Chapter 4 Effects of Nitrogen-Doping on the Microstructure, Bonding and
4.4 Summary
Vertically aligned CNx NTs have been synthesized using microwave plasma
enhanced chemical vapor deposition with different nitrogen flow rate. The
microstructure, bonding and electrochemistry properties have been investigated
utilizing Raman, XPS, CV and AC impedance measurements to establish
correlations among the N content, electronic structure, microstructure and EC
activities. The results show that nitrogen incorporation provides a simple pathway to
engineering/modifying the microstructure and electronic bonding structure, which is
crucial to the ET kinetics of the CNx NT electrode. In-situ N doping of the CNTs at
an optimal 3.5 at. % N is effective in converting the nanotubes into bamboo-type
morphology and promoting substitutional graphite-like defect structure, which is
favorable for fast ET in electrochemistry. The surface defects thus created further
enhance anchoring of Pt atom in the subsequent selective EC deposition to produce
ultrahigh density Pt NPs to form nanocomposite, which is ideal for many
highly desirable in many EC based applications.
Figure 4.1 Cross-sectional SEM images of the vertically aligned CNTs synthesized
at different flow rate of N2: (a) 0, (b) 80, and (c) 120 sccm.
( ( a a ) )
(c ( c) )
(b ( b) )
Figure 4.2 (a) Comparison of the peak intensities and the full width at half
maximum (FWHM) of the first-order Raman spectra for the vertically aligned CNx
NTs prepared with different N2 flow rate during growth. (b) D-band position as a
function of N2 flow rate. (c) ID/IG as a function of N2 flow rate.
Figure 4.3 (a) The C1s XPS spectra of the vertically aligned CNTs prepared with
various N2 flow rate during growth. The inset is the FWHM of the C-N bonding
component as a function of N2 flow rate. (b) The N1s XPS spectra of the vertically
aligned CNTs prepared with different N2 flow rate. (c) The IP and IG as a function of
the N2 flow rate for the N1s peak. (d) The N-doping concentration as a function of
Figure 4.4 (a) Cyclic voltammetry of the vertically well-aligned CNTs modified with
different N-doping level in 1 M KCl and 5 mM K4Fe(CN)6. (b) The ferricyanide
peak current versus the scan rate (v)1/2 plot for CNTs using various flow rate of N2
for both anode and cathode. (★: 0 sccm, ■: 40 sccm, ●: 80 sccm, ◆: 120 sccm, ▲:
Figure 4.5 AC impedance analysis of the vertically aligned CNTs with different
N-doping level in 1 M KCl and 5 mM K4Fe(CN)6. The inset shows the internal
resistance as a function of the N2 flow rate.
Figure 4.6 TEM images of the Pt NP-CNx NT hybrid nanostructures synthesized at
different flow rate of N2: a) 0, b) 40, and c) 120 sccm.
( ( a a ) ) ( ( b b ) )
( ( c c ) )
10 nm 5 nm
5 nm
0.0 0.2 0.4 0.6 0.8 1.0 0
10 20 30
Cur ren t D e ns ity ( mA /c m
2)
Potential(V vs. Ag/AgCl)
0 sccm 40 sccm 120 sccm
Figure 4.7 Typical CV curve of the arryed Pt NP-CNx NT nanocomposites with
different N2 flow rate at a scan rate of 50 mV/s in 1 M CH3OH + 1 M H2SO4
solution.
