials Science and Engineering, Institute of Materials for Electronics and Energy Technology (i-MEET), University of
Erlangen-Nuremberg, Erlangen, Germany
2 Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (HI ERN), Erlangen, Germany
3 Institute of Particle Technology (LFG), Friedrich-Alexander-Universität-Erlangen-Nürnberg (FAU), Erlangen, Germany
4 Department of Materials Science and Engineering, Institute for Surface Science and Corrosion WW4-LKO, University of Erlangen-Nuremberg,
Erlangen, Germany
Correspondence
Fahimeh Shahvaranfard, Ning Li and
Christoph J. Brabec, Department of Materials Science and Engineering, Institute
of Materials for Electronics and Energy
Technology (i-MEET), University of
Erlangen-Nuremberg, Martensstrasse
7, 91058 Erlangen, Germany.
Email: fahimeh.shahvaranfard@fau.de
Helmholtz Institute Erlangen-Nürnberg
for Renewable Energy (HI ERN), Erlangen, Germany
Email: ning.li@fau.de;
christoph.brabec@fau.de
We examine comparatively the performance of sputtered TiO2 rutile and anatase
thin films as an electron transport layer (ETL) in MAPbI3 -based perovskite solar
cells. Both anatase and rutile TiO2 ETLs are deposited (on fluorine-doped tin
oxide [FTO] substrates) by magnetron sputtering in the form of nanocrystalline
thin films. We systematically investigate the role of crystallographic phase composition of TiO2 ETLs on the photovoltaic performance of perovskite solar cells.
The champion power conversion efficiencies (PCEs) of 18.4% and 17.7% under
reverse scan mode are obtained for perovskite solar cells based on TiO2 anatase
and TiO2 rutile ETL, respectively. The results show that the magnetron sputtering deposited ETLs differ from each other only in their phase composition while
the overall performance of the devices is not greatly affected by the crystallographic phase of the TiO2 ETLs. Our results point to an important fact that for
a proper and reliable comparison between the performance of TiO2 anatase and
rutile ETLs, it is crucial to investigate films of similar morphology and structure
that are synthesize under similar conditions.
KEYWORDS
magnetron sputtering, perovskite solar cell, TiO2 anatase, TiO2 rutile
1
INTRODUCTION
Organic-inorganic hybrid halide perovskite solar cells
(PSCs) are one of the most promising candidates for energy
conversion in the photovoltaic field due to the outstanding
optical and electronic properties that lead to a record efficiency of over 25%.[1–3]
In PSCs, the perovskite light absorber is sandwiched
between an electron transport layer (ETL) and a hole transport layer (HTL).[4–6] ETL, as one of the key functional
components in PSCs, is often comprised of a metal oxide
scaffold (e.g., TiO2 ,[7,8] SnO2 ,[9,10] or ZnO[11,12] ). Properties
of ETL, such as transparency, surface roughness, electron
extraction ability, and bands’ position, play key roles in the
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the
original work is properly cited.
© 2021 The Authors. Nano Select published by Wiley-VCH GmbH
Nano Select 2021;1–8.
wileyonlinelibrary.com/journal/nano
1
2
overall cell efficiency. Among different metal oxides developed as ETL, TiO2 is the most frequent choice owing to its
high chemical and photochemical stability, relatively high
electron mobility, and low cost.[13–17]
Anatase and rutile are the most common crystal forms
of TiO2 . Rutile is generally considered to be the thermodynamically most stable phase for bulk titania, whereas the
anatase phase typically becomes more stable in nanostructured TiO2 .[18,19]
Recent works have compared anatase and rutile TiO2
ETLs and their crystalline phase-dependent charge extraction performance as well as the effect of the TiO2 phase on
the efficiency of perovskite solar cells. For instance, Zhu
et al.[20] have reported on the improvement of the device
performance using anatase TiO2 compared to the device
fabricated with amorphous and rutile TiO2 . Yella et al.[21]
found that nanocrystalline rutile TiO2 performed better
than a planar anatase TiO2 , an improvement which was
ascribed to the large nanocrystalline rutile TiO2 /perovskite
interface area for electron extraction.
In a recent work by Wang et al.,[22] TiO2 rutile and TiO2
anatase ETL were synthesized by two different methods;
chemical bath deposition and spin coating. Their results
indicated that rutile TiO2 ETL that was synthesized by
chemical bath deposition helped to improve the extraction of photoexcited electrons and decreased electronhole recombination, due to its better conductivity, and
enhanced ETL/perovskite light absorber interface compared to anatase TiO2 ETL that was synthesized by the spin
coating method. However, due to the different techniques
used to produce anatase versus rutile TiO2 , it may not yield
comparable ETL morphological and structural features.
In this study, we aim to systematically compare the performance of anatase TiO2 (A-TiO2 ) versus rutile TiO2 (RTiO2 ) as ETL that are synthesized with magnetron sputtering in methylammonium lead iodide (MAPI) perovskite
solar cells. Nanocrystalline anatase and rutile TiO2 thin
films are deposited on fluorine-doped tin oxide (FTO) substrates. The sputtering conditions are not only adjusted to
accurately control the ETL crystallographic phase (as pure
rutile or pure anatase) by tuning the chamber pressure during sputtering, but also tuned to achieve ETLs with comparable morphological and structural features, for example,
in terms of surface roughness, thickness, and average crystallite size. We show that this is a key point for a proper
comparison between the role of different phases TiO2 in
perovskite solar cells.
Our results, in fact, show that comparing ETLs of different phases without considering other differences in the
properties of the films may be misleading in elucidating the
impact of the ETL crystal structure on the overall device
efficiency. Here we report that the performance of perovskite solar cells is not substantially affected by the crys-
SHAHVARANFARD et al.
tallographic phase of the TiO2 ETLs. With this work we
want to draw attention to the important of a proper comparison between the performance of R-TiO2 and A-TiO2 as
ETLs. Such a proper comparison requires that layers are
(i) synthesized by the same method, and (ii) do not differ
from each other in their morphological or micro-structural
features. When such requirements are not met, the device
performance may be simultaneously influenced by various
parameters, for example, surface morphology, crystal size,
and wettability, thus making a direct anatase versus rutile
phase comparison unreliable.
It should be noted that here we focus on the factors that,
besides crystalline structure, affect the performance of the
ETLs. Indeed, high-performance light absorbers, such as
mixed cation/anion hybrid perovskite materials instead of
MAPI, further improves the efficiency of the device.
2
RESULTS AND DISCUSSION
The FTO substrates coated with TiO2 film were characterized by XRD to identify their crystallographic composition. The XRD patterns of 200 nm high pressure (HP)TiO2 and low pressure (LP)-TiO2 before and after thermal treatment in air at 450◦ C for 1 hour are shown in
Figures 1A and S1a, respectively. No information on the
crystallographic phase of 30 nm sputtered TiO2 film could
be traced by XRD, probably due to the low thickness of
the layers which is below the detection limit of the XRD
setup used. The XRD patterns confirm mainly anatase
crystalline phase composition of as-formed HP-TiO2 and
mainly rutile crystalline phase composition of as-formed
LP-TiO2 (Figure S1a). Comparing Figure S1a and Figure 1A
indicates that the degree of phase crystallinity increased
with thermal treatment. The XRD pattern of HP-TiO2
shows diffraction peaks at 25.36◦ , 48.03◦ ,and 54.75◦ which
correspond to 101, 200, and 105 crystallographic planes
of A-TiO2 , respectively.[23,24] The XRD pattern of LP-TiO2
shows diffraction peaks at 27.57◦ , 36◦ , 41.37◦ , and 54.69◦
corresponding to 110, 101, 111, and 211 crystallographic
planes of R-TiO2 , respectively.[23,24] The phase characterization of the 30 nm thick TiO2 films was performed by
confocal Raman spectroscopy and the corresponding spectra are provided in Figure 1B. The characteristic peaks
corresponding to anatase (at 397, 516, and 638 cm−1 )[25]
appear as sharp peaks in Figure 1B whereas the characteristic peaks corresponding to rutile (at 241, 445, and
∼610 cm−1 )[25] are less pronounced, due to the large overlap with the FTO characteristic peaks (see the S1b).
Figure 2A-D shows top view and cross-sectional SEM
images of 30 nm sputtered TiO2 film on FTO substrate
formed by magnetron sputtering at different chamber pressures. As shown in Figure 2A,B, the TiO2 films form a
SHAHVARANFARD et al.
3
(B)
(A)
500 nm
(C)
500 nm
(D)
27.5 nm
31.3 nm
100 nm
200 nm
(E)
(F)
RMS : 34.4 nm
RMS : 36.2 nm
F I G U R E 2 Top view and cross-sectional SEM images of (A
and C) A-TiO2 , (B and D) R-TiO2 ; AFM images of (E) A-TiO2 , (F)
R-TiO2 on FTO
F I G U R E 1 A, X-ray diffraction (XRD) patterns of 200 nm
HP-TiO2 and LP-TiO2 on FTO. Crystallite size of anatase and rutile
films was calculated from the most intense peak from XRD pattern
and Scherrer equation and is ∼100 nm. B, Raman spectra of the
30 nm HP-TiO2 and LP-TiO2 on FTO. The Raman spectra are
background corrected for the FTO coated glass substrate
continuous layer on the surface of the FTO regardless of
the sputtering pressure. The thickness of as-sputtered layers is ∼30 nm for both anatase and rutile TiO2 films (Figure 2C,D).
To further study the surface roughness, 30 nm A-TiO2
and R-TiO2 films on FTO were characterized by AFM (Figure 2E,F). The root mean square (RMS) for A-TiO2 and RTiO2 were 34.4 and 36.2 nm, respectively, which are virtually identical. For comparison, Figure S2 shows the top
view SEM image and AFM characterization of pristine
FTO. According to these results, it becomes evident that
the observed surface roughness and grain size of the TiO2
layers (Figure 2) are not entirely related to the TiO2 layer
and are predominantly affected by the FTO substrate.
The contact angle of the water droplet was measured
to investigate the wettability of two types of TiO2 films
and the formation of perovskite light absorber on TiO2
ETLs.[26,27] The contact angle of water droplets on A-TiO2
and R-TiO2 films is illustrated in Figure S3. The contact
angle of water on A-TiO2 and R-TiO2 was 65.4◦ and 68.7◦ ,
respectively, confirming similar surface wetting properties
of the two types of TiO2 ETLs.
As shown in Figure S4, similar grain size distributions
are found for perovskite layers deposited on the two ETLs.
The grain size distribution statistics extracted from SEM
imaging exhibited comparable feature sizes for the two
samples (Figure S4) suggesting that the minor differences
in roughness and wettability of TiO2 anatase and rutile
films do not impact perovskite layer growth. We conclude
that the similar surface morphology and structural features
of TiO2 anatase and rutile layers that are deposited by magnetron sputtering indeed result in comparable perovskite
layers morphology.
Perovskite solar cells were fabricated with a device architecture of FTO/TiO2 -ETL/MAPbI3 /PDCBT/MoOx /Ag.
Figure 3 shows cross-sectional SEM images of full devices
fabricated with different ETLs. Cross-sectional SEM
images of both devices show homogenous coverage of the
TiO2 layers.
To assess the photovoltaic performance of the PSCs, JV curves were measured in the forward scan mode under
AM 1.5 G illumination (100 mW cm−2 ) and the results are
compiled in Figure 4. The average performance parameters of both cells are summed up in Table 1. According to
the results, devices based on A-TiO2 show an average photovoltaic conversion efficiency (PCE) of 16%, a short circuit
4
SHAHVARANFARD et al.
F I G U R E 3 Cross-sectional SEM image
of full device fabricated with (A) 30 nm
A-TiO2 (B) 30 nm R-TiO2
current density (JSC ) of 22.2 mA cm−2 , an open circuit voltage (VOC ) of 1.08 V, and a fill factor (FF) of 67%. The best
performing device fabricated with A-TiO2 obtained a PCE
of 17.2%, a JSC of 22.6 mA cm−2 , a VOC of 1.09 V, and a FF of
70%. For devices fabricated with R-TiO2 , an average PCE of
15.6%, a JSC of 22 mA cm−2 , a VOC of 1.07 V, and a FF of 66%
are achieved. The best performing device fabricated with
R-TiO2 showed a PCE of 16.7%, a JSC of 22.8 mA cm−2 , a
VOC of 1.07 V, and a FF of 68%. Figure 4C-F shows the characterization of 20 devices constructed with different TiO2
crystallographic phase ETLs in the forward scan. It should
be noted that A-TiO2 devices show slightly higher performance parameters but not to a significant extent. To further validate the current density obtained from J-V curves,
EQE measurements were performed for devices fabricated
with A-TiO2 and R-TiO2 and the results are shown in Figure 4G. The JSC values for A-TiO2 and R-TiO2 are 20.6 and
20.3 mA cm−2 , respectively, which are consistent with the
average JSC value from the J-V curves. As shown in Figure 4A,B, the same trend is observed in the cell performance loss for both devices constructed with A-TiO2 and
R-TiO2 ETL under forward and reverse bias scans. This
indicates that hysteresis is not impacted by using A-TiO2
versus R-TiO2 ETL in these perovskite solar cells.
According to these results, we found that the photovoltaic performance of the devices is not majorly affected
by the crystallographic phase of the TiO2 ETLs. To further
clarify this hypothesis, we carried out further investigations on the structural and optical properties of the fabricated TiO2 layers.
Kelvin probe measurements, which provide information
on the surface electronic properties,[28,29] were employed
to measure the work function of TiO2 ETLs. Table 2 shows
the average work function of 30 nm anatase and rutile layers. A minor work function difference of about 50 meV
was observed for anatase (4.27 eV) versus rutile (4.22 eV),
suggesting that both TiO2 layers would follow similar
energy band alignments with reference to perovskite light
absorber.[30]
The optical properties of 30 nm A-TiO2 and 30 nm
R-TiO2 on FTO were further investigated to understand
whether the small differences in the blue regime of the
EQE can be related to different optical properties of the
interface layer. Figure 5 compares the transmission (T),
reflection (R), and absorption (A) spectra that were separately measured in an integrated sphere setup. We notice
that the spectra of A-TiO2 versus R-TiO2 are in fact very
similar. Small differences in the interference peaks (best
seen in the reflection curve) underline that both samples
not only have a similar layer thickness but also a comparable refractive index. As a control for the accuracy of the
measurements, the sum of transmittance, reflectance, and
absorptance (i.e., the trend line A+T+R) is also plotted in
Figure 5, showing that the deviation from 100% is less than
1.5%. Based on the optical analysis of the two TiO2 layers,
we conclude that the minor difference in cell performance
or the small blue shift in the EQE are not caused by the systematically different optical properties of the TiO2 layers.
Steady-state photoluminescence (PL) spectra of perovskite
(MAPI) layer on FTO, FTO/A-TiO2 and FTO/R-TiO2 are
shown in Figure 6A. The PL results confirm an improvement in charge extraction for both A & R-TiO2 /perovskite
interface compared to bare FTO/perovskite. However, both
A-TiO2 and R-TiO2 show comparable PL quenching. The
PL results further prove that anatase and rutile TiO2 thin
films act very similarly in the extraction of photo-induced
electrons from the perovskite layer.
Impedance measurements were performed at 0.8 V in
the dark to assess the recombination behavior of the device
with A-TiO2 and R-TiO2 at a bias representative for the
maximum power point (MPP). The Nyquist plots of devices
are shown in Figure 6B. The semicircle in the Nyquist
plots is discussed in terms of the most simple transient
replacement circuit, where the recombination resistance
is parallel to the geometric capacitance of the cell and in
series with a further resistance representing all parasitic as
well as internal series resistance losses.[30,31] Table S1 summarizes the extracted electrical parameters from the equivalent circuit model presented as the inset in Figure 6B.
The series resistance is slightly smaller for A-TiO2 than
its R-TiO2 counterpart; however, differences in the series
resistance of 10–20% are not expected to impact the device
performance. Similarly, a slightly higher recombination
resistance, which is indicative of reduced recombination
SHAHVARANFARD et al.
5
25
25
(A)
20
Forward
Reverse
15
10
5
PEC=18.4%
Jsc=23.1mAcm-2
VOC= 1.1V
FF=73%
0
0.0
0.2
0.4
JSC(mAcm-2)
JSC(mAcm-2)
20
PEC=17.2%
Jsc=22.6mAcm-2
VOC= 1.09V
FF=70%
0.6
0.8
15
10
5
1.0
Forward
Reverse
PEC=17.7%
Jsc=22.8mAcm-2
VOC= 1.09V
FF=71%
0
0.0
1.2
0.2
0.4
PEC=16.7%
Jsc=22.8mAcm-2
VOC= 1.07V
FF=68%
0.6
0.8
1.0
1.2
V(V)
V(V)
18.0
(C)
17.5
JSC(mAcm-2)
16.5
16.0
15.5
15.0
(D)
23.0
17.0
PCE(%)
(B)
14.5
22.5
22.0
21.5
21.0
14.0
20.5
13.5
A-TiO2
R-TiO2
A-TiO2
(E)
1.10
(F)
72
70
Fill Factor (%)
1.09
VOC(V)
R-TiO2
1.08
1.07
1.06
68
66
64
62
1.05
60
1.04
A-TiO2
R-TiO2
A-TiO2
100
R-TiO2
(G)
EQE(%)
80
60
A-TiO2
R-TiO2
40
20
0
300
400
500
600
700
800
Wavelength (nm)
F I G U R E 4 J–V curve of the best devices fabricated by (A) A-TiO2 , (B) R-TiO2 ETLs; (C)–(F) Statistics of the photovoltaic parameters; (G)
EQE spectra for devices based on A-TiO2 and R-TiO2
6
SHAHVARANFARD et al.
TA B L E 1
Summary of the photovoltaic parameters of devices based on A-TiO2 and R-TiO2 ETLs
ETL
construction
JSC [mA
cm−2 ]
VOC [V]
FF [%]
PCEavg [%]
PCEmax
[%]
A-TiO2
22.2 ± 0.4
1.08 ± 0.009
67 ± 2.2
16 ± 0.67
17.2
R-TiO2
22 ± 0.5
1.07 ± 0.01
66 ± 2.7
15.6 ± 0.66
16.7
analysis for A-TiO2 versus R-TiO2 . It should be noted
that besides morphology, grain size, and structure of the
deposited films the characteristics of the defects within
the films can also affect the performance of the deposited
layers, as is shown in our previous publications.[14]
However, the characterization of defect structures in the
deposited films falls out of the scope of the current work.
T A B L E 2 Work function values based on the Kelvin probe
measurements in dark
Work function [eV]
A-TiO2
4.27 ± 0.009
R-TiO2
4.22 ± 0.016
losses, is found for A-TiO2 , though the differences between
R-TiO2 and A-TiO2 are again small. It is expected that sample characterizations under light also result in the comparable response from the two samples. We cannot exclude
that such minor variations are originating from small variations of the properties of the sample such as slightly lower
roughness, higher wettability, or a work function shift.
