Nghiên cứu chế tạo sợi nano fe2o3 và znfe2o4 lai graphene khử từ ôxit graphene (RGO) bằng phương pháp phun tĩnh điện và ứng dụng cho cảm biến khí h2s tt tiếng anh

MINISTRY OF EDUCATION AND TRAINING
HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY

Nguyen Van Hoang

ELECTROSPINNING OF α-Fe2O3 AND ZnFe2O4
NANOFIBERS LOADED WITH REDUCED GRAPHENE
OXIDE (RGO) FOR H2S GAS SENSING APPLICATION

Major: Materials Science
Code: 9440122

ABSTRACT OF DOCTORAL DISSERTATION
OF MATERIALS SCIENCE

Hanoi - 2020


The Dissertation was completed at:
Hanoi University of Science and Technology

Supervisor: Prof. PhD. Nguyen Van Hieu

Reviewer 1: Prof. PhD. Luu Tuan Tai
Reviewer 2: Prof. PhD. Pham Thanh Huy
Reviewer 3: Prof. PhD. Vu Dinh Lam

The dissertation will be defended at the University Council of
Doctoral dissertation held at Hanoi University of Science and
Technology.
At ……….., date………..month…......year………..

The dissertation can be found at the libraries:
1. Ta Quang Buu, Hanoi University of Science and Technology
2. Vietnam National Library


INTRODUCTION

1. Background of the thesis
Recently, 1D nanostructures including nanowires (NWs),
nanorods (NRs), nanotubes (NTs), and nanofibers (NFs) have
attracted much attention for a wide application including optical
catalysis, electronic devices, optoelectronic devices, storage devices,
and gas sensors due to their high surface-to-volume ratio. Especially,
NFs are widely used in many fields such as catalysis, sensor, and
energy storage because of their outstanding properties like their large
surface area-to-volume ratio and flexible surface functionalities.
There are several approaches for NFs fabrication, for example,
drawing, template, phase separation, self-assembly, and
electrospinning, among which electrospinning is a simple, costeffective and versatile method for NFs production.
Regarding gas sensing applications, semiconductor metal oxide
(SMO) NFs sensors have a lot of promise due to their advantages of
SMO materials like low cost, simple fabrication, and high
compatibility with microelectronic processing. Furthermore, NFs
consist of many nanograins, therefore, grain boundaries are large,
surface-to-volume ratio is very high, and gases easily diffuse along
grain boundaries. As a result, an exceptionally high response was
observed in in SMO NFs gas sensors by electrospinning. Among
various SMO NFs prepared by electrospinning, α-Fe2O3 has become
a potential gas sensing material because of its low cost and thermal

stability and ability to detect many gases such as NO2, NH3, H2S, H2,
and CO. Besides, zinc ferrite ZnFe2O4 (ZFO), a Fe2O3-based ternary
spinel compounds, has been a promising material for detecting gases
thanks to its good chemical and thermal stability, low toxicity, high
specific surface area and excellent selectivity. Otherwise, H2S is a
colorless, corrosive, inflammable and extremely toxic gas which can
be rapidly absorbed by human lungs and easily causes diseases in
respiratory and nervous system, even deaths. However, until now,
very few studies on H2S gas sensing properties of α-Fe2O3 and ZFO
NFs, especially effects of parameters of fabrication process (i.e.
solution composition, heat treatment, and electrospun time) on
1


morphology, structure and H2S gas sensing properties of the sensors
have been carried out although there have been some reports about
H2S gas sensitivity of the sensor of other nanostructures of α-Fe2O3
or ZFO sensors (e.g. nanochains, porous nanospheres, and porous
nanosheets (NSs)).
Furthermore, RGO, a GP reduced from GO produced from
graphite by Hummer method, has recently received world-wide
attention owing to its exceptional physicochemical properties. The
combination between SMO NFs and RGO to enhance gas sensing
performance through the formation of heterojunction was mentioned
in many works. However, up to present, there have been no reports
on incorporation of RGO in α-Fe2O3 and ZFO NFs for enhanced H2S
gas sensing performance.
Therefore, the thesis titled "Electrospinning of α-Fe2O3 and
ZnFe2O4 nanofibers loaded with reduced graphene oxide (RGO) for
H2S gas sensing application” was carried out to answer the concerns