Chapter 5
Nanostructured ZnO Nanorod@Cu Nanoparticle as Catalyst for Microreformers
5.1 Introduction
The use of hydrogen for energy generation has attracted significant attention in the
recent years as a clean, sustainable and transportable fuel alternative, and has
consequently sparked a rapid global development of hydrogen fuel cells for electric
power generation.[1] Catalytic reformation of hydrocarbons, with careful attention
to avoid storage and safety issues,[75,76] is currently the predominant process for
hydrogen generation. One of the leading and most promising techniques for
hydrogen generation is catalytic reformation of methanol.[77,78] Cu/ZnO-based
catalysts are, therefore, of great importance for industrial scale catalytic production
of reformate hydrogen.[13] Owing to their wide commercial relevance,
Cu/ZnO-based catalysts, prepared via several preparation routes, are being
extensively investigated; and substantial improvements in their efficiency of
catalytic activity brought about by addition of suitable promoter/support,
hetero-nanostructures as reforming catalysts is still lacking to date. This inspired us
to design the core-shell nanostructured catalyst, consisting of a ZnO nanorod (NR)
core and an outer shell of Cu nanoparticles (NPs), i.e. NR@NPs, for achieving high
efficiency of catalytic conversion.
In addition, the idea of using microreformers is highly attractive for several
applications such as on-board hydrogen sources for small vehicles and portable fuel
cells (FCs).[79,80] However, two key issues have hindered the realization of
microreformers for catalysis, viz. poor adhesion between the catalyst layer and the
microchannels; and poor utilization of catalyst layer deposited in the form of thick
film.[81] Notwithstanding the several approaches investigated to overcome these
issues, catalyst immobilization and its efficient utilization inside the microchannel
remains a challenge.[82,83] Most of these approaches involve a two-step process,
wherein active catalysts are prepared in the first step, followed by its immobilization
on the surface of the microchannels in the second step. In this communication, we
report a simple and reliable method for integrating in-situ synthesis of catalyst and
its immobilization for microreformer applications.
5.2 Structural Characterization of Cu Nanoparticle/ZnO Nanorod
Nanocomposites
ZnO NR arrays at low temperature subsequently resulted in spontaneous formation
of the cable-like nanostructures. Since the ZnO NR@Cu NP nanocomposites were
in-situ synthesized directly on microreactor, the arrayed ZnO@Cu nanocomposites
were strongly anchored onto the microchannel. This was evidenced by observation
of no material loss after sonication in the water for several hours, which highlighted
the strong mechanical anchorage of nanostructured catalysts on the surface of
microchannel. The interaction between Cu NP and ZnO NR was studied by several
analytical techniques, including electron microscopy (EM), X-ray diffraction (XRD),
X-ray photoelectron spectroscopy (XPS), X-ray absorption spectroscopy (XAS), and
temperature-programmed reduction (TPR). The structure of the microreformer
design based on the ZnO NR@Cu NP nanocomposite is illustrated in Figure 5.1,
which is evidenced by the photographs comparing the microchannels before and
after the deposition of the ZnO NR@Cu NP nanocomposite (see Figure 5.2).
One of the most significant advantages of the core-shell nanocomposites, which
are clearly distinct from the traditional catalysts, is the large surface area they offer
for effective surface contact between the reactants and catalysts. Figure 5.3a
illustrates the cross-sectional SEM image of vertically aligned ZnO NRs grown on
the inner surface of the microchannel. The size of the NRs range from 35 to 50 nm
uneven surface with arrow-marked stacking faults, which is shown in greater detail
in Figure 5.3c. Figure 5.3d displays the typical TEM image of ZnO@Cu hybrid
nanocomposites at Cu decoration concentration of 2 mM. Close attachment of Cu
NPs on the ZnO NR cores can clearly be observed. More detailed TEM images with
EDX elemental mapping of Cu and Zn are shown in Figure 5.4, which further
confirms the attachment of Cu NPs on the ZnO NR cores. Further, a high-resolution
TEM image, shown in Figure 5.3e, yielded the spacing of the｛111｝lattice planes of
the fcc copper crystal to be 0.21 nm. A histogram of the diameter of Cu NPs
determined from the TEM measurement is shown in Figure 5.3f, indicating a
diameter range of 3 to 8 nm with an average of 5 nm. Moreover, a high-resolution
TEM image of the Cu NP/ZnO NR heterostructures reveals that the (111) plane of
the Cu NP is immobilized on the (002) plane of ZnO NR with clear evidence of the
lattice-mismatched region (see Figure 5.5).