Nevertheless, we note that loss analysis from impedance
spectroscopy agrees well with the trend found from J-V
3
CONCLUSION
The effect of crystallographic phase composition of TiO2
ETL on the performance of methylammonium lead iodide
(MAPI) perovskite solar cells was systematically explored.
Since TiO2 thin films play a key role as the ETL in
perovskite solar cells, we characterized these films with
100
100
80
60
Transmittance (T)
Reflectance (R)
Absorptance (A)
T+R+A
40
20
80
60
Transmittance (T)
Reflectance (R)
Absorptance (A)
T+R+A
40
20
0
0
400
FIGURE 5
(B)
Response (%)
Response (%)
(A)
500
600
Wavelength (nm)
700
800
400
500
600
700
Wavelength (nm)
800
UV–Vis transmittance (T), reflectance (R) and absorptance (A) spectra of 30 nm (A) A-TiO2 and (B) R-TiO2 on FTO
F I G U R E 6 A, Steady-state PL spectra of perovskite layer on FTO, FTO/A-TiO2 and FTO/R-TiO2 . B, Nyquist plots of full device measured
in dark at an applied voltage of 0.8 V. The inset shows the equivalent circuit used for fitting
SHAHVARANFARD et al.
respect to their physical features, optical properties, and
the ability to act as an efficient ETL in perovskite solar
cells. The film deposition parameters were tuned in order
to fabricate TiO2 layers that identical in their physical
and structural properties and differ only in the term
of their phase composition. Our results illustrated that
such anatase and rutile TiO2 that are deposited by magnetron sputtering performed relatively comparable when
processed identically. Thorough analyses explored that
anatase and rutile TiO2 layer showed only minor variations
in their morphological and optical features. Accordingly,
these minor variations in the film properties resulted only
in minor performance differences of perovskite solar cells.
Hence, our results indicate that for the proper comparison
between different phases of the TiO2 layer in an ETL, other
physical and structural properties of the fabricated films
should be identical, which can be achieved by controlling
the fabrication parameters during film deposition.
4
4.1
EXPERIMENTAL SECTION
Device fabrication
FTO coated glass substrates (7 Ω m−2 , Solaronix) were
ultrasonically cleaned in acetone, ethanol, and deionized
water and were then dried with nitrogen. TiO2 films were
deposited on FTO substrates with DC magnetron sputtering in an ultrahigh vacuum chamber (Createc, SP-P-US6 M-3Z) with pre-pump pressure of ∼10−8 mbar. The layers were deposited with two different chamber pressures,
2.5 × 10−3 and 2.5 × 10−2 mbar at 30 ◦ C. A pure Ti (99.90%)
was used as the target. For TiO2 films that were deposited
with different chamber pressure, the sputtering time was
tuned to achieve similar layer thicknesses (∼30 nm). The
sputtering rates for high and low sputtering pressure were
∼0.8 and ∼0.95 nm min-1 , respectively, and the DC sputtering was performed at 150 W operating power.
The TiO2 layers with a thickness of ∼30 nm were synthesized for fabrication of perovskite solar cells and their subsequent characterization. Samples were identified based
on the sputtering chamber pressure. TiO2 films that were
deposited with high sputtering pressure were labeled as
HP-TiO2 and TiO2 films that were deposited with low sputtering pressure were labeled as LP-TiO2 . The layers were
then annealed in an air furnace at 450 ◦ C for 1 hour.
To fabricate the perovskite solar cells, the as-prepared
FTO coated TiO2 layers were transferred to a nitrogenfilled glovebox. The perovskite solution was prepared
by dissolving 710 mg of PbI2 (Lumtec) and 240 mg of
CH3 NH3 I (Lumtec) in dimethylformamid (DMF) (SigmaAldrich) mixed with dimethylsulfoxid (DMSO) (Sigma-
7
Aldrich) at 40◦ C. The perovskite precursor was spincoated on the TiO2 film at 4000 rpm for 15 s followed by
blowing off extra solvents with a N2 stream (gas quenching process).
The samples were then dried on a hot plate at 110◦ C
for 10 min. As a hole transport layer, 10 mg mL−1
solution of poly[5,5′-bis(2-butyloctyl)-(2,2′-bithiophene)4,4′-dicarboxylate-alt-5,5′-2,2′-bithiophene](PDCBT,
1-Material) in chlorobenzene, was deposited by spin
coating on the perovskite layer at 2000 rpm for 30 s and
dried at 70◦ C for 3 minutes. Finally, a 15 nm-thick MoOx
(Alfa Aesar) and 120 nm-thick Ag counter electrode were
coated through a shadow mask (with an opening of 10.4
mm2 ) on PDCBT via thermal evaporation.
4.2
Characterization
The crystallographic composition of deposited TiO2
ETLs was assessed by X-ray diffraction (XRD, X’pert
Philips MPD diffractometer, equipped with a PanalyticalX’celeratordetector) and confocal Raman spectrometer
(LabRAM HR – Evolution, Horiba Scientific Ltd.),
whereas the morphology of the deposited TiO2 ETLs and
perovskite layers was investigated by a field-emission
scanning electron microscope (Fe-SEM, S4800, Hitachi).
The water contact angle was measured in a static mode by
OCA 20 (DataPhysics).
Current density-voltage (J-V) characterization of the
solar cells was performed under simulated AM1.5G with a
light intensity of 100 mW cm−2 which was calibrated with
a Si cell with the scan rate of 2.5 V s-1 (step: 0.05 V). The
external quantum efficiency (EQE) spectra were measured
using an Enli Technology (Taiwan) EQE system (QE-R).
The work function of the layers was measured by using
a Kelvin probe system in dark and the UV-vis light absorption spectra were obtained using a PerkinElmer spectroscopy system (Lambda 950 UV/VIS Spectrometer).
Steady-state photoluminescence (PL) was carried out
using an Argon-ion 488-nm laser as excitation source, a
Horiba monochromator and a Silicone detector. Solid state
impedance spectroscopy was performed in dark at 0.8 V
and in a frequency range from 1 MHz to 1 Hz by using a
Zahner system.
AC K N OW L E D G E M E N T S
S. Hosseinpour thanks W. Peukert and Emerging Talents Initiative (ETI) 2018/2_Tech_06, FAU, Germany, and
BMBF-MSRT (grant ID: CAlSAB) for supporting his
research.
Open access funding enabled and organized by Projekt
DEAL.
8
D A T A AVA I L A B I L I T Y S T A T E M E N T
The data that support the findings of this study are
available from the corresponding author upon reasonable
request
REFERENCES
1. N.-G. Park, K. Zhu, Nat. Rev. Mater. 2020, 5, 333.
2. A. E. Shalan, S. Narra, T. Oshikiri, K. Ueno, X. Shi, H.-P. Wu,
M. M. Elshanawany, E. Wei-Guang Diau, H. Misawa, Sustain.
Energy Fuels 2017, 1, 1533.
3. C. Chen, Y. Cheng, Q. Dai, H. Song, Sci. Rep. 2016, 5, 17684.
4. R. Wang, M. Mujahid, Y. Duan, Z. Wang, J. Xue, Y. Yang, Adv.
Funct. Mater. 2019, 29, 1808843.
5. B. Cao, H. Liu, L. Yang, X. Li, H. Liu, P. Dong, X. Mai, C. Hou, N.
Wang, J. Zhang, J. Fan, Q. Gao, Z. Guo, ACS Appl. Mater. Interfaces 2019, 11, 33770.
6. G. D. Rajmohan, F. Z. Huang, R. d’Agostino, J. du Plessis, X. J.
Dai, Thin Solid Films 2017, 636, 307.
7. A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc.
2009, 131, 6050.
8. W. S. Yang, B. W. Park, E. H. Jung, N. J. Jeon, Y. C. Kim, D. U.
Lee, S. S. Shin, J. Seo, E. K. Kim, J. H. Noh, S. Il Seok, Science
(80-.). 2017, 356, 1376.
9. W. Ke, G. Fang, Q. Liu, L. Xiong, P. Qin, H. Tao, J. Wang, H. Lei,
B. Li, J. Wan, G. Yang, Y. Yan, J. Am. Chem. Soc. 2015, 137, 6730.
10. L. Xiong, Y. Guo, J. Wen, H. Liu, G. Yang, P. Qin, G. Fang, Adv.
Funct. Mater. 2018, 28, 1802757.
11. D.-Y. Son, J.-H. Im, H.-S. Kim, N.-G. Park, J. Phys. Chem. C 2014,
118, 16567.
12. J. Kim, G. Kim, T. K. Kim, S. Kwon, H. Back, J. Lee, S. H. Lee, H.
Kang, K. Lee, J. Mater. Chem. A 2014, 2, 17291.
13. S. Zhang, L. Lei, S. Yang, X. Li, Y. Liu, Q. Gao, X. Gao, Q. Cao, Y.
Yu, Chem. Lett. 2016, 45, 592.
14. F. Shahvaranfard, M. Altomare, Y. Hou, S. Hejazi, W. Meng, B.
Osuagwu, N. Li, C. J. Brabec, P. Schmuki, Adv. Funct. Mater.
2020, 30, 1909738.
15. A. Huang, L. Lei, J. Zhu, Y. Yu, Y. Liu, S. Yang, S. Bao, X. Cao, P.
Jin, ACS Appl. Mater. Interfaces, 2017, 9, 2016.
16. C. Liang, Z. Wu, P. Li, J. Fan, Y. Zhang, G. Shao, Appl. Surf. Sci.
2017, 391, 337.
SHAHVARANFARD et al.
17. X. Yao, J. Liang, Y. Li, J. Luo, B. Shi, C. Wei, 2017, 1700008.
18. P. Roy, S. Berger, P. Schmuki, Angew. Chem. Int. Ed. 2011, 50,
2904.
19. S. J. Smith, R. Stevens, S. Liu, G. Li, A. Navrotsky, J. BoerioGoates, B. F. Woodfield, Am. Mineral. 2009, 94, 236.
20. L. Zhu, Q. Lu, L. Lv, Y. Wang, Y. Hu, Z. Deng, Z. Lou, Y. Hou, F.
Teng, RSC Adv. 2017, 7, 20084.
21. A. Yella, L.-P. Heiniger, P. Gao, M. K. Nazeeruddin, M. Grätzel,
Nano Lett. 2014, 14, 2591.
22. Y. Wang, J. Wan, J. Ding, J. S. Hu, D. Wang, Angew. Chemie - Int.
Ed. 2019, 58, 9414.
23. K. Thamaphat, P. Limsuwan, B. Ngotawornchai, Kasetsart J.
(Nat. Sci.) 2008, 42, 357.
24. K. Sakurai, M. Mizusawa, Anal. Chem. 2010, 82, 3519.
25. D. Tuschel, Spectroscopy 2019, 34, 10.
26. T.-C. MikhailPylnev, A. MariaBarbisan, Appl. Surf. Sci. 2021, 541.
27. X. Huang, Z. Hu, J. Xu, P. Wang, J. Zhang, Y. Zhu, Electrochim.
Acta 2017, 231, 77.
28. X. Zhang, Y. Lin, D. He, J. Zhang, Z. Fan, T. Xie, Chem. Phys. Lett.
2011, 504, 71.
29. R. Cohen, L. Kronik, A. Shanzer, D. Cahen, A. Liu, Y. Rosenwaks, J. K. Lorenz, A. B. Ellis, J. Am. Chem. Soc. 1999, 121, 10545.
30. W. Zeng, X. Liu, H. Wang, D. Cui, R. Xia, Y. Min, Thin Solid Films
2017, 629, 11.
31. H. Kim, I. Mora-Sero, V. Gonzalez-Pedro, F. Fabregat-Santiago,
E. J. Juarez-Perez, N. Park, J. Bisquert, Nat. Commun. 2013, 4,
2242.
S U P P O RT I N G I N F O R M AT I O N
Additional supporting information may be found in the
online version of the article at the publisher’s website.
How to cite this article: F. Shahvaranfard, N. Li,
S. Hosseinpour, S. Hejazi, K. Zhang, M. Altomare,
P. Schmuki, C. J. Brabec. Nano Select. 2021, 1.
https://doi.org/10.1002/nano.202100306
Supporting Information
Comparison of the sputtered TiO2 anatase and rutile thin films as electron transporting
layers in perovskite solar cells
Fahimeh Shahvaranfard,* Ning Li,* Saman Hosseinpour, Seyedsina Hejazi, Kaicheng Zhang,
Marco Altomare, Patrik Schmuki, Christoph J. Brabec*
F. Shahvaranfard, Dr. N. Li, K. Zhang, Prof. Dr. C. Brabec
Institute of Materials for Electronics and Energy Technology (i-MEET), Department of
Materials Science and Engineering, University of Erlangen-Nuremberg, Martensstrasse 7,
91058 Erlangen, Germany
E-mail: christoph.brabec@fau.de
Dr. N. Li, Prof. Dr. C. Brabec
Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (HI ERN), Immerwahrstr. 2,
91058, Erlangen, Germany
Dr. S. Hosseinpour
Institute of Particle Technology (LFG), Friedrich-Alexander-Universität-Erlangen-Nürnberg
(FAU), Cauerstraße 4, 91058 Erlangen, Germany
Prof. Dr. M. Altomare, Dr. S. Hejazi, Prof. Dr. P. Schmuki,
Institute for Surface Science and Corrosion WW4-LKO, Department of Materials Science and
Engineering, University of Erlangen-Nuremberg, Martensstrasse 7, 91058 Erlangen, Germany
1
* FTO
A TiO2 Anatase
R TiO2 Rutile
(a)
Intensity (a.u.)
R
*
*
R
*
*
R
*
*
*
A
20
1.0
30
R
* **
*
A*
Normalized Intensity (a.u.)
HP-TiO2
LP-TiO2
40
50
*
*
A
2q (o)
**
60
*
*
70
(b)
*
80
FTO
HP-TiO2
LP-TiO2
0.8
0.6
0.4
0.2
0.0
200
300
400
500
600
700
800
900 1000
Wavenumber (cm-1)
Figure S1. (a) X-ray diffraction (XRD) patterns of 200 nm HP-TiO2 and LP-TiO2 on FTO
before thermal treatment. (b) Normalized Raman spectra of FTO coated glass, HP and LP thin
films (~30 nm). The following parameters were chosen to acquire the Raman spectra with a
reasonable signal-to-noise ratio: Laser 532 nm, Objective 10X, Hole size 500, Grating 1800,
Slit size 100, Filter 100%, Acquisition time100 sec, and 100 times repetition.
2
Figure S2.(a)Top view SEM image and (b) AFM of pristine FTO.
Figure S3. The contact angle of water droplet on (a) A-TiO2, (b) R-TiO2 on FTO.
3
Figure S4. Top view SEM images and grain size distribution of MAPbI3 grown on (a and c)
A-TiO2, (b and d) R-TiO2.
Table S1. Impedance spectroscopy parameters of devices with different ETLs.
ETL construction
Rs (Ω)
Rrec(kΩ)
A-TiO2
19.9
R-TiO2
22.5
CPE
Q (nF)
a
5.6
21.8
0.95
4.6
23
0.95
4
Appendix C
Dewetting of PtCu Nanoalloys on TiO2 Nanocavities Provides a
Synergistic Photocatalytic Enhancement for Efficient H2 Evolution
Fahimeh Shahvaranfard, Paolo Ghigna, Alessandro Minguzzi, Ewa Wierzbicka,
Patrik Schmuki, Marco Altomare
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
97
www.acsami.org
Research Article
Dewetting of PtCu Nanoalloys on TiO2 Nanocavities Provides a
Synergistic Photocatalytic Enhancement for Efficient H2 Evolution
Fahimeh Shahvaranfard, Paolo Ghigna, Alessandro Minguzzi, Ewa Wierzbicka, Patrik Schmuki,*
and Marco Altomare*
Downloaded via FRIEDRICH-ALEXANDER-UNIV on August 3, 2021 at 10:33:56 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Cite This: ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACCESS
Metrics & More
Read Online
Article Recommendations
sı Supporting Information
*
ABSTRACT: We investigate the co-catalytic activity of PtCu alloy nanoparticles for photocatalytic H2 evolution from methanol−
water solutions. To produce the photocatalysts, a few-nanometer-thick Pt−Cu bilayers are deposited on anodic TiO2 nanocavity
arrays and converted by solid-state dewetting via a suitable thermal treatment into bimetallic PtCu nanoparticles. X-ray diffraction
(XRD) and X-ray photoelectron spectroscopy (XPS) results prove the formation of PtCu nanoalloys that carry a shell of surface
oxides. X-ray absorption near-edge structure (XANES) data support Pt and Cu alloying and indicate the presence of lattice disorder
in the PtCu nanoparticles. The PtCu co-catalyst on TiO2 shows a synergistic activity enhancement and a significantly higher activity
toward photocatalytic H2 evolution than Pt- or Cu-TiO2. We propose the enhanced activity to be due to Pt−Cu electronic
interactions, where Cu increases the electron density on Pt, favoring a more efficient electron transfer for H2 evolution. In addition,
Cu can further promote the photoactivity by providing additional surface catalytic sites for hydrogen recombination. Remarkably,
when increasing the methanol concentration up to 50 vol % in the reaction phase, we observe for PtCu-TiO2 a steeper activity
increase compared to Pt-TiO2. A further increase in methanol concentration (up to 80 vol %) causes for Pt-TiO2 a clear activity
decay, while PtCu-TiO2 still maintains a high level of activity. This suggests improved robustness of PtCu nanoalloys against
poisoning from methanol oxidation products such as CO.
KEYWORDS: TiO2 nanotube, solid-state dewetting, platinum, copper, PtCu alloy nanoparticle, photocatalysis, H2 evolution
1. INTRODUCTION
H2 evolution reaction are noble-metal nanoparticle (NPs), e.g.,
Pt, Au, or Pd. Pt, due to its higher work function (i.e., it forms
a higher metal/semiconductor Schottky barrier), can efficiently
trap photogenerated TiO2 CB electrons and mediate their
transfer to the environment. Pt offers also suitable catalytic
surface sites for hydrogen recombination. 9,10 Its low
abundance has however pushed the research toward the
identification of earth-abundant, cost-effective co-catalysts.