mentioned above.
2. The study objective
The study objective of the thesis are listed as follows:
- To successfully fabricate on-chip sensors based on α-Fe2O3, ZFO
NFs and their loading with RGO by on-chip electrospinning.
- To explore the effect of parameters (i.e. solution composition,
heat treatment, electrospun time, and RGO concentration) of
fabrication process on the NF morphology, structure and H2S gas
sensing properties.
- To clarify H2S gas sensing mechanism of the sensors of α-Fe2O3,
ZFO NFs and their incorporation with RGO.
3. Research scope and content
The thesis uses α-Fe2O3, ZFO NFs and their loading with RGO, as
well as harmful gas H2S as object studies.
The study focuses on the following contents:
- To optimize some process parameters (i.e. solution composition,
heat treatment, electrospun time, and RGO concentration) for onchip sensor fabrication of α-Fe2O3, ZFO NFs and their loading
with RGO via electrospinning method.

2


-

To characterize the NFs and to analyze the relationship between
their morphology and microstructure of NFs with fabrication
process parameters.
- To examine H2S gas-sensing properties of the NFs sensors for
clarifying the relationship among morphology, microstructure
with gas-sensing properties of the NFs sensors.

- To understand the H2S gas-sensing mechanisms of α-Fe2O3, ZFO
NFs and their loading with RGO.
4. Research Methodology
To achieve the objectives, the thesis research was conducted by
experimental methods, namely:
- The on-chip electrospinning method was employed for the
fabrication of α-Fe2O3, ZFO NFs and their loading with RGO.
- Morphology and structure of the NFs were characterized by
TGA, RAMAN, FE-SEM, TEM, HR-TEM, SAED, EDX, and
XRD.
- The gas-sensing properties of the NFs were measured by a
home-made system using flow-through technique.
5. Practical and scientific significance of the thesis
The scientific relevance: The thesis results elaborated the
relationship among processing parameters, microstructure, and gassensing properties of α-Fe2O3, ZFO NFs and their loading with RGO.
In addition, the thesis also clarified H2S gas-sensing mechanisms of
α-Fe2O3, ZFO NFs and their loading with RGO. Furthermore, the
research results have been reviewed by domestic and foreign
scientists, and published in prestigious journals such as Journal of
Hazardous Materials and Sensors and Actuator B, which shows
scientific significance of the dissertation.
The practical relevance: This dissertation focused on the
development of the effective sub-pp H2S gas sensor of α-Fe2O3, ZFO
NFs and their loading with RGO by on-chip electrospinning method.
The optimized results provide a premise to develop the sensors for
environmental monitoring, occupational health, petrochemical plant,
which showed significantly practical relevance of the dissertation.
6. The original contributions of the dissertation
3

Currently, almost NFs sensors are prepared by two-step process:
synthesis of sensing materials and then fabrication of the sensors,
which is not cost-effective for large-scale production and difficult for
reproducible sensor fabrication. In this thesis, the on-chip NFs
sensors were successfully synthesized by electrospinning.
The effect of morphology, structure and composition which were
changed in fabrication process by varying solution concentration,
electrospun time, heat treatment conditions, and RGO concentration
on H2S gas sensing properties of the sensor of α-Fe2O3, ZFO NFs
and their incorporation with RGO was systematically investigated.
In addition, H2S gas sensing mechanisms of α-Fe2O3, ZFO NFs
and their incorporation with RGO, especially when annealing
temperature was changed, were also discussed in detail.
The main research results of the thesis were published in 02 ISI
articles, 01 location article, and 02 proceedings of conference.
7. The structure of the thesis
This thesis is interpolated from the articles by the author. Apart
from the introductions, conclusions and recommendations, there are
four main chapters and a list of references and publications in this
thesis.
Chapter 1: Overview on SMO NFs and their loading with RGO
for gas-sensing application
Chapter 2: Experimental approach
Chapter 3: α-Fe2O3 NFs and their loading with RGO for H2S gas
sensing application
Chapter 4: ZFO NFs and their loading with RGO for H2S gas
sensing application