Figure 5.6a shows XRD patterns of Cu deposited on ZnO NRs prepared with
different Cu decoration concentrations from 1 to 3 mM. A large range XRD patterns
can be as reference in the Figure 5.7. No obvious Cu peak could be detected below 1
mM decoration concentration. Figure 5.6a also shows that as the decoration
concentration gradually increases, Cu(111) peak at 43.5 degree becomes more
size of Cu particles in the nanocomposite samples. Furthermore, the FWHM
increases as the decoration concentration decreases, indicating formation of finer
particles at lower decoration concentration, which is also confirmed in the results of
larger Cu surface area and higher dispersion determined via N2O decomposition
method (see Table 5.1). In addition, in Figure 5.6a, a shift in the position of this peak
to higher 2θ values as compared to the metallic Cu peak is also observed with
decreasing decoration concentration. This can be attributed to the existence of
defects at the interface between Cu NP and ZnO NR or partial dissolution of Cu into
the ZnO lattice. The defects, either microstrain or structural disorder, can originate at
the interface due to the lattice mismatch between Cu and ZnO, but can be quickly
overwhelmed by the strong metallic Cu(111) peak from larger sized particles
deposited at higher decoration concentrations, as can be observed from Figure 5.6a.
A detailed inspection of the electronic states of the surface metal species was
carried out through XPS analysis. Figure 5.6b depicts the Cu 2p core level XPS of
NR-Cu composite nanostructures prepared at different decoration concentrations. In
order to differentiate between the oxidation states of Cu, the main peak of the Cu
2p3/2 core level spectra was fitted with two components at 932.8 and 933.7 eV
corresponding to Cu0/Cu+ and Cu2+ species, respectively.[86] It can be seen that a
binding energies above 940eV further indicates a lower Cu oxidation state.[87]
When the decoration concentration was decreased, the position of the Cu main peak
shifted slightly toward higher binding energy. This is strongly related to the
modification of the electron density on smaller Cu species at lower decoration
concentrations.[86] Moreover, it appears that the fraction of atomic Cu on the
surface, as estimated from the ratio of Cu/(Cu + Zn + O) peaks tends to increase
with increasing decoration concentration. In contrast, the fraction of atomic Cu on
the surface is lower than 8 at. % while the decoration concentration is decreased to 1
mM, which is consistent with the XRD results in Figure 5.6a. Meanwhile, similar
trend can be obtained for the surface ratios of Cu/Zn and ZnO/Cu through XPS
measurements (see Table 5.1). However, it is rather unfortunate that Cu0 state cannot
be distinguished from Cu+ state by XPS analysis due to their spectral overlap. X-ray
absorption spectroscopy (XAS) measurements were hence used to resolve this issue.
The X-ray absorption near edge structure (XANES) spectra are associated with the
excitation of a core electron to bound and quasi-bound states. To determine the
valence of copper in the arrayed ZnO NR@Cu NP nanocomposites, the Cu K-edge
XANES spectra are compared with those of Cu foil in Figure 5.6c. The XANES
spectra of arrayed nanocomposites exhibit the edge absorption at 8979 eV together
nanocomposites. A more detailed understanding may be gained from the derivative
of the XANES spectra (see Figure 5.8). The intensity of the edge absorption for
arrayed nanocomposites is relatively low compared to that of Cu foil. This could be
related to the smaller (nano-sized) dimensions of the Cu particles.[88] Additionally,
it may be noted that the nanocomposite samples display lower intensity as well as a
shift towards higher energy of the peak at ca. 8990 eV, together with the shoulder
around 8984 eV, as compared to the derivative of Cu foil. These features suggest
some changes in the chemical environment around the copper species in these
nanocomposites. The reason may be attributed to a further distortion of the copper
lattice due to the presence of microstrain at the interface between Cu NPs and ZnO
NRs. Thus, the XAS results seem to agree well with the XRD results.