Metals such as Ni, Cu, and Co are more available, but their
individual activity is comparably poor. Nonetheless, their
combination with noble metals in the form of bimetallic (or
multimetallic) systems can lead to significantly higher activities
than those showed by the monometallic constituents.11−14 The
catalytic improvement, in general, is associated with the atomic
structure of the alloy catalyst and can be ascribed to electronic,
The generation of hydrogen from water on an illuminated
semiconductor has been pioneered by Fujishima and Honda in
1972.1 Ever since then, titanium dioxide (TiO2) has been
envisaged as a most promising photocatalytic material for H2
evolution due to some key features, i.e., a suitable conduction
band (CB) edge with respect to the redox potential of H2O to
H2, as well as its low cost, large availability, and high chemical
and photochemical stability.2−4
TiO2, as an oxide substoichiometric semiconductor, suffers
however from intrinsic limits such as a fast recombination of
photogenerated charge carriers, and a sluggish kinetic of charge
transfer to reactants at the semiconductor/environment
interface. This causes a poor photocatalytic performance.5,6
One-dimensional (1D) TiO2 nanostructures, such as nanotubes (NTs), nanorods, or nanowires, can provide a short
diffusion distance and a directional pathway for photogenerated charge carriers, hence limiting charge recombination
and improving the photoactivity.7,8
In addition to nanostructuring, another strategy to enhance
the photocatalytic performance is to deposit on the TiO2
surface a suitable co-catalyst. Co-catalysts that can promote the
© 2020 American Chemical Society
Received: June 16, 2020
Accepted: July 24, 2020
Published: July 24, 2020
38211
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
geometric, and synergistic effects.12 Bimetallic catalysts can
also block undesired reaction paths or prevent poisoning from
byproducts.15,16
In this context, PtCu alloys have attracted large attention as
they can provide a remarkable activity enhancement compared
to Pt alone in different catalytic processes such as oxygen
reduction,17 CO2 conversion,18 or NO reduction.19 Here,
alloying allows one also to minimize the amount of Pt used. In
addition, bimetallic PtCu nanoparticles or PtCu alloys
demonstrated superior catalytic activity and tolerance for the
electrochemical oxidation of light molecules, such as methanol
or formic acid.12,20−22
Combinations of noble metals (e.g., Pt or Au) and Cu
catalysts have been explored also in photocatalysis. TiO2
powders modified with grafted Cu species and Pt NPs showed
promising photocatalytic H2 evolution performances. This was
explained in terms of a reversible switch of the Cu oxidation
state (Cu2+ ↔ Cu+) that favors an efficient electron transfer via
Pt to protons for H2 evolution.23 Other studies on PtCuTiO224 or Au-Cu2O-TiO225 photocatalysts propose a reduction
of oxidized Cu species to Cu metal during the early stage of
illumination, leading to bimetallic (PtCu or AuCu) systems
where Cu promotes the photocatalytic performance by
increasing the electron density on Pt and by providing
catalytic surface sites for hydrogen recombination.
PtCu NPs can be produced by different methods including
photodeposition,26 galvanic replacement,12 impregnation,27
chemical reduction,28 or colloidal chemistry.29 In a recent
work,30 we reported on a straightforward “sputtering−
dewetting−alloying” approach to produce bimetallic (e.g.,
NiCu) nanoparticles directly on photocatalytic TiO2 surfaces.
Herein, we use such a dewetting−alloying concept to
produce a TiO2 photocatalyst for H2 evolution co-catalyzed by
bimetallic PtCu nanoparticles. For this, we first sequentially
coat TiO2 NT arrays with a few-nanometer-thick Pt and Cu
films. Then, we trigger solid state dewetting of the Pt−Cu
bilayers by a suitable heat treatment,31 that is, the Pt and Cu
films break up, agglomerate, and intermix, forming dewetted−
alloyed PtCu NPs at the NT surface. We vary the Pt and Cu
content in the NPs by adjusting the initial thicknesses of the Pt
and Cu films, i.e., by tuning the bilayer initial composition.
Our results show that PtCu NPs of an optimized
composition can lead to a remarkable synergistic enhancement
in photocatalytic hydrogen evolution compared to Pt or Cu
monometallic co-catalysts due to an improved charge transfer
efficiency to adsorbed H+. Moreover, our results indicate that
the presence of Cu in the bimetallic co-catalyst promotes an
enhanced performance even for high concentration of
methanol (used as hole scavenger) in the reaction phase,
while Pt-TiO2 under such reaction conditions is subject to a
substantial loss of activity. This suggests either that the
bimetallic PtCu co-catalyst promotes the complete oxidation of
methanol to CO2 (hence minimizing the formation of CO that
can poison Pt active sites) or that Cu improves the resistance
of the co-catalyst to poisoning, likely by weakening the binding
strength of methanol oxidation intermediates (CO) to Pt sites.
This results in a photocatalyst for photoreforming of organic−
water mixtures that allows us to generate H2 at a higher rate
than Pt-TiO2 systems and well tolerates high organic
concentrations in the reaction phase.
2. EXPERIMENTAL SECTION
Research Article
2.1. Growth of TiO2 Nanotube Arrays. Key to a high degree of
self-ordering for anodic TiO2 nanotube arrays is to establish
electrochemical conditions that during anodization lead to high rate
of oxide growth along with a high rate of oxide dissolution.32−34 This
can be achieved by anodizing Ti metal in hot o-H3PO4/hydrogen
fluoride (HF) electrolytes.34,35 The growth conditions can be adjusted
to form a short tube length that resembles a nanocavity. The resulting
arrays of TiO2 nanotubes can be formed over large surfaces, even up
to some 10 cm2,36 and present a virtually ideal hexagonal packing.
In the present work, to grow such nanotube arrays, Ti foils (30 ×
15 mm2, 0.125 mm thick, Advent Research Materials, +99.6% purity)
were cleaned by ultrasonication in acetone, ethanol, and deionized
(DI) water and then dried with nitrogen. Anodization was carried out
in an electrolyte consisting of 3 M HF in o-H3PO4 (Sigma-Aldrich) at
105 °C, by applying a potential of 15 V for 2 h. After anodization, the
samples were rinsed with ethanol, soaked in ethanol overnight, and
then dried with nitrogen.
2.2. Metal-Film Deposition and Solid-State Dewetting. To
form Pt, Cu, and alloyed PtCu nanoparticles on TiO2 nanotube
arrays, an Ar-plasma sputtering machine (EM SCD500, Leica) was
used to deposit thin metal films, followed by a solid-state dewetting
step.
Metal (Pt or Cu) films or metal bilayers (Pt−Cu) of different
thicknesses were deposited on the anodic TiO2 nanotube structures.
As metal targets, we used a 99.99% pure Pt target (Hauner
Metallische Werkstoffe) or a 99.90% pure Cu target (Baltic
Praeparation e.K.). Sputtering was performed at an applied current
of 16 mA and at a chamber pressure of 10−2 mbar of Ar.
A solid-state dewetting31 approach was then used to convert thin
metal films into dispersions of defined metal particles at the
nanostructured oxide surface. The overall driving force for dewetting
is the minimization of the free surface energy of the metal film, of the
substrate, and of the metal−substrate interface. Given that the thinner
the metal film, the higher its surface-to-volume ratio and thus its
surface energy, the driving force for dewetting increases dramatically
when the film thickness decreases. In other words, the thinner the
metal film, the lower the activation energy for metal atom surface
mobility. This is the key reason why dewetting can be triggered at
temperatures that are well below the film melting point, i.e., the film
dewets while remaining in the solid state.31,36
In the present work, after metal-film sputtering, the samples were
annealed at 500 °C, for 1 h in an argon atmosphere (Ar flux = 10 L
h−1), to induce solid-state dewetting of the metal films and form
directly at the NT surface Pt, Cu, or bimetallic PtCu nanoparticles.
Samples are named according to the co-catalyst nature and
composition, depending on the initial thickness of the dewetted metal
film(s): for example, “PtxCuy-TiO2”, where “x” and “y” are the
nominal initial thicknesses of the sputtered metal films expressed in
nanometer. TiO2 samples loaded with monometallic (Pt or Cu) NPs
are labeled as “Ptx-TiO2” or “Cuy-TiO2”.
2.3. Characterization of the Structures. A field emission
scanning electron microscope (FE-SEM, Hitachi S4800) was used to
study the morphology of the samples. The crystallographic properties
of the samples were examined by X-ray diffraction (XRD) with an
X’Pert Philips MPD (equipped with a Panalytical X’celerator
detector). X-ray photoelectron spectroscopy (XPS, PHI 5600) was
employed to analyze the chemical composition of the samples.
Diffusive reflectance spectra of the different samples were measured
by a fiber-based UV−vis−IR spectrophotometer (Avantes, ULS2048)
equipped with an integrating sphere AvaSphere-30 using an AvaLightDH-S-BAL balanced power light source.
X-ray absorption spectroscopy (XAS) at the Pt LIII-edge (11 564
eV) and Cu K edge (8979 eV) was carried out in the fluorescence
mode at the P65 beamline of the DESY - Petra III synchrotron
radiation facility (Hamburg, Germany), using a Si(311) double-crystal
monochromator and a passivated implanted planar silicon (PIPS)
detector, allowing for collecting a full extended X-ray absorption fine
structure (EXAFS) spectrum in ca. 15−20 min. The energy
38212
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
calibration was performed by measuring the absorption spectrum of a
metallic Cu foil at the Cu K edge. The spectra of the reference
samples (metallic Cu and Pt, CuO and Cu2O) were acquired in the
transmission mode. For these measurements, a proper amount of
sample, as to give a unit increase in the absorption coefficient, was
mixed with cellulose and pressed into a pellet. All data were obtained
at room temperature. The X-ray signal extraction and normalization
were performed by means of the ATHENA code, belonging to the set
of interactive programs IFEFFIT.37 The pre-edge background was
fitted by means of a straight line, and the post-edge background by
means of a cubic spline. The EXAFS data analysis was performed with
the EXCURVE code using a k2 weighing scheme.
2.4. Photocatalytic Experiments. Photocatalytic measurements
for H2 generation were conducted by irradiating the metal NPdecorated oxide films in water, methanol, methanol−water (with
different MeOH contents), or ethanol−water (20% EtOH) mixtures
in a quartz tube sealed with a rubber septum. As irradiation source, a
UV LED (Opsytec, Germany λ = 365 nm, beam size = 0.785 cm2,
power of ∼100 mW cm−2) was used. H2 evolution experiments were
carried out for the most active sample (Pt2.5Cu2.5) also under visible
light irradiation provided by a filtered (420 nm cutoff filter) AM 1.5G
simulated solar light (100 mW cm−2, 300 W Xe lamp with solar light
optical filter) and by a monochromatic 450 nm laser (2 W, OdicForce
Lasers).
Prior to photocatalysis, to remove oxygen, the cell and liquid phase
were purged with Ar gas for 20 min. The amount of accumulated H2
in the head space of the quartz tube during irradiation was determined
by means of gas chromatography (GC-MSQO2010SE, Shimadzu,
Japan). The GC column was equipped with a thermal conductivity
detector (TCD), a Restek micro packed Shin Carbon ST column (2
m × 0.53 mm), and a Zebron capillary column ZB05 MS (30 m ×
0.25 mm). GC measurements were conducted at oven temperature,
45 °C (isothermal conditions), with the temperature of the injector
set at 280 °C and that of the TCD fixed at 260 °C. The carrier gas
(argon) flow rate was 14.3 mL min−1.
Research Article
Figure 1. SEM images of different TiO2 nanotube arrays: (a) pristine
NTs (the inset shows the cross-sectional view); (b−f) NTs decorated
with NPs dewetted from (b) 5 nm thick Pt film; (c) PtCu bilayer with
4 nm Pt−1 nm Cu; (d) PtCu bilayer with 2.5 nm Pt−2.5 nm Cu (the
inset shows the cross-sectional view); (e) PtCu bilayer with 1 nm Pt−
4 nm Cu; and (f) 5 nm thick Cu films.
The dewetted nanoparticles show a similar average size
(Figure S3). The Pt film and the different Pt−Cu bilayers
dewet into particles with a mean size in the 5−15 nm range.
Minor differences are observed for the pure Cu NPs, the size of
which is in average slightly larger, i.e., 5−20 nm. A similar
trend was observed also in previous work30 and can be
explained taking into account the lower melting point, and
likely a higher surface mobility, of Cu.
A series of TiO2 NT arrays decorated with dewetted Pt, Cu,
or PtCu NPs (prepared by systematically varying the nominal
thickness of the Pt and Cu films) were characterized by XRD
in view of their crystallographic features.
The XRD results are shown in Figure 2. All patterns (Figure
2a) feature dominant diffraction signals at 38.4 and 40.1°
ascribed to the 002 and 101 reflections of metallic Ti. Such
signals originate from the Ti metal substrate. Anodic TiO2 NT
layers, as reported in the literature, are typically amorphous in
the as-grown state.2 The thermal treatment at 500 °C in argon
leads to the crystallization of the NT layers into a mixed
anatase−rutile TiO2 phase, as proved by the peaks at 25.3 and
27.5° that can be ascribed to the anatase 101 and rutile 110
reflections, respectively. The crystallization of the NTs forms
also minor amounts of a Ti oxyfluoride phase, giving rise to the
peak at 22.8° that can be attributed to the 200 reflection of
TiOF2. Such a compound may originate from a solid-state
reaction (during the thermal treatment) between the oxide and
fluoride species uptaken from the anodizing electrolyte during
the NT growth. Besides, the content of anatase phase is in
general higher than that of rutile for most samples, though a
different anatase−rutile relative composition was observed for
NTs decorated with pure Cu NPs (sample Cu5). In this case,
the presence of Cu alone seems to favor the crystallization of
3. RESULTS AND DISCUSSION
Figure 1a shows the SEM images of highly ordered TiO2
nanotubes (“nanocavities”) grown by anodization of Ti foils in
a hot HF/o-H3PO4 electrolyte.38 Additional SEM images are
provided in Figure S1. These nanostructures show an ideal
hexagonal packing (top view) and have an average length of
∼200 nm and inner diameter of ∼100 nm (cross-sectional
viewinset in Figure 1a). Such semiconductor geometry is
ideal for the subsequent metal film deposition−dewetting step.
Figure 1b−f shows the result of metal sputter-coating of Pt
(5 nm) and Cu (5 nm) films or Pt−Cu bilayers on TiO2 NTs
followed by dewetting at 500 °C in Ar. For the bilayers, we
sputtered first Pt and then Cu. As shown in Figure S2, the
TiO2 NTs were homogeneously covered by the sputtered
metal (nominal film thickness, 5 nm), regardless of the film
composition (Pt, Cu, or Pt−Cu). One can see from Figure
1b−f that the thermal treatment causes the conversion via
surface diffusion of all metal films and bilayers into metal
nanoparticles.
It is reported that dewetting generally initiates at temperatures (Tdewet) between 1/3 and 1/2 of the metal melting
point.31 The occurrence of dewetting of Cu at 500 °C is well in
line with its melting points (Tm = 1085 °C).37 Dewetting of Pt
films (and Pt−Cu bilayers) would in principle be expected to
take place at higher temperatures, given the melting point of Pt
of 1768 °C. Nevertheless, the films studied in the present work
are rather thin (nominal thickness ≤5 nm) and can hence
dewet at lower temperatures as, in general, the thinner the film
the lower the Tdewet value.31
38213
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
Research Article
Figure 2. (a) X-ray diffraction (XRD) patterns of TiO2 nanotube
decorated with dewetted Pt, Cu, or alloyed PtCu NPs of different
compositions; (b) enlarged view of the XRD patterns in (a) in the 2θ
range of 44−52°.
the NTs into rutile phase (as observed also in previous
work30,37).
Figure 2b shows an enlarged view in the 2θ region of 45−
52° of the patterns shown in Figure 2a. The XRD pattern of
sample Pt5 shows signals at 39.7, 46.4, and 67.8° that
correspond to the Pt 111, 200, and 220 diffraction peaks,
confirming the metallic nature of the dewetted Pt NPs. The
XRD pattern of sample Cu5 shows signals at 43.3, 50.5, and
74.0° that can be attributed to the 111, 200, and 220
crystallographic planes of Cu metal, respectively.11,39
Interestingly, samples produced by dewetting Pt−Cu
bilayers show a shift of the signals of Pt toward larger angles
and of Cu toward smaller angles, i.e., for sample Pt2.5Cu2.5,
no Pt 200 peak is found at 46.4° and no Cu 200 reflection
appears at 50.5°. Instead, only one signal can be seen at 48.1°
that can be assigned to the 200 reflection of a bimetallic PtCu
phase. This confirms the occurrence of Pt and Cu alloying
during dewetting (in line with what is also reported for other
dewetted metal combinations37,40), i.e., the bilayer, when
treated at 500 °C in Ar, breaks up and agglomerates causing Pt
and Cu to intermix forming dewetted−alloyed PtCu NPs at
the TiO2 NT surface. The position of the 200 peak is well in
line with Vegard’s law,30,41 that is, the lattice constant of the
PtCu phase correlates with a linear combination of the 200
peak positions of pure Pt and Cu phases and fits to the Pt and
Cu molar fractions in the binary PtCu system. In addition, the
XRD peaks of the dewetted NPs are broad and relatively weak,
which suggest a low degree of crystallinity or the presence of
structural disorder in the bimetallic NPs (see the discussion of
XAS data below).
To further study the physicochemical differences between
mono- and bimetallic co-catalysts, samples Pt2.5, Pt2.5Cu2.5,
and Cu2.5 were characterized by XPS and XAS.
XPS surveys for these samples are shown in Figure S4, while
their surface composition data are compiled in Table S1. In
general, the data show the structures to be composed of Ti, O,
adventitious C, and Pt and/or Cu in line with the composition
of the sputtered-dewetted metal films.
High-resolution XPS spectra are shown in Figure 3a,b, while
Figure 3c shows the Cu and Pt speciation determined by fitting
the HR XPS spectra. Figure 3a provides the Pt 4f XPS spectra
for the samples Pt2.5 and Pt2.5Cu2.5. The fit of the spectrum
of sample Pt2.5 shows signals at 70.7 and 74.0 eV that can be
attributed to the Pt 4f7/2 and Pt 4f5/2 signals of metallic Pt. The
small peaks at 71.9 and 75.4 eV can account for minor contents
Figure 3. XPS spectra for the (a) Pt 4f region of samples Pt2.5 and
Pt2.5Cu2.5 and (b) Cu 2p region of samples Cu2.5 and Pt2.5Cu2.5;
(c) Pt and Cu speciation determined by fitting the spectra in (a) and
(b).
