没有合适的资源?快使用搜索试试~ 我知道了~
研究从非均相到光催化的过程的多功能模型系统:TiO2(110)上的外延RuO2(110)
0 下载量 41 浏览量
2021-03-24
20:24:44
上传
评论
收藏 1.12MB PDF 举报
温馨提示
试读
11页
研究从非均相到光催化的过程的多功能模型系统:TiO2(110)上的外延RuO2(110)
资源推荐
资源详情
资源评论
Versatile Model System for Studying Processes Ranging from
Heterogeneous to Photocatalysis: Epitaxial RuO
2
(110) on TiO
2
(110)
Yunbin He,
†,‡
Daniel Langsdorf,
†
Lei Li,
‡
and Herbert Over*
,†
†
Department of Physical Chemistry, Justus-Liebig-University, Heinrich-Buff-Ring 58, D-35392 Gießen, Germany
‡
Hubei Collaborative Innovation Center for Advanced Organic Chemical Materials; Key Laboratory of Green Preparation and
Application for Functional Materials, Ministry of Education; Faculty of Materials Science & Engineering, Hubei University, Wuhan
430062, China
*
S
Supporting Information
ABSTRACT: The binary model system RuO
2
/TiO
2
(110) can be
prepared with single crystallinity and excellent control of the
morphology of the RuO
2
(110) nanoislands. The interface of
RuO
2
/TiO
2
(110) is structurally well-defined since RuO
2
grows
with the same lattice constants as TiO
2
(110). The actual growth of
RuO
2
on TiO
2
(110) single crystals starts from square-shaped 3−4
ML thick RuO
2
islands with narrow size and thickness
distributions. After TiO
2
(110) is completely covered by RuO
2
, the further growth proceeds via a step flow mechanism,
forming very large and flat RuO
2
(110) terraces with well-defined thickness. Both the flat RuO
2
(110) films and RuO
2
(110)
nanoislands are very reactive toward CO oxidation, and the RuO
2
(110) nanoislands are robust in the redox reactions, i.e., easily
recovering their morphology after reoxidation from the reduced state. The RuO
2
/TiO
2
(110) heterojunction forms a Schottky
barrier of 1.4 eV which is important for photocatalysis.
1. INTRODUCTION
Molecular insight into physical and chemical processes needs
the intimate collaboration of experiment and theory that in turn
resorts on the design and preparation of highly sophisticated
model systems. Proper model systems are at the heart of
physical chemistry, requiring lowest possible structural
complexity, while still capturing the most important properties
of the real system under study. Always the scientific question in
mind dictates the proper choice of the model system.
In heterogeneous catalysis the active component in the form
of nanoparticles is brought onto a supporting material, most
frequently an oxide support to stabilize the dispersion of the
active component.
1,2
Assuming the interaction between the
support and the active component is important for the catalytic
performance, then in a well-designed model system the support
and the metal particles but also the interfaces should exhibit a
simple structure.
Sumitomo Chemical has demonstrated that the Deacon
process, i.e., the gas phase oxidation of HCl by oxygen to
recover molecular chlorine, is most efficiently catalyzed with
RuO
2
supported on rutile TiO
2
.
3
The choice of the supporting
material has shown to be decisive. The activity is not only
determined by the chemical nature of the support but also by
the type of polymorph. For instance, RuO
2
supported on
anatase TiO
2
is by a factor of 3 less active than RuO
2
supported
on rutile TiO
2
.
3
Transmission electron microscopy (TEM)
investigations of the Sumitomo catalyst indicated that ultrathin
RuO
2
islands are formed on rutile TiO
2
preferentially oriented
along the (110) direction.
4
Obviously, a well-designed model
system for the Deacon reaction over a Sumitomo catalyst
should consist of RuO
2
(110) grown on a single-crystal rutile
TiO
2
(110).
This rutile RuO
2
/TiO
2
(110) heterostructure is also a proper
model system for the electrochemical oxidation of HCl,
mimicking the dimensionally stable anodes (DSA) in the
chlorine evolution reaction (CER).
5,6
Besides the CER, the
oxygen evolution reaction (OER) can be efficiently catalyzed by
DSA. As a four-electron process, the OER is sluggish and
constitutes therefore the bottleneck in various technologically
important electrolysis reactions such as the water splitting to
produce hydrogen.
Photocatalysis can be employed to produce hydrogen by
solar water splitting.
7
Here the photoabsorber generates first an
electron−hole pair by the absorption of a single photon with
suitable energy. Subsequently, the electron and hole can be
utilized in the reduction and oxidation of water to produce
hydrogen and oxygen, respectively. Frequently cocatalysts are
required for speeding up the sluggish OER.