4

CHAPTER 1. OVERVIEW ON SMO NFs AND THEIR
LOADING WITH RGO FOR GAS-SENSING APPLICATION
In this chapter, an overview on electrospinning, one of the most
simple, cost-effective and flexible methods for NFs fabrication with
such various kinds of materials as polymers, SMO, and composites,
was introduced. NFs formation made use of electrostatic forces to
stretch a viscoelastic solution. A high voltage was applied to a
solution droplet suspended at a tip of a syringe needle. When the
electric field reached a critical value, a charged jet of the solution
was ejected and stretched to form a continuous and thin fiber from
the tip of needle to a collector. Subsequently, the as-spun fibers were
calcined to decompose polymer and crystallite to form SMO NFs.
Some parameters in fabrication process, which affected
morphologies and microstructures of NFs, were also mentioned. NFs
morphologies and microstructures depended on such factors as
electrospinning parameters, solution, and environmental conditions.
Furthermore, collectors and needles also had a strong influence on
morphologies and microstructures of obtained NFs. The most
commonly used collector was the rotary drum collector which was
suitable for mass production of aligned NFs. In addition, the
conditions of heat treatment process greatly affected NFs
morphologies and microstructures. Any changes in the annealing
temperature, annealing time, or heating rate could lead to changes in
NFs morphologies and microstructures, resulting in the varied NFs
properties.
SMO NFs have been widely used in gas sensing application.
Many works showed that the NFs sensor have high response and fast
response-recovery time due to their high porosity and large specific

surface area structures. Especially, there have been many studies on
H2S gas sensing properties of SMO NFs and composite NFs. The
results showed that NFs structure had higher response and faster
response time than other nanostructures. In addition, the response
and selectivity of the composites sensors were enhanced compared to
those of binary SMO sensors. However, there are not many
researches on the sensors based on NFs of α-Fe2O3 or ZFO to
5


different gases in general and to H2S gas in particular, especially at
sub-ppm concentrations.
This chapter also reviewed on RGO and its application in gas
sensing field. In general, RGO was widely used in gas sensors thanks
to its incompletely reduced functional groups, and many dangling
bonds or defects that created favorable positions for gases to absorb.
However, RGO sensors also had some limitations such as low
response, drift resistance, irreversibility, long response and recovery
time. Therefore, combining RGO with other materials like noble
metals or SMO helped to solve these problems. There were two
trends to combine RGO and SMO to enhance the gas sensing
properties. On the one hand, SMO particles were attached to the
surface of RGO NSs, resulting in SMO-loaded RGO sensors. The
sensors were conducted through continuously connected RGO NSs.
The sensors showed gas sensing characteristics of a p-type
semiconductor of RGO while SMO enhanced the response and
response-recovery time. However, the SMO-loaded RGO sensors
failed to solve some inherent limitations of RGO sensors like long
response time, irreversibility and low response. In particular, the
sensor response to reducing gas was very low. On the other hand, the

RGO-loaded SMO sensor had much higher response to reducing gas
than the SMO-loaded RGO sensor thanks to their inherited gas
sensing characteristics of SMO. The main conducting path of the
sensor went through SMO. RGO concentration was usually below 5
wt% and RGO NSs were dispersed and disconnected in composites.
The sensors behaved gas sensing characteristics of SMO. The RGOloaded SMO sensors had higher response than the pure SMO sensors
due to the formation of heterojunction between RGO and SMO.
Sensors based on SMO NFs loaded with RGO combined
advantages of RGO-loaded SMO sensors and NFs sensors. SMO
NFs loaded with RGO were composed of SMO NFs and RGO NSs,
in which RGO were distributed randomly and discontinuously
among SMO nanograins or on NFs surface. The RGO-loaded SMO
NFs structure had high porosity and large specific surface area;
therefore, the sensor of this structure often had excellent sensitivity
and fast response time. Many works reported that the RGO loaded6