(∼10 atom %) of Pt(II) species, likely Pt(II) oxides (see
quantitative data in Figure 3c).42 Similar spectra were obtained
for the sample decorated with the bimetallic co-catalyst, i.e.,
Pt2.5Cu2.5. Here, however, one can note a positive shift of 0.3
eV in the binding energy (BE) values of the Pt 4f7/2 and Pt
38214
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
4f5/2 peaks. This proves an intimate interaction of Pt and Cu in
the dewetted bimetallic nanoparticles and hence corroborates
the XRD results supporting the formation of PtCu nanoalloys.26,42,43 Besides, small peaks at 72.2 and 75.7 eV are also
observed for sample Pt2.5Cu2.5, which can account for
comparable contents of Pt(II) oxides. It should be mentioned
that the presence of Pt(II) oxides (as well as Pt(IV) species)
on dewetted Pt nanoparticles has been reported also in
previous work.44
Figure 3b shows the Cu 2p spectra for samples Cu2.5 and
Pt2.5Cu2.5. The Cu 2p spectrum of sample Cu2.5 clearly
shows the coexistence of different oxidation states of copper,
such as metallic Cu with the characteristic Cu 2p3/2 and Cu
2p1/2 doublet located at 932.4 and 952.2 eV, and Cu(II) (likely
CuO), the signal of which can be fitted with two peaks at 934.4
(Cu 2p3/2) and 954.4 eV (Cu 2p1/2). The Cu(II) species also
show the characteristic satellite features at ca. 940 and 962
eV,45 and hence one can conclude that Cu(I) species are either
not present at the surface of the NPs or that their content is
minor. The Cu 2p spectrum of sample Pt2.5Cu2.5 shows a
shift of 0.3 eV to lower binding energy (i.e., a negative BE
shift) with respect to that of pure Cu NPs (sample Cu2.5)
this provides further evidence that Cu and Pt are in intimate
contact in the dewetted−alloyed PtCu NPs.
Based on such results, one may suggest that the dewetted
NPs feature a metallic core carrying an oxide shell. The latter
forms likely after dewetting, when the samples are exposed to
oxygen under ambient conditions. This explains, for example,
the particularly high O surface content measured by XPS for
sample Cu2.5 (see data in Figure 3c).
We characterized by XPS also samples Pt5-TiO2 and Cu5TiO2. The results are shown in Figure S5a,b, along with the
samples’ composition in terms of relative metal vs oxide
content obtained from XPS fitting (Figure S5c). These data
should be compared to results for Pt2.5-, Cu2.5-, and
Pt2.5Cu2.5 systems in Figure 3. A comparison of the
quantitative data for Pt, Cu, and PtCu systems shows that
the relative metal vs oxide content in the dewetted nanoparticles does not change significantly with the particle size.
Interestingly, the Cu 2p XPS spectrum of sample Pt2.5Cu2.5
(Figure 3b) shows that the co-catalyst NPs contain Cu mainly
in the metallic state, in line with the dominant Cu 2p3/2 and
Cu 2p1/2 doublet located at 932.1 and 951.9 eV, while the
signal of Cu(II) species is negligible (see for comparison the
quantitative data in Figure 3c). This supports once more the
interaction in the bimetallic NPs between Cu and Pt, where in
this case the presence of Pt seems to limit Cu oxidation under
ambient conditions. It should be noted that the Gibbs free
energy of dissolution of metallic Cu in Pt is negative at any
mixing ratio.46 The formation of a Pt−Cu alloy therefore gives
a stabilizing contribution term for Cu(0)this may be the
origin of the protection effect against Cu oxidation observed
for the PtCu NPs.
In principle, based on plain electronegativity effects, the XPS
shift occurring with alloying is expected for Pt to be negative
(shift toward lower BE values), while it should be positive for
Cu (shift toward higher BE values). Our results, though
apparently in contradiction with the expected shifts, are
nonetheless in line with what reported in various previous
studies on PtCu nanoalloys.26,42,47,48 This suggests that not
only the alloy composition but likely also electronic or
structural effects, or the presence of additional compounds
(oxides, as in our case), may affect the Pt and Cu BE values
and the resulting shifts observed.
We found that, for bimetallic systems, the XPS signals can
show anomalies in the binding energy shift.26,42,47−49 Such
anomalies have been explained, e.g., by a charge compensation
model, as reported by Watson et al.49 In the case of Pt−Cu
systems, positive (“anomalous”) shifts for the Pt signal have
been observed,26,42,48 and particularly relevant are the findings
reported for Pt−Co alloys.47 It has been proposed that: when
Pt atoms are alloyed with a second component, the total
number of electrons per Pt atom increases, while the number
of 5d electrons decreases, accompanied by a rehybridization of
the d-band as well as of the sp-band, which leads to an upshift
of the reference level (Fermi energy), resulting in an opposite
downshift of the Pt 4f7/2 level. Such electronic effects were
shown to cause, as a net effect, a positive BE shift for Pt when
alloyed with metals such as Co or Cu.
X-ray absorption near-edge structure (XANES) spectra for
samples Pt5, Pt2.5Cu2.5, and Cu5, as well as reference
materials (Pt and Cu foils, and Cu2O and CuO pellets) are
shown in Figure 4a,b. At the Pt L3-edge, the edge energy
Research Article
Figure 4. XAS results at the Pt LIII and Cu K edges. (a, b) XANES
spectra at the LIII and Cu K edges, respectively. The spectra were
shifted along the y axis for a clearer comparison. (c, d) Pt LIII-edge
EXAFS spectra for samples Pt2.5Cu2.5 and Pt5, respectively. Blue and
magenta lines: experimental; dark yellow lines: fit according to the
model described in the text. (e, f) Fourier transforms of the EXAFS
spectra of (c) and (d), respectively.
position and shape show that Pt is present in the metallic state.
The structure at ca. 11 580 eV is broadened in sample Pt5 with
respect to the Pt foil reference, and even more broadening is
apparent in the spectrum of sample Pt2.5Cu2.5. In general,
broadening of features in XAS spectra is indicative of the
presence of disorder, for example, Lytle et al.50 attributed the
38215
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
broader structure of the Pt L3 edge in Pt catalysts (compared
to a reference metal foil) to nanosize effects in the former. We
can therefore conclude that Pt in the dewetted NPs is more
disordered than the metallic Pt reference (foil), and the
disorder is in sample Pt2.5Cu2.5 even more pronounced with
respect to sample Pt5 due to alloying effects. The increased
disorder is in agreement with the diffraction patterns shown in
Figure 2 (peak broadening). The EXAFS spectra at the Pt L3
edge of the two samples were fitted using a Pt metal model.
The fits are shown in Figure 4c−f. The fitting results are given
in Table S2 (n: coordination number, r: coordination distance,
r0: crystallographic distance in metallic Pt, σ2: EXAFS Debye−
Waller factor). The nearest-neighbor distance in sample
Pt2.5Cu2.5 is lower than in sample Pt5. This likely indicates
that in sample Pt2.5Cu2.5, Cu is incorporated in the Pt lattice
(atomic radii: Pt 1.75 Å, Cu 1.40 Å).
The XANES data at the Cu K edge are shown in Figure 4b,
along with the spectra of standard compounds for the different
oxidation states of Cu. The spectrum of sample Pt2.5Cu2.5
shows the presence of mostly Cu(II) with some amounts of
Cu(I), while the spectrum of sample Cu5 shows the presence
of both Cu(I) and Cu(II) in comparable amounts. It should be
noted, however, that a precise quantification of the amounts of
Cu in the different oxidation states is made difficult by the lack
of an evident edge structure. Using the same argument as
above, this can be attributed to the presence of disorder in the
dewetted NPs, in agreement with their nanostructured nature
as evidenced by XRD. This disorder prevents further analysis
of these spectra. However, XANES results at the Cu K edge
agree well with XPS data, when the sensitivity of XPS to nearsurface information is taken into account.
The different samples were then studied as photocatalysts
for H2 evolution from methanol−water mixtures under UV
light illumination. Methanol was selected as a hole scavenger
(e.g., instead of ethanol) as in preliminary experiments the
former was found to lead to higher H2 evolution rates (Figure
S6), as also reported in previous work. More details can be
found in the Supporting Information.
Figure 5a shows the photocatalytic activity of pristine TiO2
NTs along with that of NT layers decorated with dewetted Pt,
Cu, or PtCu NPs. The data indicate that, as expected, pristine
TiO2 NTs show a negligible H2 generation rate (0.06 μL h−1
cm−2), while structures decorated with Pt NPs (sample Pt5)
evolve H2 at a higher rate, i.e., 12.9 μL h−1 cm−2. The activity
of NTs modified with dewetted Cu NPs is 1.0 μL h−1 cm−2,
i.e., clearly lower than that of Pt-TiO2 systems, this being in
line with previous work.30 As widely reported in the
literature,51−53 the presence of Pt NPs on the TiO2 surface
is beneficial to the photoactivity due to the high work function
of Pt (5.1−5.9 eV) with respect to the Fermi level of TiO2
(4.6−4.7 eV). This results in the formation of a relatively high
Schottky barrier that allows for efficient trapping of TiO2 CB
electrons in the Pt NPs and for their transfer to H+ for H2
evolution. The lower work function of Cu (4.5−5.1 eV)
explains the comparably poor activity of Cu-TiO2 systems, i.e.,
the height of the formed Schottky barrier is comparably small,
leading to a less efficient electron trapping.
Most importantly, dewetted bimetallic NPs with equal
nominal loadings (initial film thickness) of Pt and Cu (sample
Pt2.5Cu2.5) co-catalyze TiO2 NTs for H2 generation with the
highest efficiency, leading to H2 evolution rates as high as 50.7
μL h−1 cm−2, which is ∼4 and ∼50 times higher than the
activity of NTs modified with pure Pt (sample Pt5) and pure
Research Article
Figure 5. (a) Photocatalytic H2 evolution rate under UV light
illumination measured for different Pt/Cu ratios; (b) photocatalytic
H2 evolution rate under UV light illumination measured for samples
dewetted from films of different nominal thicknesses but constant Pt/
Cu ratio of 1:1; (c) photocatalytic H2 evolution rate under UV light
illumination measured for samples dewetted from films of different
nominal thicknesses of Pt; (d) evolved H2 amount over time for
samples Pt2.5Cu2.5-TiO2 and Pt2.5-TiO2 under UV light illumination; (e) stability test for sample Pt2.5Cu2.5-TiO2; (f) photocatalytic
H2 evolution rate for samples Pt-TiO2 and PtCu-TiO2 measured in
methanol−water mixtures of different compositions under UV light
illumination; (g) difference in H2 evolution rate (ΔrH2) between
samples Pt-TiO2 and PtCu-TiO2 plotted as a function of the
methanol−water mixture composition.
Cu (Cu5) NPs, respectively. The nominal relative Pt/Cu
content for sample Pt2.5Cu2.5 is 0.70:0.30 wt % or 0.44:0.56
atom %. Similar activity enhancements for Pt and Cu comodified photocatalysts are reported also in previous reports,
but smaller relative amounts of Cu (≤0.1 wt %), i.e., higher Pt
relative contents, were shown to enable a clear performance
enhancement.23
The activity enhancement we observe is remarkable and is
far larger than simple additive effects, i.e., it is synergistic as the
sum of the activity of pure Pt and Cu co-catalyst NPs yields a
H2 evolution rate of ∼14.0 μL h−1 cm−2, which is substantially
38216
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
lower than the rate of 50.7 μL h−1 cm−2 delivered by the
bimetallic co-catalyst.
The co-catalytic activity of other bimetallic systems, e.g.,
sample Pt4Cu1 (13.6 μL h−1 cm−2) or Pt1Cu4 (19.5 μL h−1
cm−2), is only slightly higher than that of sample Pt5.
Nevertheless, it is remarkable that for Cu NPs the
incorporation of small amounts of Pt, as in sample Pt1Cu4,
leads to activities that are higher than that of sample Pt5, which
indeed contains a (nominally) 5 times larger Pt loading.
To evaluate the effects of the overall co-catalyst loading on
the H2 evolution activity, a series of samples were produced by
dewetting on NTs bilayers with different nominal thicknesses
but a constant Pt/Cu thickness ratio of 1:1. These structures
are shown in Figure S7. The SEM data show that in any case
the NTs are homogeneously decorated with bimetallic PtCu
NPS. It is clear that the thicker the bilayer, the larger the PtCu
NP size, which shows average diameters of 5−10, 5−15, and
25−35 nm for samples Pt1Cu1, Pt2.5Cu2.5, and Pt5Cu5,
respectively. The photocatalytic performance of these
structures is shown in Figure 5b. The results show that
dewetting of bilayers with a nominal thickness of Pt2.5 nm and
Cu2.5 nm leads to the most effective co-catalytic effects.
After evaluating the effects of the relative Pt/Cu amount, we
carried out control experiments by producing a series of TiO2
NT layers decorated with different loadings of dewetted Pt
NPs. For this, the initial Pt loading was systematically screened
by sputter-depositing 0.5, 1, 2.5, 5, and 10 nm thick Pt films
followed by dewetting (500 °C, Ar, 1 h). Figure 5c shows the
photocatalytic H2 evolution rates of Pt-TiO2 structures. The
data show a typical volcano trend, where the increase of the cocatalyst loading leads initially to an activity enhancement.
Sample Pt2.5 is the most active photocatalyst, with a H2
evolution rate of 25.8 μL h−1 cm−2, which is still 50% lower
than the activity of the optimized bimetallic co-catalyst
(sample Pt2.5Cu2.5) in spite of the virtually identical Pt
content. Higher Pt loadings lead to suboptimum photocatalytic
efficiencies (as also observed in previous work8,44,54) likely due
to a larger size of the dewetted NPs and consequently to their
lower active surface area.
The higher activities reported in previous studies for Pt- and
Cu-modified photocatalysts were attributed to Cu plasmonic
effects27,55 or Cu2+/Cu+ photosensitization mechanisms.56 To
enable the latter effect in Pt−Cu bimetallic systems, the
formation of PtCu alloys was purposely avoided, being
reported to be detrimental for the photoactivity.23 Irie et al.,
and Dozzi and co-workers in follow-up work, proposed that
since the redox potential of the Cu2+/Cu+ couple (for
amorphous Cu(II) oxide decorations) and the CB minima of
crystalline CuO are less negative than the CB of TiO2, Cu2+
species on the TiO2 surface may capture TiO2 CB electrons
causing a consequent one-electron reduction of oxidized
copper species (Cu2+ to Cu+), meanwhile enhancing the
charge carrier separation efficiency.23,56 This effect was
reported to come along with visible light activation as electrons
in the VB of TiO2 are directly transferred to the discrete energy
levels of grafted Cu2+ species (hence causing visible light
absorption).
For what concerns our work, annealing titania in argon (or
in a reducing environment, e.g., H2/Ar) may produce oxygen
vacancies in the oxide, as extensively reported in the literature
(examples can be found in refs57−61). Structures crystallized
by air annealing would provide a suitable reference material to
assess the effect of the Ar treatment and the associated
formation of oxygen vacancies. Nevertheless, air annealing is
not suitable to induce dewetting (and alloying) of Pt, Cu, or
Pt−Cu films. As found in previous work,13,44,54,61,62 such
metals are susceptible to oxygen and tend to easily oxide,
particularly when exposed to oxygen at high temperatures
their oxidation, in turn, prevents metal atom surface mobility
and impair dewetting and NP formation. In other words, airtreated structures would differ from samples annealed in argon
not only for their electronic properties (e.g., density of oxygen
vacancies) but also in the nature (morphology and
composition) of the co-catalyst, thus making the direct
comparison meaningless.
We should however point out that, for the argon-treated
samples, possible effects associated with oxygen vacancies
formation should be comparable for any samples, since all
samples were treated under the same conditions (500 °C, pure
argon, 1 h). Such effects may be, e.g., a higher mobility for
majority carriers (increased conductivity for CB electrons),58,63
or visible light activation, e.g., due to the introduction of sub
CB energy levels causing a narrowing of the oxide optical band
gap.58,59
We assessed the optical properties of pristine NTs and Cu-,
Pt-, and PtCu-modified structures by measuring UV−vis
diffuse reflectance spectra. The results are shown Figure S8.
According to the spectra, the band gap (Eg) absorption, for λ ∼
≤400 nm, is clear for pristine NTs (black curve). The onset of
light absorption is in line with the mixed anatase/rutile
composition of the TiO2 NTs.
The absorption band(s) in the visible spectral range, as
observed in recent work,64 can be assigned to interference
fringes due to the NT barrier layer (NT bottom) at the
substrate/NT interface. The presence of dewetted NPs on the
NTs introduces additional complex optical features (likely due
to their reflectance). Note that in the presence of the dewetted
metal NPs, not only different optical characteristics appear in
the visible spectral range but also the TiO2 band gap
absorption (with predicted onset at λ ∼ 400 nm) becomes
indistinguishable.
We carried out H2 evolution experiments with the most
active sample (Pt2.5Cu2.5) under irradiation provided by a
filtered (420 nm cutoff) AM 1.5G simulated solar light and by
a monochromatic 450 nm laser (2 W, OdicForce Lasers).
Under such experimental conditions, no significant amount of
evolved hydrogen could be detected by gas chromatography.
This suggests that the PtCu-modified photocatalysts are active
under UV light, and hence the improved activity provided by
dewetted−alloyed PtCu NPs cannot be ascribed to the NP
light absorption features, e.g., to plasmonic effects or to the
formation of p−n junctions (in the presence of Cu oxides) or
via Cu(II) photosensitization effects. Analyzing the UV−vis
spectra (Figure S8), we noted in fact that the reflectance of
samples 2.5Pt2.5Cu and 5Cu is at 365 nm virtually identical,
while the activity of the former is ∼50 times higher than that of
the latter.
Therefore, one can conclude that:
Research Article
(i) the optical (reflectance) features of the different samples
seem not to match the photocatalytic activity trend;
(ii) the oxide charge transport properties cannot be invoked
as a key factor affecting the activity, since such features
are expected to be comparable for all of the structures
(due to the identical Ar annealing conditions used);
38217
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
Research Article
the cycles were repeated four times in a row, after degassing
the photocatalytic cell and refreshing the methanol−water
mixture. The results (Figure 5e) show the H2 evolution activity
to be stable and reproducible for an overall illumination period
of 16 h, hence confirming that no photocatalyst deterioration
phenomena (NPs detachment or Cu loss) take place.
Finally, we evaluated the effect of the composition of the
methanol−water mixtures on the H2 evolution activity. Figure
5f shows the activity of sample Pt2.5Cu2.5 and its
monometallic counterpart (Pt2.5) as a function of the
methanol content. Such data are also compiled in Figure 5g
to show the activity difference (ΔrH2) between samples
Pt2.5Cu2.5 and Pt2.5 for each mixture composition.
According to the results, by increasing the concentration of
methanol from 0 to 50 vol % in the reaction solution, the
bimetallic catalyst, i.e., sample Pt2.5Cu2.5, shows a more
pronounced increase in photoactivity compared to sample
Pt2.5. When the methanol concentration is further increased to
80%, sample Pt2.5 undergoes an evident deactivation, losing
50% of its activity (Figure 5f). On the contrary, the bimetallic
co-catalyst is affected to a minor extent (∼15% activity loss)
and, as a result, the activity difference (ΔrH2) between samples
Pt2.5Cu2.5 and Pt2.5 monotonically increases when increasing
the methanol concentration from 0 to 80 vol % (Figure 5g).