8
A suitable model is
again RuO
2
on TiO
2
(110) where TiO
2
serves as the
photoabsorber, and RuO
2
acts as the cocatalyst to facilitate
the OER. Of particular importance in this system is the
interface between RuO
2
and TiO
2
, since photogenerated holes
need to traverse this interface to induce the OER at the RuO
2
surface.
In this paper we will report on the binary model system
RuO
2
/TiO
2
(110) that can be prepared with single crystallinity
Received: December 5, 2014
Revised: January 12, 2015
Published: January 12, 2015
Article
pubs.acs.org/JPCC
© 2015 American Chemical Society 2692 DOI: 10.1021/jp5121405
J. Phys. Chem. C 2015, 119, 2692−2702
and excellent control on the size distribution of the RuO
2
islands and the morphology of the RuO
2
(110) films. The
interface of RuO
2
/TiO
2
(110) is structurally well-defined since
RuO
2
grows pseudomorphically on the TiO
2
(110) substrate,
i.e., with the same lattice constants as TiO
2
(110), establishing a
Schottky barrier of 1.4 eV. The growth of RuO
2
on TiO
2
(110)
single crystals at 600 K is quite complex in that square-shaped
3−4 monolayer (ML) thick RuO
2
(110) islands are initially
formed with narrow size and thickness distributions. After
TiO
2
(110) is completely covered by 3−4 ML of RuO
2
, the
further growth proceeds via a step flow mechanism leading to
very large and atomically flat RuO
2
(110) terraces. Both flat
RuO
2
(110) films and RuO
2
(110) nanoislands were tested with
regard to the reactivity in the model CO oxidation reaction.
2. EXPERIMENTAL DETAILS
The experiments were conducted in a home-built three-
chamber ultrahigh-vacuum (UHV) system.
9
The sample can
be introduced via the load lock chamber that contains a small
sample manipulator and a magnetic rod for sample transfer to
the long-traveling sample manipulator in the analysis chamber.
This chamber houses a quadrupole mass spectrometer and a
dual X-ray source together with a hemispherical analyzer (PSP
Vacuum Technology) to perform X-ray photoelectron spec-
troscopy (XPS) experiments. In addition, the analysis chamber
contains an electron beam evaporator (tectra e-flux). The
scanning tunneling microscope (STM) chamber is separated
from the analysis chamber by a CF150 gate valve and separately
pumped by an ion getter pump (100 L/s). The sample can be
transferred from the analysis chamber to the STM chamber just
by translating the manipulator through the open CF150 gate
valve and placing then the sample plate into the STM (VT-
STM, Omicron) with a wobble stick. In general, we used
homemade tungsten tips for the STM experiments.
The sample temperature was measured with an infrared (IR)
pyrometer, which was precalibrated with a K type thermo-
couple. Prior to the film growth, the TiO
2
(110) crystal was
cleaned by cycles of Ar-ion sputtering (p(Ar) = 10
−6
mbar, U =
1.0−1.5 kV, I
emission
= 20 mA) and subsequent annealing at 950
Kin10
−7
mbar of oxygen. Ruthenium was deposited on
TiO
2
(110) by physical vapor deposition using a well-outgased
electron beam evaporator. With STM and XPS the deposition
rate of ruthenium was calibrated to be approximately 1
monolayer (ML) per 4 min. During the deposition of Ru the
TiO
2
(110) crystal was kept at 600 K in an oxygen atmosphere
of 10
−6
mbar in order to directly form RuO
2
rather than
metallic Ru on the titania surface. The temperature of 600 K
was chosen to keep mobility of the surface species high while
facing no problem of intermixing of Ti and Ru during the
growth.
10
All STM images presented in this paper were taken at
room temperature in the constant current mode with the
sample positively biased, representing empty states of the
surface. Typical sample voltage and tunneling current used for
scanning were 0.5−4.5 V and 0.8−1.5 nA, respectively.
For STM data processing the WsxM freeware was used. For
large scale areas the raw data were treated by a global plane
filter where the whole image was fitted to a single plane, which
was then subtracted from the original image. Additionally for
images of areas with a high number of steps a flatten filter
function was employed. On small scale areas the raw data were
flattened with a local plane filter where the image is fitted to a
plane which is defined by a chosen terrace in STM image.
Independent of the scale area, image optimization w as
per form ed by manual adjustme nt of the brightness, the
contrast, and a z control treatment which adjusts the user z-
scale of the image.