To investigate the reducibility of NR@Cu NP nanocomposites prepared at
different decoration concentrations, TPR measurements were carried out in the
temperature range of 140 to 300 oC, as shown in Figure 5.6d. It is known that the
reduction of bulk CuO is indicated by a single reduction peak at a considerably
higher temperature of 320 oC.[89] Clearly, the Cu NP-immobilized ZnO NR
nanoarchitectures with 2 mM decoration concentration exhibits a significant shift in
the position of the reduction peak to lower temperature in addition to the reduction
NR@Cu NP catalysts can be emphasized by the lower reduction temperature than
the commercial catalysts (Cu-ZnO-Al2O3; MDC-3: Sud-Chemie) as shown in the
same figure. The reason can be attributed to the enhanced dispersion of fine Cu NPs
and strong metal-support interaction (SMSI) effect under NR@NP nanosystems,
which facilitates the reduction of the supported Cu species. However, at decoration
concentrations of higher than 2 mM, the increase in the size of Cu particles plays a
predominant role in the structural change of ZnO@Cu nanocomposites, which also
weakens the interaction between Cu and ZnO and consequently undermines the
catalytic activity.
5.3 Test of Methanol Reforming Reaction
Methanol conversion and hydrogen production rate for the arrayed ZnO@Cu
nanocomposites and the commercial catalysts are shown in Figure 5.9a and 5.9b,
respectively. Each data point represents a 12-hour experimental run with very small
deviation observed with repeated runs, which is indicated by the small error bars in
the figures. The methanol conversion rate over the arrayed ZnO@Cu
nanocomposites is as high as 93 % with a hydrogen production rate of 183 mmol
gcat-1 h-1 at 250 oC. Both these rates are significantly higher than those obtained with
the commercial catalysts. Small amounts of CO were detected by CO detector in the
than ca. 1000 ppm formed by the commercial catalysts. The present results thus
provide a promising alternative to the commercial catalyst for use in catalytic
generation of high purity hydrogen without necessitating additional processes to
remove CO down stream. For a better comparison of the catalytic activities of
NR@NP nanoarchitectures and commercial catalysts, kinetic parameters were
evaluated from the conversion data.[90] The kinetic constant at a given temperature
can be calculated from the equation:
κ = 1/τ [εx + (1+ε)ln (1-x)],
where κ, τ, x, and ε are the kinetic constant, contact time, fractional conversion,
and fractional expansion of the system, respectively. As shown in Figure 5.9c, the
kinetic constant data manifest higher activity of the nanostructured NR@NP arrays
than the commercial catalysts. This can be attributed to the greater external surface
area, the existence of SMSI effect, the presence of microstrain at the interface, and
the modification of electronic structures in the nanocomposites. These effects are
consistent with the TEM, XRD, XPS, XANES, and TPR data. Based on the reaction
temperature dependence of the kinetic constant, the apparent activation energy was
calculated from the slope of the Arrhenius plot (Figure 5.9d). It is evident that the
reaction activation energy for the nanostructured NR@NP arrays is noticeably lower
nanostructured catalysts than on the commercial catalysts. Last but not least, the
advantage of the nanostructured NR@NP arrays catalyst is further exemplified by
the higher stability (11.5 % reduction after 36 hours operation) in methanol
reforming reaction, which is significantly superior to the 33.8 % reduction for the
commercial catalysts (see Figure 5.10).
5.4 Summary
In summary, we have successfully demonstrated an easy-to-fabricate route to
prepare arrayed ZnO NR@Cu NP heterostructures on the inner surface of
microchannels via direct in-situ synthesis for microreformer applications. The
superb catalytic performance and stability of the Cu NP-decorated ZnO NR
nanostructures can be attributed to the larger surface area and enhanced dispersion
of fine Cu NPs, formation of microstrain, the modification of electronic structure of
Cu species, and the existence of SMSI effect. These results present new
opportunities in the development of highly active and selective NR@NP
nanoarchitectures for a wide range of different catalytic reaction systems.