These results may provide an additional explanation for the
enhanced activity of bimetallic NPs. One may in fact speculate
that PtCu systems provide, compared to Pt NPs, a substantially
enhanced tolerance against poisoning from methanol oxidation
products. Cu atoms in the bimetallic NPs can in fact weaken
the binding energy of intermediate adsorbates (such as CO) to
Pt surface sites, as also reported for PtCu catalysts used in
methanol fuel cells.11,71 It is also worth mentioning that
previous work on Cu- and PtCu-modified TiO2 photocatalysts23 showed that the presence of Cu is key to promote
the VB hole-mediated oxidation of the hole scavenger to
carbon dioxide (complete oxidation) rather than to carbon
monoxide (partial oxidation). This was shown to be
accompanied by a higher rate of hydrogen production. In
other words, the higher selectivity to CO2 may not only limit
the formation of CO, hence preventing CO from blocking H2
evolution on Pt sites, but can also make TiO2 CB electrons
more readily available for H+ reduction due to a more efficient
VB hole consumption.72
(iii) the key factor affecting the H2 evolution performance is
thus the co-catalyst composition.
The presence of Pt, as clearly shown by XPS and XANES,
affects the oxidation state of Cu, which in the bimetallic NPs is
present mainly in the metallic form. This may provide Cu in
the PtCu system, compared to oxidized Cu species in the pure
Cu NPs, with a metal-like co-catalytic ability, i.e., with the
ability to trap TiO2 CB electrons via a metal/semiconductor
Schottky junction effect. Pt, on the other hand, is present in
the dewetted NPs mainly as metal, and its metallic nature
remains almost unaltered regardless of the presence or absence
of Cu as an alloying element. It is also worth to point out that
in recent work65 on Ni-, Cu-, and NiCu-modified TiO2 we
demonstrate that, under illumination, oxidized Ni and Cu
species such as native oxides at the co-catalyst surface are
promptly reduced to a metallic phase. This can occur either via
solid-state reduction or through a dissolution/redeposition
processboth mechanisms are triggered by photopromoted
CB electrons. In situ reduction under photocatalytic conditions
was also shown for oxidized Pt species in Pt-based photocatalytic systems66,67 and proposed for CuOx−PtO2−TiO2
photocatalysts.24 Thus, we assume that under the experimental
conditions here adopted, oxidized Cu (or Pt) compounds are
reduced to their metallic states, which are the active cocatalytic species responsible for the photocatalytic H 2
generation observed. In other words, the different initial Cu
speciation in Cu- and PtCu-modified photocatalysts should not
be invoked to explain the higher photoactivity of the latter.
Hence, as also proposed by Amal et al.,24 a most plausible
cause for the synergistically enhanced co-catalytic performance
is the electronic interaction between Pt and Cu atoms in
bimetallic NPs. Such interaction, as supported in their work by
XPS results and density functional theory (DFT) calculations,
is based on the lower electronegativity of Cu with respect to
that of Pt. The result is an uneven distribution of the electron
density between Pt and Cu that can increase the net electron
density on Pt, enhancing the trapping efficiency of photogenerated electrons from the TiO2 CB. This decreases charge
recombination in the semiconductor and consequently
improves the photocatalytic H 2 evolution (Scheme
S1).12,24,68 Similar beneficial effects when alloying Pt with
Cu were reported also recently for the electrochemical
hydrogen evolution reaction (HER).69
We assessed the PtCu-TiO2 photocatalysts also in terms of
stability. Figure 5d shows the amounts of evolved hydrogen vs
illumination time for sample Pt2.5Cu2.5. Data for sample Pt2.5
are also shown as reference. Sample Pt2.5Cu2.5 under
illumination produces H2 at a constant rate, and no initial
induction period or deactivation effects at later stages causing
loss of activity could be observed. The steady H2 evolution in
the early stage also indicates that the possible in situ reduction
of oxidized Cu species occurs rapidly, i.e., in a timespan of a
few minutes, according to what we recently observed by in situ
time resolved X-ray absorption spectroscopy.65 Given the
steady H2 evolution, one may even speculate that reduction of
oxidized Cu species occurs in the solid state, i.e., via charge
transfer across the PtCu/TiO2 interface and hence does not
involve in this case Cu dissolution/redeposition mechanisms
reported elsewhere.65,70
To further validate the photocatalyst stability, we carried out
with the same sample (Pt2.5Cu2.5) a series of repeated
photocatalytic cycles. Each photocatalytic run lasted 4 h, and
4. CONCLUSIONS
We used a dewetting−alloying principle to produce, from
nanometer thick PtCu bilayers, bimetallic PtCu co-catalytic
nanoparticles on arrays of TiO2 nanocavities. We characterized
these structures in view of their physicochemical features and
co-catalytic ability in photocatalytic H2 evolution. A series of
complementary characterization techniques including XRD,
XPS, and XANES proved the bimetallic nature of the PtCu
alloy nanoparticles. TiO2 nanocavities co-catalyzed by such
PtCu nanoalloys can synergistically promote the photocatalytic
generation of H2 from methanol−water mixtures at rates that
are substantially higher than those yielded by Pt or Cu
monometallic counterparts. The H2 evolution improvement is
ascribed to a Pt−Cu electronic interaction in the co-catalyst
NPs that causes a more efficient trapping of TiO2 CB electrons
and transfer to the environment for H2 evolution.
38218
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
The results herein discussed contribute another interpretation to the growing debate on the copper oxidation state and
its role in the photocatalytic hydrogen evolution reaction.
From a more general viewpoint, our work shows that
alloying via solid-state dewetting of earth-abundant transition
metals such as Cu with noble elements (Pt, Pd, or Au) can
provide intriguing catalytic effects and remarkable activity
enhancements along with a more sustainable catalyst economy.
Such a principle can be adapted to a variety of bimetallic
systems of key relevance in photocatalytic or electrochemical
conversion processes.
Alexander University Erlangen-Nuremberg, Germany. M.A.,
P.G., and A.M. acknowledge DESY (Hamburg, Germany), a
member of the Helmholtz Association HGF, for the provision
of experimental facilities. Parts of this research were carried out
at DESYPETRA III (Project I-20190283), and Dr. Edmund
Welter is acknowledged for technical assistance in using the
photon beamline P65. The research leading to this result has
been supported also by the project CALIPSOplus under the
Grant Agreement 730872 from the EU Framework Programme
for Research and Innovation HORIZON 2020. E.W. acknowledges the Alexander von Humboldt Foundation for providing
financial support. A.M. acknowledges ″Piano di Sostegno alla
Ricerca, Università degli Studi di Milano″. P.G. acknowledges
financial support by MIUR through the grant “PRIN 2017,
2017KKP5ZR, MOSCATo”. Helga Hildebrand, Anja Friedrich, Ulrike Marten-Jahns, and Alexander Tesler are gratefully
acknowledged for technical help.
■
ASSOCIATED CONTENT
* Supporting Information
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acsami.0c10968.
Additional catalyst characterization data (SEM, particle
size statistics, XPS, photocatalytic H2 evolution activity,
UV−vis, EXAFS) (Figures S1−S8, Scheme S1, and
Tables S1 and S2) (PDF)
■
■
Research Article
REFERENCES
(1) Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at
a Semiconductor Electrode. Nature 1972, 238, 37−38.
(2) Roy, P.; Berger, S.; Schmuki, P. TiO2 Nanotubes: Synthesis and
Applications. Angew. Chem., Int. Ed 2011, 50, 2904−2939.
(3) Spanu, D.; Recchia, S.; Mohajernia, S.; Schmuki, P.; Altomare,
M. Site-Selective Pt Dewetting on WO3 -Coated TiO2 Nanotube
Arrays: An Electron Transfer Cascade-Based H 2 Evolution
Photocatalyst. Appl. Catal., B 2018, 237, 198−205.
(4) Chuangchote, S.; Jitputti, J.; Sagawa, T.; Yoshikawa, S.
Photocatalytic Activity for Hydrogen Evolution of Electrospun
TiO2 Nanofibers. ACS Appl. Mater. Interfaces 2009, 1, 1140−1143.
(5) Linsebigler, A. L.; Lu, G.; Yates, J. T. Photocatalysis on TiOn
Surfaces: Principles, Mechanisms, and Selected Results. Chem. Rev.
1995, 95, 735−758.
(6) Chen, X.; Mao, S. S. Titanium Dioxide Nanomaterials: Synthesis,
Properties, Modifications, and Applications. Chem. Rev. 2007, 107,
2891−2959.
(7) Jennings, J. R.; Ghicov, A.; Peter, L. M.; Schmuki, P.; Walker, A.
B. Dye-Sensitized Solar Cells Based on Oriented TiO2 Nanotube
Arrays: Transport, Trapping, and Transfer of Electrons. J. Am. Chem.
Soc. 2008, 130, 13364−13372.
(8) Nguyen, N. T.; Altomare, M.; Yoo, J.; Schmuki, P. Efficient
Photocatalytic H2 Evolution: Controlled Dewetting-Dealloying to
Fabricate Site-Selective High-Activity Nanoporous Au Particles on
Highly Ordered TiO2 Nanotube Arrays. Adv. Mater. 2015, 27, 3208−
3215.
(9) Wang, D.; Liu, Z. P.; Yang, W. M. Proton-Promoted Electron
Transfer in Photocatalysis: Key Step for Photocatalytic Hydrogen
Evolution on Metal/Titania Composites. ACS Catal. 2017, 7, 2744−
2752.
(10) Chen, T.; Feng, Z.; Wu, G.; Shi, J.; Ma, G.; Ying, P.; Li, C.
Mechanistic Studies of Photocatalytic Reaction of Methanol for
Hydrogen Production on Pt/TiO2 by in Situ Fourier Transform IR
and Time-Resolved IR Spectroscopy. J. Phys. Chem. C 2007, 111,
8005−8014.
(11) Liao, Y.; Yu, G.; Zhang, Y.; Guo, T.; Chang, F.; Zhong, C.
Composition-Tunable PtCu Alloy Nanowires and Electrocatalytic
Synergy for Methanol Oxidation Reaction. J. Phys. Chem. C 2016, 120,
10476−10484.
(12) Dimitrova, N.; Dhifallah, M.; Mineva, T.; Boiadjieva-Scherzer,
T.; Guesmi, H.; Georgieva, J. High Performance of PtCu@TiO2
Nanocatalysts toward Methanol Oxidation Reaction: From Synthesis
to Molecular Picture Insight. RSC Adv. 2019, 9, 2073−2080.
(13) Ji, L.; Spanu, D.; Denisov, N.; Recchia, S.; Schmuki, P.;
Altomare, M. A Dewetted-dealloyed Nanoporous Pt Co-catalyst
Formed on TiO2 Nanotube Arrays Leads to Strongly Enhanced
Photocatalytic H2 Production. Chem. - Asian J. 2020, 15, 301−309.
AUTHOR INFORMATION
Corresponding Authors
Patrik Schmuki − Institute for Surface Science and Corrosion
WW4-LKO, Department of Materials Science and Engineering,
University of Erlangen-Nuremberg, 91058 Erlangen, Germany;
Chemistry Department, Faculty of Sciences, King Abdulaziz
University, 80203 Jeddah, Kingdom of Saudi Arabia;
orcid.org/0000-0002-9208-5771; Email: schmuki@
ww.uni-erlangen.de
Marco Altomare − Institute for Surface Science and Corrosion
WW4-LKO, Department of Materials Science and Engineering,
University of Erlangen-Nuremberg, 91058 Erlangen, Germany;
orcid.org/0000-0002-7237-8809;
Email: marco.altomare@fau.de
Authors
Fahimeh Shahvaranfard − Institute for Surface Science and
Corrosion WW4-LKO, Department of Materials Science and
Engineering, University of Erlangen-Nuremberg, 91058
Erlangen, Germany
Paolo Ghigna − Dipartimento di Chimica, Università degli Studi
di Pavia, 27100 Pavia, Italy; orcid.org/0000-0002-86807272
Alessandro Minguzzi − Dipartimento di Chimica, Università
degli Studi di Milano, 20133 Milan, Italy; orcid.org/00000002-8130-4465
Ewa Wierzbicka − Institute for Surface Science and Corrosion
WW4-LKO, Department of Materials Science and Engineering,
University of Erlangen-Nuremberg, 91058 Erlangen, Germany;
orcid.org/0000-0001-8064-4823
Complete contact information is available at:
https://pubs.acs.org/10.1021/acsami.0c10968
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
The authors acknowledge the ERC, DFG, and the DFG cluster
of excellence EAM for financial support. M.A. acknowledges
financial support from the Emerging Talents Initiative ETI
(ETI2018/2_Tech_11) provided by the FAU Friedrich
38219
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
(14) Kugai, J.; Moriya, T.; Seino, S.; Nakagawa, T.; Ohkubo, Y.;
Nitani, H.; Mizukoshi, Y.; Yamamoto, T. A. Effect of Support for
PtCu Bimetallic Catalysts Synthesized by Electron Beam Irradiation
Method on Preferential CO Oxidation. Appl. Catal., B 2012, 126,
306−314.
(15) Okuhara, T.; Miura, H. Metallic Catalysts. Hyomen Kagaku
1988, 9, 676−683.
(16) Papadimitriou, S.; Armyanov, S.; Valova, E.; Hubin, A.;
Steenhaut, O.; Pavlidou, E.; Kokkinidis, G.; Sotiropoulos, S. Methanol
Oxidation at Pt-Cu, Pt-Ni, and Pt-Co Electrode Coatings Prepared by
a Galvanic Replacement Process. J. Phys. Chem. C 2010, 114, 5217−
5223.
(17) Lyu, L.-M.; Kao, Y.-C.; Cullen, D. A.; Sneed, B. T.; Chuang, Y.C.; Kuo, C.-H. Spiny Rhombic Dodecahedral CuPt Nanoframes with
Enhanced Catalytic Performance Synthesized from Cu Nanocube
Templates. Chem. Mater. 2017, 29, 5681−5692.
(18) Lee, S.; Jeong, S.; Kim, W. D.; Lee, S.; Lee, K.; Bae, W. K.;
Moon, J. H.; Lee, S.; Lee, D. C. Low-Coordinated Surface Atoms of
CuPt Alloy Cocatalysts on TiO2 for Enhanced Photocatalytic
Conversion of CO2. Nanoscale 2016, 8, 10043−10048.
(19) Zhou, S.; Varughese, B.; Eichhorn, B.; Jackson, G.; McIlwrath,
K. Pt-Cu Core-Shell and Alloy Nanoparticles for Heterogeneous NOx
Reduction: Anomalous Stability and Reactivity of a Core-Shell
Nanostructure. Angew. Chem., Int. Ed. 2005, 44, 4539−4543.
(20) Xiang, H.; Zheng, Y.; Sun, Y.; Guo, T.; Zhang, P.; Li, W.; Kong,
S.; Ouzounian, M.; Chen, H.; Li, H.; Hu, T. S.; Yu, G.; Feng, Y.; Liu,
S. Bimetallic and Postsynthetically Alloyed PtCu Nanostructures with
Tunable Reactivity for the Methanol Oxidation Reaction. Nanoscale
Adv. 2020, 2, 1603−1612.
(21) Yang, H.; Dai, L.; Xu, D.; Fang, J.; Zou, S. Electrooxidation of
Methanol and Formic Acid on PtCu Nanoparticles. Electrochim. Acta
2010, 55, 8000−8004.
(22) Khan, I. A.; Qian, Y.; Badshah, A.; Zhao, D.; Nadeem, M. A.
Fabrication of Highly Stable and Efficient PtCu Alloy Nanoparticles
on Highly Porous Carbon for Direct Methanol Fuel Cells. ACS Appl.
Mater. Interfaces 2016, 8, 20793−20801.
(23) Dozzi, M. V.; Chiarello, G. L.; Pedroni, M.; Livraghi, S.;
Giamello, E.; Selli, E. High Photocatalytic Hydrogen Production on
Cu(II) Pre-Grafted Pt/TiO2. Appl. Catal., B 2017, 209, 417−428.
(24) Jung, M.; Hart, J. N.; Boensch, D.; Scott, J.; Ng, Y. H.; Amal, R.
Hydrogen Evolution via Glycerol Photoreforming over Cu−Pt
Nanoalloys on TiO2. Appl. Catal., A 2016, 518, 221−230.
(25) Sinatra, L.; LaGrow, A. P.; Peng, W.; Kirmani, A. R.; Amassian,
A.; Idriss, H.; Bakr, O. M. A Au/Cu2O−TiO2 System for PhotoCatalytic Hydrogen Production. A Pn-Junction Effect or a Simple
Case of in Situ Reduction? J. Catal. 2015, 322, 109−117.
(26) Kar, P.; Zhang, Y.; Mahdi, N.; Thakur, U. K.; Wiltshire, B. D.;
Kisslinger, R.; Shankar, K. Heterojunctions of Mixed Phase TiO2
Nanotubes with Cu, CuPt, and Pt Nanoparticles: Interfacial Band
Alignment and Visible Light Photoelectrochemical Activity. Nanotechnology 2018, 29, No. 14002.
(27) Shiraishi, Y.; Sakamoto, H.; Sugano, Y.; Ichikawa, S.; Hirai, T.
Pt-Cu Bimetallic Alloy Nanoparticles Supported on Anatase TiO2:
Highly Active Catalysts for Aerobic Oxidation Driven by Visible
Light. ACS Nano 2013, 7, 9287−9297.
(28) Srabionyan, V. V.; Pryadchenko, V. V.; Kurzin, A. A.; Belenov,
S. V.; Avakyan, L. A.; Guterman, V. E.; Bugaev, L. A. Atomic Structure
of PtCu Nanoparticles in PtCu/C Catalysts from EXAFS Spectroscopy Data. Phys. Solid State 2016, 58, 752−762.
(29) Toshima, N.; Wang, Y. Preparation and Catalysis of Novel
Colloidal Dispersions of Copper/Noble Metal Bimetallic Clusters.
Langmuir 1994, 10, 4574−4580.
(30) Spanu, D.; Recchia, S.; Mohajernia, S.; Tomanec, O.; Kment,
Š .; Zboril, R.; Schmuki, P.; Altomare, M. Templated Dewetting−
Alloying of NiCu Bilayers on TiO2 Nanotubes Enables Efficient
Noble-Metal-Free Photocatalytic H2 Evolution. ACS Catal. 2018, 8,
5298−5305.
(31) Thompson, C. V. Solid-State Dewetting of Thin Films. Annu.
Rev. Mater. Res. 2012, 42, 399−434.
(32) Ono, S.; Saito, M.; Asoh, H. Self-Ordering of Anodic Porous
Alumina Formed in Organic Acid Electrolytes. Electrochim. Acta 2005,
51, 827−833.