3. RESULTS AND DISCUSSION
3.1. Growth of RuO
2
on TiO
2
(110): Experimental
Results. The STM images in Figure 1 summarize the
morphology evolution of ruthenium with deposition time in
an oxygen atmosphere of 10
−6
mbar. After standard cleaning
procedures the TiO
2
(110) crystal typically shows atomically flat
surface with terraces separated by single atomic steps (cf. Figure
1a). For nominal 1 monolayer coverage (cf. Figure 1b),
ruthenium related islands grow preferentially at step edges of
Figure 1. STM images (300 nm × 300 nm) for RuO
2
islands and films grown on TiO
2
(110) for increasing deposition times: (a) clean TiO
2
(110),
(b) 4 min, (c) 8 min, (d) 12 min, (e) 16 min, (f) 20 min, (g) 24 min, and (h) 28 min. Four minutes of deposition time corresponds to about 1
monolayer of ruthenium. During growth the sample was kept at a temperature of 600 K in an O
2
atmosphere of 1 × 10
−6
mbar.
The Journal of Physical Chemistry C Article
DOI: 10.1021/jp5121405
J. Phys. Chem. C 2015, 119, 2692−2702
2693
TiO
2
(110). When starting from a defect-rich TiO
2
(110) surface
the islands are more homogeneously dispersed over the surface
(Figure S1 in Supporting Information). Additional deposition
of 1 ML of ruthenium leads the islands to grow in lateral size
(cf. Figure 1c). Quite surprisingly, the islands are quasi-square-
shaped, although the unit cell of the TiO
2
(110) substrate is
rectangular. Already the cuboid-like shape of the islands
indicates that RuO
2
rather than metallic Ru is formed. Metallic
Ru forms rather clusters or round-shaped particles o n
TiO
2
(110) (cf. Figure S2 in Supporting Information). From
accompanying XPS experiments the o xidation of Ru is
corroborated: RuO
2
is clearly visible in the Ru 3d emission
(cf. chapters 3.3 and 3.4). Therefore, we conclude that flat
RuO
2
islands can be formed on TiO
2
(110) under an oxygen
atmosphere of 10
−6
mbar at 600 K.
Adding one more monolayer of ruthenium on the
TiO
2
(110), the RuO
2
islands grow two-dimensionally and
coalesce partially into larger islands with a flat top layer (cf.
Figure 1d). No steps or islands are observed on most of the
merged RuO
2
islands. Further deposition of ruthenium (cf.
Figures 1e,f) leads to the development of large merged islands
which may be considered as terraces of RuO
2
. Deposition
beyond 6 ML of RuO
2
(cf. Figures 1g,h) the surface clearly
exposes terraces of RuO
2
; the TiO
2
(110) is now fully covered
by RuO
2
.
Further growth of the RuO
2
film proceeds via a step flow
mode (cf. Figure 2); i.e., ruthenium atoms are exclusively
incorporated into the step edges of RuO
2
, while the terrace
width does not vary with the deposition time of ruthenium.
Most notably, the step edges are not straight but rather
roughened with long finger-like structures.
The initial growth and nucleation of RuO
2
on TiO
2
(110) is
studied in more detail with STM, as summarized in Figure 3.
The deposition of 1/4 ML of RuO
2
leads to islands which
decorate the step edges of TiO
2
(110) while fewer islands are
located on the terraces of TiO
2
(110) (cf. Figure 3a). The
height distribution of these islands is indicated in Figure 3b.
Most of the RuO
2
islands are 3 ML thick with a narrow height
distribution and none of the islands shows a thickness of 1 ML.
Some islands overgrow the step edges of the TiO
2
(110) with a
flat top; the heights of these islands are accounted for by half-
order monolayer thickness in the distribution histograms.
Upon further deposition of ruthenium the islands grow
laterally in size (cf. Figures 3c,e,g), while keeping the thickness
in the range of 3−4 ML (cf. Figures 3d,f,h). Before the islands
coalesce into larger entities the top layers of the square-shaped
islands are always flat, without any indication of steps or
Figure 2. STM images (500 nm × 500 nm) of RuO
2
films grown for different times: (a) 40, (b) 80, and (c) 128 min. The step flow growth mode is
clearly discernible.
Figure 3. STM images for the initial growth of RuO
2
on TiO
2
(110) in the form of nanoislands for varying deposition time of (a) 1 min, (c) 4 min,
(e) starting from (a) and depositing additional 7 min of Ru: (1 + 7) min, and (g) 12 min. The deposition rate of ruthenium is approximately 1 ML
per 4 min. From these STM images the distribution of island heights is extracted and shown as histograms.
The Journal of Physical Chemistry C Article
DOI: 10.1021/jp5121405
J. Phys. Chem. C 2015, 119, 2692−2702
2694
剩余10页未读,继续阅读
资源评论
weixin_38739164
- 粉丝: 8
- 资源: 951
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功