Microchannel Reactor
Gas In
Gas
Out Arrayed Arrayed ZnOZnONRsNRs Cu NPs
Out Arrayed Arrayed ZnOZnONRsNRs Cu NPs
Figure 5.1 Schematic diagram of the novel catalyst ZnO NR@Cu NP arrays grown
on the inner surface of microchannel reactor.
Figure 5.2 Photographs of the microchannels a) before and b) after homogeneously
depositing the ZnO NR@Cu NP nanocomposites.
20 nm
Figure 5.3 a) Cross-sectional SEM image of vertically well-aligned ZnO NRs. b)
Typical TEM image of a single ZnO NR showing the presence of stacking faults as
marked with arrows. c) HRTEM image of the ZnO NR. d) Typical TEM image of
the ZnO NR@Cu NP nanocomposites. e) HRTEM image of Cu NPs on the surface
of one single ZnO NR. f) The size histogram of Cu NPs analyzed from the HRTEM
20 nm
a)
20 nm
b)
20 nm
c)
20 nm
a)
20 nm 20 nm
a)
20 nm
b)
20 nm 20 nm
b)
20 nm
c)
20 nm 20 nm
c)
Figure 5.4 a) TEM image and corresponding EDX elemental mapping of b) Cu and c)
Zn.
ZnO NR 2 nm
Cu NP
ZnO(002) Cu(111)
mismatch
ZnO NR 2 nm
Cu NP
ZnO(002) Cu(111)
mismatch
Figure 5.5 HRTEM image of ZnO NR@Cu NP heterostructures.
41 42 43 44 45 46
8960 8980 9000 9020 9040
2 mM
925 930 935 940 945 950
Cu 2p3/2
8960 8980 9000 9020 9040
2 mM
925 930 935 940 945 950
Cu 2p3/2
Figure 5.6 a) XRD patterns of Cu NPs on the surface of ZnO NRs prepared with
different decoration concentrations from 1 to 3 mM. b) Cu 2p XPS core level spectra
of arrayed ZnO NR@Cu NP nanocomposites prepared with different Cu decoration
concentrations from 0.5 to 3 mM. c) Cu K-edge XANES spectra of arrayed ZnO
NR@Cu NP nanocomposites and bulk reference sample Cu foil. d) Comparative
TPR profiles of ZnO NR@Cu NP nanocomposites and commercial catalysts
(MDC-3: Sud-Chemie).
30 35 40 45 50 55 60 65
In te n s it y ( a .u .)
2 θ (degree)
3 mM
2 mM
1 mM
■
▲ ▲
▲▲
▲
30 35 40 45 50 55 60 65
In te n s it y ( a .u .)
2 θ (degree)
3 mM
2 mM
1 mM
■
▲ ▲
▲▲
▲
Figure 5.7 Large range of XRD patterns of Cu NPs on the surface of ZnO NRs
prepared with different decoration concentrations from 1 to 3 mM. Peaks marked
with ▲ are due to ZnO, the other marked with ■ is due to Cu.
4.4
8976 8980 8984 8988 8992
Cu K-edge
Table 5.1 Microstructure properties of Cu NPs on the surface of ZnO NRs prepared
with different decoration concentrations.
Figure 5.8 First derivative of the Cu K-edge XANES spectra of arrayed ZnO
NR@Cu NP nanocomposites and bulk reference sample Cu foil.