(33) Houser, J. E.; Hebert, K. R. The Role of Viscous Flow of Oxide
in the Growth of Self-Ordered Porous Anodic Alumina Films. Nat.
Mater. 2009, 8, 415−420.
(34) Hebert, K. R.; Albu, S. P.; Paramasivam, I.; Schmuki, P.
Morphological Instability Leading to Formation of Porous Anodic
Oxide Films. Nat. Mater. 2012, 11, 162−166.
(35) Yoo, J. E.; Schmuki, P. Critical Factors in the Anodic
Formation of Extremely Ordered Titania Nanocavities. J. Electrochem.
Soc. 2019, 166, C3389−C3398.
(36) Altomare, M.; Nguyen, N. T.; Schmuki, P. Templated
Dewetting: Designing Entirely Self-Organized Platforms for Photocatalysis. Chem. Sci. 2016, 7, 6865−6886.
(37) Ravel, B.; Newville, M. ATHENA and ARTEMIS Interactive
Graphical Data Analysisusing IFEFFIT. Phys. Scr. 2005, T115,
No. 1007.
(38) Yoo, J. E.; Lee, K.; Altomare, M.; Selli, E.; Schmuki, P. SelfOrganized Arrays of Single-Metal Catalyst Particles in TiO2 Cavities:
A Highly Efficient Photocatalytic System. Angew. Chem., Int. Ed. 2013,
52, 7514−7517.
(39) El-Deeb, H.; Bron, M. Microwave-Assisted Polyol Synthesis of
PtCu/carbon Nanotube Catalysts for Electrocatalytic Oxygen
Reduction. J. Power Sources 2015, 275, 893−900.
(40) Herz, A.; Wang, D.; Müller, R.; Schaaf, P. Formation of
Supersaturated Au-Ni Nanoparticles via Dewetting of an Au/Ni
Bilayer. Mater. Lett. 2013, 102−103, 22−25.
(41) Denton, A. R.; Ashcroft, N. W. Vegard’s Law. Phys. Rev. A
1991, 43, 3161−3164.
(42) Liu, Y.; Huang, Y.; Xie, Y.; Yang, Z.; Huang, H.; Zhou, Q.
Preparation of Highly Dispersed CuPt Nanoparticles on Ionic-LiquidAssisted Graphene Sheets for Direct Methanol Fuel Cell. Chem. Eng. J.
2012, 197, 80−87.
(43) Liu, J.; Liu, M.; Yang, X.; Chen, H.; Liu, S. F.; Yan, J. PhotoRedeposition Synthesis of Bimetal Pt−Cu Co-Catalysts for TiO2
Photocatalytic Solar-Fuel Production. ACS Sustainable Chem. Eng.
2020, 8, 6055−6064.
(44) Yoo, J.; Altomare, M.; Mokhtar, M.; Alshehri, A.; Al-Thabaiti,
S. A.; Mazare, A.; Schmuki, P. Photocatalytic H2 Generation Using
Dewetted Pt-Decorated TiO2 Nanotubes: Optimized Dewetting and
Oxide Crystallization by a Multiple Annealing Process. J. Phys. Chem.
C 2016, 120, 15884−15892.
(45) Mohajernia, S.; Hejazi, S.; Andryskova, P.; Zoppellaro, G.;
Tomanec, O.; Zboril, R.; Schmuki, P. Conductive Cu-Doped TiO2
Nanotubes for Enhanced Photoelectrochemical Methanol Oxidation
and Concomitant Hydrogen Generation. ChemElectroChem 2019, 6,
1244−1249.
(46) Myles, K.; Darby, J. Thermodynamic Properties of Solid
Palladuim-Copper and Platinum-Copper Alloys. Acta Metall. 1968,
16, 485−492.
(47) Wakisaka, M.; Mitsui, S.; Hirose, Y.; Kawashima, K.; Uchida,
H.; Watanabe, M. Electronic Structures of Pt-Co and Pt-Ru Alloys for
CO-Tolerant Anode Catalysts in Polymer Electrolyte Fuel Cells
Studied by EC-XPS. J. Phys. Chem. B 2006, 110, 23489−23496.
(48) Liu, J.; Liu, M.; Yang, X.; Chen, H.; Liu, S. F.; Yan, J. PhotoRedeposition Synthesis of Bimetal Pt−Cu Co-Catalysts for TiO2
Photocatalytic Solar-Fuel Production. ACS Sustainable Chem. Eng.
2020, 8, 6055−6064.
(49) Watson, R. E.; Hudis, J.; Perlman, M. L. Charge Flow and d
Compensation in Gold Alloys. Phys. Rev. B 1971, 4 (12), 4139.
(50) Lytle, F. W.; Wei, P. S. P.; Greegor, R. B.; Via, G. H.; Sinfelt, J.
H. Effect of Chemical Environment on Magnitude of X-ray
Absorption Resonance at L I I I Edges. Studies on Metallic Elements,
Compounds, and Catalysts. J. Chem. Phys. 1979, 70, 4849−4855.
(51) Bamwenda, G. R.; Tsubota, S.; Kobayashi, T.; Haruta, M.
Photoinduced Hydrogen Production from an Aqueous Solution of
Ethylene Glycol over Ultrafine Gold Supported on TiO2. J.
Photochem. Photobiol., A 1994, 77, 59−67.
38220
Research Article
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
ACS Applied Materials & Interfaces
www.acsami.org
(52) Bamwenda, G. R.; Tsubota, S.; Nakamura, T.; Haruta, M.
Photoassisted Hydrogen Production from a Water-Ethanol Solution-a
Comparison of Activities of Au@TiO2 and Pr@TiO2. J. Photochem.
Photobiol., A 1995, 89, 177−189.
(53) Naldoni, A.; D’Arienzo, M.; Altomare, M.; Marelli, M.; Scotti,
R.; Morazzoni, F.; Selli, E.; Dal Santo, V. Pt and Au/TiO2
Photocatalysts for Methanol Reforming: Role of Metal Nanoparticles
in Tuning Charge Trapping Properties and Photoefficiency. Appl.
Catal., B 2013, 130−131, 239−248.
(54) Nguyen, N. T.; Yoo, J.; Altomare, M.; Schmuki, P. “Suspended”
Pt Nanoparticles over TiO 2 Nanotubes for Enhanced Photocatalytic
H 2 Evolution. Chem. Commun. 2014, 50, 9653−9656.
(55) Zhang, L.; Jia, C.; He, S.; Zhu, Y.; Wang, Y.; Zhao, Z.; Gao, X.;
Zhang, X.; Sang, Y.; Zhang, D.; Xu, X.; Liu, H. Hot Hole Enhanced
Synergistic Catalytic Oxidation on Pt-Cu Alloy Clusters. Adv. Sci.
2017, 4, No. 1600448.
(56) Irie, H.; Kamiya, K.; Shibanuma, T.; Miura, S.; Tryk, D. A.;
Yokoyama, T.; Hashimoto, K. Visible Light-Sensitive Cu(II)-Grafted
TiO2 Photocatalysts: Activities and X-Ray Absorption Fine Structure
Analyses. J. Phys. Chem. C 2009, 113, 10761−10766.
(57) Mohajernia, S.; Hejazi, S.; Mazare, A.; Nguyen, N. T.; Schmuki,
P. Photoelectrochemical H2 Generation from Suboxide TiO2
Nanotubes: Visible-Light Absorption versus Conductivity. Chem. Eur. J. 2017, 23, 12406−12411.
(58) Naldoni, A.; Altomare, M.; Zoppellaro, G.; Liu, N.; Kment, Š .;
Zbořil, R.; Schmuki, P. Photocatalysis with Reduced TiO2: From
Black TiO2 to Cocatalyst-Free Hydrogen Production. ACS Catal.
2019, 9, 345−364.
(59) Mohajernia, S.; Andryskova, P.; Zoppellaro, G.; Hejazi, S.;
Kment, S.; Zboril, R.; Schmidt, J.; Schmuki, P. Influence of Ti 3+
Defect-Type on Heterogeneous Photocatalytic H2 Evolution Activity
of TiO2. J. Mater. Chem. A 2020, 8, 1432−1442.
(60) Liu, N.; Häublein, V.; Zhou, X.; Venkatesan, U.; Hartmann, M.;
Mačković, M.; Nakajima, T.; Spiecker, E.; Osvet, A.; Frey, L.;
Schmuki, P. “Black” TiO2 Nanotubes Formed by High-Energy Proton
Implantation Show Noble-Metal-Co-Catalyst Free Photocatalytic H2Evolution. Nano Lett. 2015, 15, 6815−6820.
(61) Naldoni, A.; Fabbri, F.; Altomare, M.; Marelli, M.; Psaro, R.;
Selli, E.; Salviati, G.; Dal Santo, V. The Critical Role of Intragap States
in the Energy Transfer from Gold Nanoparticles to TiO2. Phys. Chem.
Chem. Phys. 2015, 17, 4864−4869.
(62) Nguyen, N. T.; Altomare, M.; Yoo, J. E.; Taccardi, N.; Schmuki,
P. Noble Metals on Anodic TiO2 Nanotube Mouths: Thermal
Dewetting of Minimal Pt Co-Catalyst Loading Leads to Significantly
Enhanced Photocatalytic H2 Generation. Adv. Energy Mater. 2016, 6,
No. 1501926.
(63) Mohajernia, S.; Hejazi, S.; Mazare, A.; Nguyen, N. T.; Hwang,
I.; Kment, S.; Zoppellaro, G.; Tomanec, O.; Zboril, R.; Schmuki, P.
Semimetallic Core-Shell TiO2 Nanotubes as a High Conductivity
Scaffold and Use in Efficient 3D-RuO2 Supercapacitors. Mater. Today
Energy 2017, 6, 46−52.
(64) Tesler, A. B.; Altomare, M.; Schmuki, P. Morphology and
Optical Properties of Highly Ordered TiO2 Nanotubes Grown in
NH4F/O-H3PO4 Electrolytes in View of Light Harvesting and
Catalytic Applications ACS Appl. Nano Mater. 2020, DOI: 10.1021/
acsanm.0c01859.
(65) Spanu, D.; Minguzzi, A.; Recchia, S.; Shahvardanfard, F.;
Tomanec, O.; Zbořil, R.; Schmuki, P.; Ghigna, P.; Altomare, M. An
Operando X-Ray Absorption Spectroscopy Study of a NiCu-TiO2
Photocatalyst for H2 Evolution. ACS Catal. 2020, 10, 8293−8302.
(66) Chiarello, G. L.; Dozzi, M. V.; Scavini, M.; Grunwaldt, J. D.;
Selli, E. One Step Flame-Made Fluorinated Pt/TiO2 Photocatalysts
for Hydrogen Production. Appl. Catal., B 2014, 160−161, 144−151.
(67) Khare, R.; Jentys, A.; Lercher, J. A. Development of
photochemical and electrochemical cells for operando X-ray
absorption spectroscopy during photocatalytic and electrocatalytic
reactions. Phys. Chem. Chem. Phys. 2020, DOI: 10.1039/
D0CP00654H.
(68) Venkateswara Rao, C.; Viswanathan, B. ORR Activity and
Direct Ethanol Fuel Cell Performance of Carbon-Supported Pt−M
(M = Fe, Co, and Cr) Alloys Prepared by Polyol Reduction Method.
J. Phys. Chem. C 2009, 113, 18907−18913.
(69) Du, Y.; Ni, K.; Zhai, Q.; Yun, Y.; Xu, Y.; Sheng, H.; Zhu, Y.;
Zhu, M. Facile Air Oxidative Induced Dealloying of Hierarchical
Branched PtCu Nanodendrites with Enhanced Activity for Hydrogen
Evolution. Appl. Catal., A 2018, 557, 72−78.
(70) Hejazi, S.; Mohajernia, S.; Wu, Y.; Andryskova, P.; Zoppellaro,
G.; Hwang, I.; Tomanec, O.; Zboril, R.; Schmuki, P. Intrinsic Cu
Nanoparticle Decoration of TiO2 Nanotubes: A Platform for Efficient
Noble Metal Free Photocatalytic H2 Production. Electrochem.
Commun. 2019, 98, 82−86.
(71) Lu, L.; Chen, S.; Thota, S.; Wang, X.; Wang, Y.; Zou, S.; Fan, J.;
Zhao, J. Composition Controllable Synthesis of PtCu Nanodendrites
with Efficient Electrocatalytic Activity for Methanol Oxidation
Induced by High Index Surface and Electronic Interaction. J. Phys.
Chem. C 2017, 121, 19796−19806.
(72) Chiarello, G. L.; Aguirre, M. H.; Selli, E. Hydrogen Production
by Photocatalytic Steam Reforming of Methanol on Noble MetalModified TiO2. J. Catal. 2010, 273, 182−190.
38221
Research Article
https://dx.doi.org/10.1021/acsami.0c10968
ACS Appl. Mater. Interfaces 2020, 12, 38211−38221
Supporting Information
Dewetting of PtCu Nanoalloys on TiO2 Nanocavities Provides a
Synergistic Photocatalytic Enhancement for Efficient H2 Evolution
Fahimeh Shahvaranfard,a Paolo Ghigna,b Alessandro Minguzzi,c
Ewa Wierzbicka,a Patrik Schmuki,a,d* Marco Altomarea*
a
Institute for Surface Science and Corrosion WW4-LKO, Department of Materials Science and
Engineering, University of Erlangen-Nuremberg, Martensstrasse 7, 91058 Erlangen, Germany
b
Dipartimento di Chimica, Università degli Studi di Pavia, Viale Taramelli 13, 27100 Pavia,
Italy
c
Dipartimento di Chimica, Università degli Studi di Milano, Via Golgi 19, 20133 Milan, Italy
d
Chemistry Department, Faculty of Sciences, King Abdulaziz University, 80203 Jeddah,
Kingdom of Saudi Arabia
* Corresponding Author. E-mail:
schmuki@ww.uni-erlangen.de
marco.altomare@fau.de
Figure S1 SEM images of pristine TiO2 NT structures taken for one sample at different locations.
The SEM images in Fig. S1 show the morphology of the TiO2 nanocavities used in the present
work, taken for one sample at different locations. While the tube arrays show in general a high
degree of self-ordering (a), some structural inhomogeneities (e.g. gaps between tubes, deviations
from ideal hexagonal packing or variation in the nanotube diameter (b-d)) can be observed at
specific points – likely at (or close to) grain boundaries of the parent polycrystalline Ti metal
substrate.1–3
S2
Figure S2 SEM images of different TiO2 NT structures: (a) pristine; (b-d) sputter-coated with
different metal films: (b) 5 nm Pt; (c) 2.5 nm Pt and 2.5 nm Cu; (d) 5 nm Cu.
S3
Figure S3 Particle size distribution for samples (a) Pt5, (b) Pt4Cu1, (c) Pt2.5Cu2.5, (d) Pt1Cu4
and (e) Cu5.
S4
Figure S4 XPS surveys for samples Pt2.5, Pt2.5Cu2.5 and Cu2.5.
Figure S5 XPS spectra for (a) Pt 4f region of samples Pt5, (b) Cu 2p region of samples Cu5, (c)
Pt and Cu speciation determined by fitting the spectra in (a,b).
S5
Figure S6 Photocatalytic H2 evolution rate under UV light illumination measured for sample
Pt2.5Cu2.5 in different alcohol-water mixtures.
The H2 evolution activity of the PtCu-TiO2 system was measured not only in 20% MeOH-water
mixtures but also in a 20% EtOH aqueous solution for comparison.
We observed a substantially higher activity (~ 2 times higher) in methanol-water mixtures than in
the presence of ethanol, despite using the same concentration for the different hole scavengers.
These results are well in line with the findings of Al-Azri et al.5 The effect of different sacrificial
agents on the photocatalytic H2 evolution rate has been also studied in other works – examples are
refs.6–9 In general, it was found that for a given photocatalyst (e.g. Pt-TiO2), the ability of the
scavenger in capturing valence band holes depends primarily on factors such as the scavenger
redox potential and its polarity. The more negative its redox potential and the higher its polarity,
the better the hole capturing efficacy.5 A more negative redox potential provides larger
thermodynamic driving force for the oxidation of the scavenger via VB holes. The redox potential
of methanol (E°ox = 0.016 VNHE) is more negative than that of ethanol (E°ox = 0.084 VNHE), which
explains the higher photocatalytic rate measured with methanol. The polarity affects the scavenger
oxidation rate by influencing its adsorption kinetics on the photocatalyst surface.
S6
Figure S7 SEM images of different TiO2 nanotube arrays decorated with NPs dewetted from (a)
1 nm Pt – 1 nm Cu, (b) 2.5 nm Pt – 2.5 nm Cu and (c) 5 nm Pt – 5 nm Cu.
The dewetted metal nanoparticles feature a certain size distribution (Fig. 1, S3 and S6) and their
size depends on the location at which they form due to various factors. As outlined in previous
work4 on solid state dewetting of thin metal films, inherent mechanism and its occurrence
specifically on TiO2 NT surfaces, a most relevant factor determining the particle size and
distribution is the initial thickness of the sputtered metal film. Due to the porous morphology of
the NT substrate and the directionality of the film deposition technique used, the formed metal
films are typically slightly thicker at the tube top rims and at the tube bottoms. The metal films are
on the other hand slightly thinner at the inner surface of the NT side walls. Bearing in mind that
the thinner the film the smaller the dewetted particles (and vice versa, i.e. the thicker the film the
bigger the particles), particles formed at the surface of the NT side walls are smaller, and are bigger
when formed at the tube top rims and in the tube bottoms – this specific size distribution of the
dewetted particles is evident from the SEM images in Fig. 1 and S6.
Another factor can be related to the distribution of species from the anodizing electrolyte (e.g. F
or P species) adsorbed/incorporated in specific zones of the NT structure.1 Such species may
provide different surface chemistry at specific locations of the oxide substrate and hence different
wettability properties inducing different dewetting modes for the metal film. These features
however play likely a minor role compared to fluctuations of the metal film thickness.
S7
Figure S8 UV-Vis reflectance spectra of pristine TiO2 NTs and of Pt-, Cu- and PtCu-decorated
TiO2 nanotubes.
Scheme S1 Charge carrier generation (in TiO2), trapping (in the PtCu NPs) and transfer (to H+ for
H2 evolution).
The enhanced photocatalytic activity of TiO2 NTs decorated with dewetted-alloyed PtCu NPs is
ascribed to the electronic interaction between Pt and Cu atoms in the bimetallic NPs 10 – such
interaction can lead to a net increase of the electron density on Pt, enhancing the trapping efficiency
of photo-generated electrons from the TiO2 CB, decreasing charge recombination in the
semiconductor and consequently improving the photocatalytic H2 evolution rate.