140 160 180 200 220 240 260 280 300
140 160 180 200 220 240 260 280 300 0
H2 production rate (mmol gcat -1 h-1 )
Temperature (oC)
140 160 180 200 220 240 260 280 0
140 160 180 200 220 240 260 280 300 0
140 160 180 200 220 240 260 280 300 0
H2 production rate (mmol gcat -1 h-1 )
Temperature (oC)
140 160 180 200 220 240 260 280 0
Figure 5.9 a) Methanol reforming reaction profiles for 2 mM arrayed Cu NPs/ZnO
NRs nanocomposites (□) and commercial catalysts (●). b) Hydrogen production rate
as a function of reaction temperature for 2 mM arrayed Cu NPs/ZnO NRs
nanocomposites (□) and commercial catalysts (●). c) Kinetic constants as a function
of reaction temperature for 2 mM arrayed Cu NPs/ZnO NRs nanocomposites (□)
and commercial catalysts (●). d) Arrhenius plots for methanol reforming reaction for
2 mM arrayed Cu NPs/ZnO NRs nanocomposites (□) and commercial catalysts (●).
0 4 8 12 16 20 24 28 32 36 0
20 40 60 80 100
M e tha nol c o nv e rs io n ( % )
Time on Stream (hr)
11.5 ％ loss 33.8 ％ loss
0 4 8 12 16 20 24 28 32 36 0
20 40 60 80 100
M e tha nol c o nv e rs io n ( % )
Time on Stream (hr)
11.5 ％ loss 33.8 ％ loss
Figure 5.10 Stability tests in methanol reforming reaction over 2 mM arrayed Cu
NPs/ZnO NRs nanocomposites (□) and commercial catalysts (○). Reaction
conditions: H2O/O2/MeOH = 1/0.125/1, Temperature = 250 oC, W/F = 21 kgcat s
mol-1methanol.
Chapter 6
Microwave-Activated CuO Nanotip/ZnO Nanorod Nanoarchitectures for Efficient Hydrogen Production
6.1 Introduction
In the field of development of advanced energy-related devices, nanostructured
materials have attracted great interest in recent years due to their intriguing
physical and chemical properties that are significantly different from their bulk
counterparts.[91-93] Nanostructured catalysts are expected to be the key
components in the advancement of future energy technologies. Hence, new
strategies for the synthesis of high-performance nanomaterials are widely
pursued.[94] In particular, Cu/ZnO-based catalysts have been extensively
investigated because of their importance in several industrial applications, such
as methanol synthesis, CO removal, and hydrogen generation.[17,35,78,95-99]
Nanostructured catalysts consisting of microporous structures are attractive for
flow-type gas-solid reactions due to their effective diffusion of reactants and heat.
In our previous work, strong metal-support interaction (SMSI) effect in arrayed
These findings led us to design and develop a much effective technique to tailor
the Cu/ZnO interface.
In recent years, microwave (MW) irradiation has become an effective tool in
synthetic organic chemistry, yielding dramatic increases in reaction rates and
yields.[101-104] Moreover, since the magnitude of MW absorption is related to
dipole oscillation in a material, MW annealing can provide selective heating and
lead to a new route for material processing. Recently, application of MW as selective
annealing technique has attracted tremendous attention for improving the efficiency
of polymer organic photovoltaic devices.[105] In this communication, we report
MW treatment of nanoarchetectures of CuO nanotip (NT) on ZnO nanorod (NR)
framework as catalyst precursors for methanol reforming reaction (MRR).
6.2 Structural Characterization of CuO Nanotip/ZnO nanorod Nanocomposites
Our initial attempts employing a surfactant-free approach
(ammonia-evaporation-induced synthetic method) for synthesis of CuO
nanostructures[106] resulted in formation of radiating nanosheets as shown in
Figure 6.1a. The high-resolution transmission electron microscopy (HRTEM)
image in Figure 6.2 provides specific structural information about an individual
different morphologies of CuO were observed, ranging from nanosheet to NT
(Figure 6.1b). Figure 6.3 presents the TEM and HRTEM images of CuO NT/ZnO
NR catalyst precusors synthesized in this work, showing that NTs not only
adhere to but also directly conjoin with the NR. While detailed understanding of
adhere to but also directly conjoin with the NR. While detailed understanding of