S8
at% (XPS)
Sample
Ti
O
C
Pt
Cu
Pt2.5
7.9
25.2
46.6
20.3
-
Pt2.5Cu2.5
7.0
26.2
45.5
9.5
11.8
Cu2.5
15.1
66.2
9.4
-
9.3
Table S1 Surface atomic concentration for samples Pt2.5, Pt2.5Cu2.5 and Cu2.5 determined from
XPS data.
Sample Pt5
Atom
n
r (Å)
r0 (Å)
2 (Å2)
Pt1
12
2.762(5)
2.774
0.0054(5)
Pt2
6
3.94(5)
3.924
0.009(5)
Pt3
24
4.79(4)
4.805
0.012(4)
Sample Pt2.5Cu2.5
Atom
n
r (Å)
r0 (Å)
2 (Å2)
Pt1
12
2.71(1)
2.774
0.0061(5)
Pt2
6
3.94(3)
3.924
0.006(2)
Table S2 Results of fitting of EXAFS data at the Pt L3-edge.
S9
References
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
Yoo, J. E.; Schmuki, P. Critical Factors in the Anodic Formation of Extremely Ordered
Titania Nanocavities. J. Electrochem. Soc. 2019, 166 (11), C3389–C3398.
https://doi.org/10.1149/2.0381911jes.
Lee, K.; Mazare, A.; Schmuki, P. One-Dimensional Titanium Dioxide Nanomaterials:
Nanotubes. Chem. Rev. 2014, 114 (19), 9385–9454. https://doi.org/10.1021/cr500061m.
Riboni, F.; Nguyen, N. T.; So, S.; Schmuki, P. Aligned Metal Oxide Nanotube Arrays:
Key-Aspects of Anodic TiO2 Nanotube Formation and Properties. Nanoscale Horizons
2016, 1 (6), 445–466. https://doi.org/10.1039/c6nh00054a.
Altomare, M.; Nguyen, N. T.; Schmuki, P. Templated Dewetting: Designing Entirely SelfOrganized Platforms for Photocatalysis. Chem. Sci. 2016, 7 (12), 6865–6886.
https://doi.org/10.1039/C6SC02555B.
Al-Azri, Z. H. N.; Chen, W.-T.; Chan, A.; Jovic, V.; Ina, T.; Idriss, H.; Waterhouse, G. I.
N. The Roles of Metal Co-Catalysts and Reaction Media in Photocatalytic Hydrogen
Production: Performance Evaluation of M/TiO2 Photocatalysts (M = Pd, Pt, Au) in
Different Alcohol–water Mixtures. J. Catal. 2015, 329, 355–367.
https://doi.org/10.1016/j.jcat.2015.06.005.
Galińska, A.; Walendziewski, J. Photocatalytic Water Splitting over Pt−TiO 2 in the
Presence of Sacrificial Reagents. Energy & Fuels 2005, 19 (3), 1143–1147.
https://doi.org/10.1021/ef0400619.
Denisov, N.; Yoo, J.; Schmuki, P. Effect of Different Hole Scavengers on the
Photoelectrochemical Properties and Photocatalytic Hydrogen Evolution Performance of
Pristine and Pt-Decorated TiO2 Nanotubes. Electrochim. Acta 2019, 319, 61–71.
https://doi.org/10.1016/j.electacta.2019.06.173.
Li, Y.; Wang, B.; Liu, S.; Duan, X.; Hu, Z. Synthesis and Characterization of Cu2O/TiO2
Photocatalysts for H2 Evolution from Aqueous Solution with Different Scavengers. Appl.
Surf. Sci. 2015, 324, 736–744. https://doi.org/10.1016/j.apsusc.2014.11.027.
Bahruji, H.; Bowker, M.; Davies, P. R.; Pedrono, F. New Insights into the Mechanism of
Photocatalytic Reforming on Pd/TiO2. Appl. Catal. B Environ. 2011, 107 (1–2), 205–209.
https://doi.org/10.1016/j.apcatb.2011.07.015.
Jung, M.; Hart, J. N.; Boensch, D.; Scott, J.; Ng, Y. H.; Amal, R. Hydrogen Evolution via
Glycerol Photoreforming over Cu–Pt Nanoalloys on TiO2. Appl. Catal. A Gen. 2016, 518,
221–230. https://doi.org/10.1016/j.apcata.2015.10.040.
S10
Appendix D
Photoelectrochemical performance of facet-controlled TiO2 nanosheets
grown hydrothermally on FTO
Fahimeh Shahvardanfard, Gihoon Cha, Nikita Denisov, Benedict Osuagwua
and Patrik Schmuki
Nanoscale Adv., 2021, 3, 747–754
119
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Nanoscale
Advances
View Article Online
PAPER
View Journal | View Issue
Cite this: Nanoscale Adv., 2021, 3, 747
Photoelectrochemical performance of facetcontrolled TiO2 nanosheets grown hydrothermally
on FTO†
Fahimeh Shahvardanfard,‡a Gihoon Cha,‡a Nikita Denisov,‡a Benedict Osuagwua
and Patrik Schmuki *abc
Single crystal anatase TiO2 nanosheets (TiO2-NSs) are grown hydrothermally on fluorine-doped tin oxide
(FTO). By systematically changing the hydrothermal conditions such as reaction time, initial
concentration of Ti precursor, F precursor, and HCl as an additive, a wide variety of TiO2-NSs, with
different morphologies and faceting have been synthesized. For the different morphologies and different
facet ratios (anatase S001/S001+101), the photoelectrochemical response is characterized and compared.
We find that for photoanodes in neutral electrolytes, the magnitude of the photocurrent depends
strongly on the growth parameters, that is, peak IPCEs can vary from 11.7% to 61%. For a wide range of
parameters, the key parameter deciding on the photocurrent is the effective electrochemically active
Received 4th December 2020
Accepted 14th December 2020
DOI: 10.1039/d0na01017k
rsc.li/nanoscale-advances
area of the electrode. Only for very high facet ratios >91%, the photoresponse can be strongly influenced
by faceting – for samples with a S001/S001+101 of 91%, IPCE value of z84% is obtained. This work defines
not only optimized synthesis conditions for a most effective growth of these single crystalline electrode,
but also represents fundamental data for further applications of such electrodes.
Introduction
TiO2 in various geometries and polymorphs has over decades
attracted strong scientic and technological interest, due to its
large variety of functional features.1–6 Among others, its semiconductive nature made the oxide a versatile part of solar cells,
photoanodes or photocatalysts. In general, among the different
titania polymorphs, it is conceived that anatase is the most
reactive compound, due to a favorable energetics relative to the
red-ox potential of water.7–10
Even more, for various photocatalytic reactions it has been
shown that the different facets of anatase play a crucial role
for electron and hole transfer. On an anatase single crystal
that exposes the major (101) and (001) planes, typically lightgenerated electrons exit from the (101) plane while holes exit
from the (001) plane, if in contact with an aqueous environment – this due to the intrinsic electronic junction established by the different surface energies of the two facets.11,12
a
Institute for Surface Science and Corrosion WW4-LKO, Department of Materials
Science and Engineering, University of Erlangen-Nuremberg, Martensstrasse 7,
91058 Erlangen, Germany. E-mail: schmuki@ww.uni-erlangen.de
b
Chemistry Department, Faculty of Sciences, King Abdulaziz University, 80203 Jeddah,
Kingdom of Saudi Arabia
c
Regional Centre of Advanced Technologies and Materials, Palacky University
Olomouc, 17. listopadu 50A, 772 07 Olomouc, Czech Republic
† Electronic supplementary
10.1039/d0na01017k
information
(ESI)
available.
See
DOI:
Well dened growth of anatase crystallites with a controllable
facet ratio was only established convincingly in 2008, when
Yang et al. reported synthesis of nanoscale single crystal
anatase TiO2-nanosheets (NSs).13 The authors showed that
crystallites with dominant (001) facets could be fabricated by
using hydrothermal synthesis in solutions containing uorine ions (F). Fluoride termination of anatase can decrease
the surface energy of the (001) facets to a value lower than that
of (101) facets, and thus single crystal anatase TiO2-NSs with
a high percentage of (001) facets could be produced.13–16
Meanwhile these sheets have been widely investigated not
only in photocatalysis but also for many other
applications.17–19
However, in order to use faceted powders (nanosheets) in
photoelectrochemical applications, the sheets need to be
immobilized on a conductive substrate. Most elegantly this can
be achieved by growing faceted nanosheets directly via
a hydrothermal process on an FTO-substrate.20,21
Here, we grow these shape-controlled anatase single crystals on FTO under different hydrothermal parameters to attain
directly photoelectrodes consisting of a layer of anatase single
crystals with a range of individual crystallite morphology and
faceting. We then study the photoelectrochemical performance of the different electrodes and evaluate particularly the
effect of the electrochemically active surface area and the ratio
of facets ratios (S001/S001+101) on the magnitude of the
photocurrent.
‡ These authors contributed equally to this work.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Nanoscale Adv., 2021, 3, 747–754 | 747
View Article Online
Nanoscale Advances
Experimental section
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Preparation of anatase TiO2 nanosheets
FTO substrates (7 U m2, Solaronix) were cleaned by ultrasonication in acetone and ethanol, and then dried with
nitrogen. TiO2-NSs were grown through a hydrothermal process
directly on the FTO surfaces. For the basic recipe synthesis,
1.5 mL of titanium isopropoxide (Sigma Aldrich) was dropped
into an equal volume (30 mL : 30 mL) of a mixture of HCl (37%,
Sigma Aldrich) and DI water in a Teon-lined stainless steel
autoclave with a capacity of 250 mL. Aer stirring for 15
minutes, 0.5 g ammonium hexauorotitanate (Sigma Aldrich)
was added to the solution and further stirred for 15 minutes.
Then, the pre-cleaned FTO was inserted in the solution facing
down. The hydrothermal synthesis was conducted at 150 C for
8–20 hours. Finally, the as-prepared layers were washed with DI
water and annealed at 450 C for 1 hour in air to remove
residual uoride.
XPS results for the samples fabricated with the basic recipe
and higher concentration of HCl (HCl/DI: 33/27) are shown in
Fig. S1.† In general, the data conrm the as grow structures to
be terminated by F. The role of HCl is mainly to alter the H+
concentration (see ref. 21).
Characterization
A eld-emission scanning electron microscope (FE-SEM, S4800,
Hitachi), X-ray diffraction (XRD, X'pert Philips MPD diffractometer, using a Panalytical X'celerator detector and graphite
monochromized Cu Ka radiation, l ¼ 1.54056 Å), high resolution transmission electron microscopy (HR-TEM, Philips
CM30), and a selected area electron diffraction (SAED) analysis
were used to study the morphology and crystalline structure of
the nanostructured TiO2. X-ray photoelectron spectroscopy
Paper
(XPS, PHI 5600, US) was performed to study the chemical
composition of the samples. The density of the nanosheets
(number of NSs per mm2) was measured by counting the
number of nanosheets in one square micrometer via SEM
images of 20k magnication. The facet ratio of S001/S001+101 was
evaluated via SEM images by calculating the ratio of surface area
of (001) to the total surface area of facets.
Photocurrent spectra were measured in 0.1 M Na2SO4 in
presence and absence of MeOH at an applied potential of
500 mV (vs. Ag/AgCl) in a three-electrode setup using 150 W Xelamp (Oriel 6365) and Oriel Cornerstone 7400 1/8 m
monochromator.
Electrochemical impedance spectra (EIS) were measured in
a 3-electrode cell, in 0.1 M Na2SO4 at voltages of 0.2 to 0.8 V
in a frequency range of 0.1 Hz to 1 MHz. The curves were tted
with a Randles circuit using the Zahner IM5d soware. The
surface area was extracted from the double-layer capacitance
(Cdl), as C is a function of A (A ¼ electrochemically active surface
area).
Results and discussion
In a rst set of experiments, various growth parameters for the
faceted crystallites were varied, as described in the experimental
section. In line with literature, the morphology and faceting of
the NS, grown on FTO, could be greatly inuenced by the
hydrothermal synthesis conditions, that is specically: the
concentration of the titania and uoride precursor as well as
reaction time and additions of DI water and HCl.22–24 This is
illustrated step-by-step below.
The SEM images in Fig. 1a shows akes that are grown for
15 h in a solution containing 0.024 g mL1 titanium isopropoxide with a variation in the concentration of ammonium
Top view and cross-sectional SEM images of TiO2 nanosheets grown on FTO using (a) different concentrations of F precursor, (b) different
concentrations of Ti precursor.
Fig. 1
748 | Nanoscale Adv., 2021, 3, 747–754
© 2021 The Author(s). Published by the Royal Society of Chemistry
View Article Online
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Paper
hexauorotitanate (F ions) from 0.004, 0.008, 0.0125 to 0.016 g
mL1. Fig. S2† shows a TEM image and SAED pattern for the
TiO2 nanosheets synthesized with the above recipe using
a uoride concentration of 0.008 g mL1 (basic recipe). The
nanosheets show crystalline lattice fringes with a spacing of 3.5
Å and 2.4 Å (Fig. S2a†), which correspond to (101) and (004)
crystallographic planes of anatase, respectively. These planes
have also been identied from the SAED patterns (Fig. S2b†)
along with other characteristic anatase planes. The akes in
Fig. 1b show the same basic recipe using a uoride concentration of 0.008 g mL1 but with a change in the concentration of
titanium isopropoxide (Ti4+ ions) from 0.016, 0.024 to 0.032 g
mL1. The results in Fig. 1a and b illustrate that by increasing
the F precursor from 0.004 g mL1 to 0.0125 g mL1, the
nanosheets become increasingly preferentially faceted towards
(001). Also the akes become larger and thus the entire layer on
FTO thicker. The layer thickness of the sheets typically
increased from 0.6 mm to 2.1 mm. At the same time, the akes
are initiated at a lower density on the FTO substrate. Roughly
the density of the nanosheets decreases from z16.8 NSs/mm2
for the 0.004 g F mL1 to z5.3 NSs/mm2 and z2.3 NSs/mm2 for
0.008 and 0.012 g F mL1, respectively. At a F precursor
concentration of 0.016 g mL1, the thinnest nanosheets (73 nm)
are obtained but they are very sparsely grown on the FTO
substrate (z3.2 NSs/mm2). From SEM images of Fig. 1b, it can
be seen that according to the change in Ti to F ratio, the layer
Nanoscale Advances
thickness and density of grown nanosheets on FTO increases
with an increasing amount of Ti precursor from 0.016 g mL1 to
0.032 g mL1. For 0.016 g mL1 Ti precursor, the nanosheets
grow in a dispersed manner, with a high ratio of (001) to (101)
facets (S001/S001+101 z 93%).
Fig. 2a and b show XRD spectra corresponding to the TiO2NSs shown in Fig. 1a and b. Overall, the diffraction peaks that
originate from titania (i.e. the oxide peaks) t well to the TiO2
tetragonal anatase phase.25 All other XRD peaks can be assigned
to the FTO substrate. The XRD spectra in Fig. 2b shows hardly
any anatase peak for low concentration of Ti precursor (0.016 g
mL1), due to the low density of sheets under these synthesis
conditions (Fig. 1b).
The different morphologies of Fig. 1a and b then were
characterized by photocurrent spectra that were measured in
0.1 M Na2SO4 at 500 mV (vs. Ag/AgCl). The photocurrent spectra
for the different samples (Fig. 2c) show the photon to current
conversion efficiency (IPCE) peaks at z40% and 42.6%,
respectively, for the sample produced with 0.004 g mL1 and
0.008 g mL1 of F precursor. Although a higher photocurrent
value is obtained at wavelengths higher than z340 nm for
nanosheets synthesized with 0.0125 g mL1 F precursor, the
samples synthesized with a lower concentration of F show
a higher magnitude of the maximum photocurrent. In general,
the results show that the coverage (as expected) is a rst
dominant factor that determines the overall photoresponse of
Fig. 2 (a) X-ray diffraction (XRD) patterns of TiO2 nanosheets grown on FTO using different concentrations of F precursor, (b) XRD pattern of
TiO2 nanosheets grown on FTO with different concentrations of Ti precursor, (c) IPCE spectra of TiO2 nanosheets fabricated with different
concentrations of F and Ti precursors in 0.1 M Na2SO4 at 0.5 V (vs. Ag/AgCl), (d) band gap evaluation of TiO2 nanosheets fabricated with different
concentrations of F and Ti precursors.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Nanoscale Adv., 2021, 3, 747–754 | 749
View Article Online
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Nanoscale Advances
Fig. 3
Paper
Top view and cross-sectional SEM images of TiO2 nanosheets grown on FTO for different synthesis times.
the photoanodes. The low density of nanosheets (z1.1 NSs/
mm2) obtained with lower concentration of Ti (0.016 g mL1)
shows a maximum IPCE of only z11.7%.
Fig. 2d shows the band gap evaluation of different layers of
Fig. 1a and b. All evaluations show a value of z3.1 eV, i.e.
consistent with literature data for anatase,26,27 except the sample
synthesized with a concentration of Ti precursor of 0.016 g
mL1. This sample shows a seemingly higher band gap (>3.2 eV)
– this result is however due FTO that is exposed to the
electrolyte.
In order to systematically change the coverage and layer
thickness of the nanosheets, we prepared a series of ake
electrodes grown for different times. I.e. we kept the precursor
composition of the best performing sample in Fig. 1 but
changed the hydrothermal treatment time. The effect of
reaction time on the morphology of the resulting TiO2-NSs is
shown in Fig. 3. It is apparent that an increase in treatment time
yields an increase in TiO2-NSs growth density (coverage) as well
as the thickness of the TiO2-NS-layer. The cross sectional SEM
images (Fig. 3) show that the layer thickness gradually increases
from 0.98 mm obtained for a reaction time of 8 h to 1.1 mm, 1.5
mm and 1.9 mm for reaction times of 11 h, 15 h and 20 h,
respectively. Fig. 4a shows the layer thickness of the nanosheets
as a function of the synthesis time, suggesting that the growth
rate of nanosheets is approximately linear throughout hydrothermal synthesis time, until the whole precursor is
consumed.20 Thus, the hydrothermal reaction time can be used
to adjust the thickness of the synthesized layers.
Fig. 4b shows XRD patterns of the samples obtained for
different hydrothermal treatment time. The data all conrm the
Fig. 4 (a) Layer thickness as a function of synthesis time, (b) X-ray diffraction (XRD) patterns of TiO2 nanosheets grown on FTO for different
synthesis times (c) IPCE spectra of TiO2 nanosheets for different synthesis times in 0.1 M Na2SO4 at 0.5 V (vs. Ag/AgCl) (d) band gap evaluation of
TiO2 nanosheets fabricated for different synthesis time.
750 | Nanoscale Adv., 2021, 3, 747–754
© 2021 The Author(s). Published by the Royal Society of Chemistry
View Article Online
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Paper
formation of anatase. The disappearance of the FTO peaks for
samples exposed for 8 h up to 20 h is in line with a growth of the
layer thickness from z1 mm to 2 mm for the longest exposition
time.
Photocurrent spectra in Fig. 4c show that with an increase in
reaction time and layer thickness, the IPCE rst increases and
then decreased with further treatment time. A maximum IPCE
of z32, 34, 42.6 and 37% is obtained for samples grown for 8,
11, 15 and 20 h respectively – i.e. the TiO2-NS arrays grown by
15 h reaction to a thickness of 1.5 mm show the highest photoelectrochemical performance.
The band gap evaluation in Fig. 4d for all samples show
a value of z3.1 eV which again conrms anatase.
To assess the inuence of the exposed surface area to the
electrolyte we characterized the samples formed for the
different reaction times with electrochemical impedance spectroscopy (EIS). Here the electrochemically active surface area of
the TiO2-NSs samples can be derived from the double layer
capacitance.28 In brief, if a semiconductor is forward biased,
that is, for an n-type material it is held at potentials negative to
the at band potential (U), it behaves like a metal and this
allows us to obtain the double layer capacitance from the
capacitance data (for titania in a neutral solution U is at
approximately 0.2 to 0.4 V (vs. Ag/AgCl)).29,31 Fig. 5a shows
examples of Nyquist plots for the different samples at a potential of 0.4 V Ag/AgCl. The data are tted with a classic Randall's
circuit consisting of a constant phase element (CPE) parallel to
the charge transfer resistance (Rct) and in series with the ohmic
resistance (Rs). The capacitance extracted is then taken as the
Nanoscale Advances
double layer capacitance. Fig. 5b shows the double layer
capacitance (Cdl) as a function of voltage obtained from a series
of EIS measurements. Cdl then is related to the surface area by
Cdl f A. The evaluation of capacitance yields for a TiO2-NS layer
grown for 8 h a capacitance of 20.9 mF. For a higher reaction
time, 15 h, Cdl increases up to 41 mF. Further increase of reaction time to 20 h then yield a reduced Cdl of 19.4 mF (A/A0 ¼ C/C0,
where A0 and C0 are the active surface area and capacitance of
the sample synthesized with the basic recipe: 1.5 mL Ti
precursor, 0.5 g F precursor, HCl/DI: 30/30 for 15 h). This
reects rst an increase of the electrochemically active surface
area followed by a decrease that can be attributed to the fact that
the thicker layers become also increasingly intergrown. In other
words, the data obtained from Cdl measurements (active area)
correlates well with differences observed in photoelectrochemical activities of TiO2-NSs in Fig. 4c. Hence, variations in the photoresponse can be ascribed mainly to the
inuence of the electrochemically active surface area.
We then evaluate the variation in the surface area of the
samples in Fig. 1 as well. Fig. S3† indicates the double layer
capacitance as a function of voltage for samples in Fig. 1 which
are synthesized with different concentrations of Ti and F
precursors. As one can see, a coherent trend is observed for all
samples between the photoelectrochemical activity and extracted surface area from EIS.
In other words, the parameters investigated up to here show
that all variations in the IPCE can be explained by the active
surface area. The different faceting seems not to majorly affect
the photoresponse.
Fig. 5 (a) Nyquist plot measured in 0.1 M Na2SO4 at 0.4 V vs. Ag/AgCl fitted with an equivalent circuit for TiO2 nanosheets grown for different
synthesis times, (b) double layer capacitance as a function of voltage of TiO2 nanosheets grown for different synthesis times, (c) impedance
parameters at 0.4 V.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Nanoscale Adv., 2021, 3, 747–754 | 751
View Article Online
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Nanoscale Advances
However, another tool reported in literature used to tune
crystal growth to achieve even more pronounced faceting is the
addition of HCl (H+ ions) to the hydrothermal synthesis solution. Therefore, in an additional set of experiments we used
different volume ratios of HCl to DI water of 33 : 27, 30 : 30 and
20 : 40 (in a total volume of 60 mL) as an addition to the base
recipe of 0.024 g mL1 titanium isopropoxide and 0.008 g mL1
ammonium hexauorotitanate and a synthesis time of 15 h.
The SEM images in Fig. 6a shows that HCl addition has a great
effect on the morphology and the facet ratio of the nal layer.
For a high amount of HCl (H+ ions) and low amount of DI water
(HCl to DI of 40 : 20), no nanosheets grew on the surface of FTO
(Fig. S4†). By using a HCl to DI water ratio of 33 : 27, very thin
nanosheets with the thickness of z49 nm with (001)-dominant
Fig. 6 Top view and cross-sectional SEM images of TiO2 nanosheets
grown on FTO with different ratios of HCl to DI water.
Paper
facet (S001/S001+101 z 91%) could be grown on the FTO substrate
(Fig. 6a). On the other hand, by decreasing the amount of HCl,
very thick nanosheets with a thickness of z760 nm with (101)dominant facet (S001/S001+101 z 50%) could be grown. From
cross sectional SEM images (Fig. 6b) one can see that by
decreasing the amount of HCl, the layer thickness increases
from 0.9 mm obtained for HCl to DI water ratio of 33 : 27 to 1.8
mm for an HCl to DI water ratio of 20 : 40. This example shows
that in order to fully design TiO2 nanosheets on FTO, adjusting
the aqueous phase by HCl is very effective.
Fig. 7a shows the XRD pattern for these TiO2 nanosheets that
all conrm the formation of TiO2 anatase.
Photocurrent spectra of the nanosheet layers in Fig. 7b show
a similar bandgap (Fig. 7c) of 3.1 eV as obtained previously with
all synthesized samples. The highest IPCE value (61%) is
observed for the sample synthesized with a volume ratio of HCl
to DI water of 33 : 27. The capacitance data for the best performing strongly faceted sample in Fig. S3a† of 38–41 mF is
virtually the same as for the best performing basic sample
(Fig. 5) – reecting a similar electrochemical area. However, the
highly preferentially (001) faceted sample in Fig. 6 yields an
IPCE of 61% while at the “regular” facet ratio only an IPCE of
approximately 40% is reached.
An evaluation of the facet ratio (S001/S001+101) yields z 91%
for highly faceted sample in Fig. 6 and z67% for the regularly
faceted sample in Fig. 6. The higher IPCE thus may be associated with higher percentage of (001) facets that are reported to
be favorable for oxidative reactions.18,30 Similarly, samples with
(101)-dominant facet (S001/S001+101 z 50%), despite their high
Fig. 7 (a) X-ray diffraction (XRD) patterns of TiO2 nanosheets grown on FTO with different ratios of HCl to DI water, (b) IPCE spectra of TiO2
nanosheets with different ratios of HCl to DI water, in 0.1 M Na2SO4 at 0.5 V (vs. Ag/AgCl) (c) band gap evaluation of TiO2 nanosheets fabricated
with different ratio of HCl to DI water.
752 | Nanoscale Adv., 2021, 3, 747–754
© 2021 The Author(s). Published by the Royal Society of Chemistry
View Article Online
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Paper
Nanoscale Advances
Fig. 8 (a) IPCE spectra of TiO2 nanosheets grown on FTO with different ratio of HCl to DI in 0.1 M Na2SO4 at 0.5 V (vs. Ag/AgCl) with and without
MeOH, (b) band gap evaluation of TiO2 nanosheets fabricated with different ratio of HCl to DI water.
active surface area (Fig. S5†), show even lower IPCE values
(z40%) when compared with the other two samples, due to the
lower percentage of (001) facets which retards the oxidative
reactions.21,30,31
To further elucidate the effects of faceting on electrochemical reaction, photocurrent spectra of a series of samples
in Fig. 6 were measured in presence of a hole capture agent
(20% methanol).32,33 The results are shown in Fig. 8a. Only for
the highest faceting ratio (91%), the hole transfer rate to the
electrolyte becomes rate determining. For lower facet ratio
samples, the photocurrent data are hardly at all affected by the
presence of hole capture agent. As a note, the band gap evaluation in Fig. 8b shows that the band gap values are similar in
presence and absence of MeOH.
Overall, this shows that for photoelectrochemical applications of this type of facet controlled anatase layers, rst of all the
effective electrochemical area is dominating as a critical factor
for the photoresponse. The gain additional benet from faceting, a high preferential faceting (S001/S001+101 > 90%) is needed.
Such layers than can reach IPCE maxima for >80% under ideal
conditions.
from the benecial effect of the larger electrochemically active
surface area and secondly from a high percentage of (001)
crystallographic facets. However, facet effects become only
apparent, if the electrode has a very high ratio of 001/101 facets.
Under optimized conditions, these electrodes provide
a remarkable IPCE of >80%. These electrodes therefore represent a highly valuable nanostructure for wide applications in
photoelectrohemistry.
Conflicts of interest
There are no conicts to declare.
Acknowledgements
The authors would like to acknowledge the DFG and the
Operational Research Program, Development and Education
(European Regional Development Fund, Project No. CZ.02.1.01/
0.0/0.0/15_003/0000416 of the Ministry of Education, Youth and
Sports of the Czech Republic) for nancial support.
Conclusion
References
The present work systematically investigates the synthesis
conditions on the morphology and structure of hydrothermally
grown TiO2-NSs on FTO and their subsequent photoelectrochemical activity. The morphology and structure of TiO2NSs can be largely affected by hydrothermal conditions, such as
reaction time and reactant concentrations. We further explored
the effects of morphology and structure of the synthesized
nanosheets on their photoelectrochemical activity. We nd that
optimizing of the morphology and faceting of the nanosheet
layers can strongly improve the photoelectrochemical performance of the layer. Performance improvement arises rstly
1 P. Roy, S. Berger and P. Schmuki, Angew. Chem., Int. Ed.,
2011, 50, 2904–2939.
2 M. Pelaez, N. T. Nolan, S. C. Pillai, M. K. Seery, P. Falaras,
A. G. Kontos, P. S. M. Dunlop, J. W. J. Hamilton,
J. A. Byrne, K. O'Shea, M. H. Entezari and D. D. Dionysiou,
Appl. Catal., B, 2012, 125, 331–349.
3 I. Paramasivam, H. Jha, N. Liu and P. Schmuki, Small, 2012,
8, 3073–3103.
4 J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang,
Y. Horiuchi, M. Anpo and D. W. Bahnemann, Chem. Rev.,
2014, 114, 9919–9986.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Nanoscale Adv., 2021, 3, 747–754 | 753
View Article Online
Open Access Article. Published on 21 December 2020. Downloaded on 8/3/2021 11:36:08 AM.
This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
Nanoscale Advances
5 K. Lee, A. Mazare and P. Schmuki, Chem. Rev., 2014, 114,
9385–9454.
6 H. Tsuchiya and P. Schmuki, Nanoscale, 2020, 12, 8119–8132.
7 T. Luttrell, S. Halpegamage, J. Tao, A. Kramer, E. Sutter and
M. Batzill, Sci. Rep., 2015, 4, 4043.
8 N. Strataki, V. Bekiari, D. I. Kondarides and P. Lianos, Appl.
Catal., B, 2007, 77, 184–189.
9 M. Ni, M. K. H. Leung, D. Y. C. Leung and K. Sumathy,
Renewable Sustainable Energy Rev., 2007, 11, 401–425.
10 K. Hashimoto, H. Irie and A. Fujishima, Jpn. J. Appl. Phys.,
2005, 44, 8269–8285.
11 K. Lv, B. Cheng, J. Yu and G. Liu, Phys. Chem. Chem. Phys.,
2012, 14, 5349.
12 T. Ohno, K. Sarukawa and M. Matsumura, New J. Chem.,
2002, 26, 1167–1170.
13 H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith,
H. M. Cheng and G. Q. Lu, Nature, 2008, 453, 638–641.
14 W. Q. Fang, J. Z. Zhou, J. Liu, Z. G. Chen, C. Yang, C. H. Sun,
G. R. Qian, J. Zou, S. Z. Qiao and H. G. Yang, Chem.–Eur. J.,
2011, 17, 1423–1427.
15 M. Dozzi and E. Selli, Catalysts, 2013, 3, 455–485.
16 Q. Wu, M. Liu, Z. Wu, Y. Li and L. Piao, J. Phys. Chem. C,
2012, 116, 26800–26804.
17 Z. Xiong, Z. Lei, Y. Li, L. Dong, Y. Zhao and J. Zhang, J.
Photochem. Photobiol., C, 2018, 36, 24–47.
18 A. Meng, J. Zhang, D. Xu, B. Cheng and J. Yu, Appl. Catal., B,
2016, 198, 286–294.
19 P. Liu, X. Huo, Y. Tang, J. Xu, X. Liu and D. K. Y. Wong, Anal.
Chim. Acta, 2017, 984, 86–95.
20 D. Zhong, Q. Jiang, B. Huang, W.-H. Zhang and C. Li, J.
Energy Chem., 2015, 24, 626–631.
754 | Nanoscale Adv., 2021, 3, 747–754
Paper
21 T. Butburee, P. Kotchasarn, P. Hirunsit, Z. Sun, Q. Tang,
P. Khemthong, W. Sangkhun, W. Thongsuwan,
P. Kumnorkaew, H. Wang and K. Faungnawakij, J. Mater.
Chem. A, 2019, 7, 8156–8166.
22 J. Li, Y. Yu, Q. Chen, J. Li and D. Xu, Cryst. Growth Des., 2010,
10, 2111–2115.
23 J. Yang, W. Li, J. Li, D. Sun and Q. Chen, J. Mater. Chem.,
2012, 22, 17744.
24 C. Canevali, S. Polizzi, A. Testino, I. R. Bellobono and
V. Buscaglia, J. Am. Chem. Soc., 2017, 129, 3564–3575.
25 S. Feng, J. Yang, H. Zhu, M. Liu, J. Zhang, J. Wu and J. Wan, J.
Am. Ceram. Soc., 2011, 94, 310–315.
26 S. Tanemura, L. Miao, W. Wunderlich, M. Tanemura,
Y. Mori, S. Toh and K. Kaneko, Sci. Technol. Adv. Mater.,
2005, 6, 11–17.
27 A. D. Paola, M. Bellardita and L. Palmisano, Catalysts, 2013,
3, 36–73.
28 M.-L. Tremblay, M. H. Martin, C. Lebouin, A. Lasia and
D. Guay, Electrochim. Acta, 2010, 55, 6283–6291.
29 A. Adán-Más, T. M. Silva, L. Guerlou-Demourgues and
M. F. Montemor, Electrochim. Acta, 2018, 289, 47–55.
30 X. Hu, S. Lu, J. Tian, N. Wei, X. Song, X. Wang and H. Cui,
Appl. Catal., B, 2019, 241, 329–337.
31 L. Ruan, X. Wang, T. Wang, Z. Ren, Y. Chen, R. Zhao,
D. Zhou, G. Fu, S. Li, L. Gao, Y. Lu, Z. Wang, H. Tian,
X. Kong and G. Han, ACS Appl. Mater. Interfaces, 2019, 11,
37256–37262.
32 D. Dong, C. Yan, J. Huang, N. Lu, P. Wu, J. Wang and
Z. Zhang, J. Mater. Chem. A, 2019, 7, 24180–24185.
33 M. Shen and M. A. Henderson, J. Phys. Chem. Lett., 2011, 2,
2707–2710.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Electronic Supplementary Material (ESI) for Nanoscale Advances.
This journal is © The Royal Society of Chemistry 2020
Supporting Information
Photoelectrochemical performance of facet-controlled TiO2 nanosheets grown
hydrothermally on FTO
Fahimeh Shahvaranfard,a,# Gihoon Cha,a,# Nikita Denisov,a,# Benedict Osuagwu,a
Patrik Schmukia,b,c*
a Institute for Surface Science and Corrosion WW4-LKO, Department of Materials Science and
Engineering, University of Erlangen-Nuremberg, Martensstrasse 7, 91058 Erlangen, Germany
b Chemistry Department, Faculty of Sciences, King Abdulaziz University, 80203 Jeddah,
Kingdom of Saudi Arabia
c Regional Centre of Advanced Technologies and Materials, Palacky University Olomouc, 17.
listopadu 50A, 772 07 Olomouc, Czech Republic
# These authors contributed equally to this work.
* Corresponding Author. E-mail:
schmuki@ww.uni-erlangen.de
1
(e)
at% (XPS)
Sample
Ti
O
C
Sn
F
Cl
HCl/DI:30/30
25.38
56.44
10.13
0.12
7.58
0.34
HCl/DI:33/27
22.13
56.16
15.43
0.61
4.96
0.71
Fig. S1. XPS spectra for (a) F 1s and (b) Cl 2p region of samples grown with the basic recipe
(a,b), and of sample grown with higher concentration of HCl (HCl to DI of 33:27), (c,d) and
2
surface atomic concentration determined from XPS data (e). Evident is a different level of
fluoride termination for the two samples but only a very weak influence of chloride.
Fig. S2. HR-TEM image (a) and the corresponding SAED pattern (b) for the anatase TiO2
nanosheets grown by hydrothermal treatment using the basic recipe at 150 °C for 15 h.
3
(c)
*A/A0 = C/C0
CPE
a
252
Cdl
(µF)
12.3
Normalized
Surface
Area*
0.97
0.3
109
64
35.9
0.95
0.9
0.008 g mL-1
119
243
41
0.96
1
0.0125 g mL-1
139
146
25.7
0.93
0.6
0.016 g mL-1
330
530
17.5
0.96
0.4
Sample
Rs
(Ω)
Rct
(kΩ)
FTO
83
0.016 g mL-1
0.024 g mL-1
0.032 g mL-1
Sample
Rs
(Ω)
Rct
(kΩ)
FTO
83
0.004 g mL-1
(d)
CPE
a
252
Cdl
(µF)
12.3
Normalized
Surface
Area*
0.97
0.3
620
509
13.1
0.96
0.3
119
243
41
0.96
1
97
906
23.5
0.96
0.6
*A/A0 = C/C0
Fig. S3 Double layer capacitance as a function of voltage for TiO2 nanosheets for (a) different
concentrations of F precursor, (b) different concentrations of Ti precursor (c) Impedance
parameters at -0.4 V for TiO2 nanosheets at different concentrations of F precursor, (d)
Impedance parameters at -0.4 V for TiO2 nanosheets at different concentrations of Ti precursor.
4
Fig. S4 Top view and cross-sectional SEM images of TiO2 nanosheets grown on FTO at HCl to
DI water: 40/20.
5
(b)
Sample
FTO
33:27
30:30
20:40
CPE
Normalized
Surface
Cdl
a
Area*
(µF)
83 252 12.3 0.97
0.3
107 183
38 0.96
0.9
119 243
41 0.96
1
91 1100 45.6 0.91
1.1
Rs
(Ω)
Rct
(kΩ)
*A/A0 = C/C0
Fig. S5 (a) Double layer capacitance as a function of voltage for TiO2 nanosheets grown with
different ratios of HCl to DI water, (b) Impedance parameters at -0.4 V.
6