SMO NFs sensors had high response to both oxidizing and reducing
gases. The sensors also had good selectivity and fast response time.
RGO enhanced the sensor response by forming heterojunctions
between RGO and SMO. Besides, RGO had many functional groups,
dangling bonds and defects that increased gas absorption, thereby
increasing the sensor response. However, until now, H2S gas sensing
properties of the RGO-loaded SMO NFs sensors in general and on
Fe2O3 NFs loaded with RGO and ZFO NFs loaded with RGO in
particular have not been investigated, which were studied on the
flowing chapters.
Finally, gas sensing mechanisms of NFs and RGO-loaded SMO
NFs were also discussed in this chapter, which was related to the
formation depletion surface on NFs surfaces and potential barriers at

homojunctions among nanograins and heterojunctions between SMO
and RGO. Moreover, the sensor gas sensing mechanisms to H2S was
elaborately mentioned.
CHAPTER 2. EXPERIMENTAL APPROACH
This chapter presented the fabrication process of the sensing
materials. Briefly, α-Fe2O3 and ZFO NFs were synthesized on chip
by electrospinning. Precursor solution content, electrospun time and
heat treatment conditions were changed to obtain the on-chip NFs
sensors with different morphologies, structures and densities. RGO
was reduced by L-ascorbic acid from graphene oxide (GO)
synthesized from graphite power by Hummers method. A series of
the sensors of 0, 0.5, 1.0, and 1.5 wt% RGO-loaded α-Fe2O3 and
ZFO NFs was also fabricated on chip by electrospinning. The onchip electrospun sensors were calcined at different temperatures to
form RGO-loaded α-Fe2O3 and ZFO NFs.
Then, some characterization methods like TGA, RAMAN,
FESEM, TEM, HRTEM, SAED, EDX, and XRD were employed to
analyze the synthesized NFs. Finally, gas sensing properties of the
synthesized sensors were measured by flow-through technique which
7


used a home-made system of a test chamber with controlled working
temperature, a series of mass flow controllers to obtained a desired
gas concentrations, and Keithley 2602 controlled by a software
program to record the electrical-resistance response of the test
sensors under various concentrations and operating temperatures.
CHAPTER 3. α-Fe2O3 NFs AND THEIR LOADING WITH
RGO FOR H2S GAS SENSING APPLICATION
3.1.
Introduction

Hematite α-Fe2O3, an n-type semiconductor with the band gap Eg
of 2.1 eV and rhombohedral crystal structure, has been widely used
in gas sensors due to its high stability, low cost, non-toxicity,
environmental friendliness and multiple functions. The H2S gas
sensing properties of α-Fe2O3 with different nanostructures have
been published in many works. However, H2S gas sensitivity at subppm concentrations of α-Fe2O3 NFs sensors has not been
investigated. Furthermore, despite some studies on effects of
processing parameters on morphology, structure and gas sensitivity
properties of the obtained NFs, similar studies on H2S gas sensing
properties of α-Fe2O3 NFs have not been carried out.
In addition, the RGO-loaded α-Fe2O3 NFs sensors have also
attracted much attention. The studies proved that RGO enhanced gas
sensitivity of the RGO-loaded α-Fe2O3 NFs sensor. However, H2S
gas sensitivity, especially at low sub-ppm concentrations, of the
RGO-loaded α-Fe2O3 NFs sensors has not been reported.
In this chapter, α-Fe2O3 NFs were synthesized by electrospinning
method. The precursor solution composition (i.e. polymer
concentration and salt concentration) and technological parameters
(i.e. electrospinning time and annealing temperature) were altered to
obtain the different morphologies and structures of α-Fe2O3 NFs,
leading to the effects on H2S gas sensing performance at sub-ppm
concentration of α-Fe2O3 NFs sensors. Besides, RGO influence on
morphologies, structures and H2S gas sensing properties of the RGOloaded α-Fe2O3 NFs sensors was also discussed in detail.
8


3.2.
H2S gas sensors based on α-Fe2O3 NFs
3.2.1. Morphologies and structures of α-Fe2O3 NFs
The XRD results at different annealing temperatures confirmed

rhombohedral structure of α-Fe2O3 NFs (JCPDS 33–0664). More
diffraction peaks appeared and became sharper with the increased
annealing temperature, indicating an increase in NFs crystallinity and
nanograin size.
The precursor solution content strongly influenced the
(d)

(a)

3 µm

3 µm

150 nm

150 nm
(b)

(e)
3 µm

3 µm

150 nm

150 nm

(c)

(f)

3 µm

3 µm

150 nm

150 nm

Figure 3.7. FESEM images of as-spun fibers (a) and α-Fe2O3
NFs prepared at different annealing temperatures: 400 (b), 500
(c), 600 (d), 700 (e), and 800°C (f). Inset figures are lowmagnification images.
morphology and structure of the synthesized α-Fe2O3 NFs. With low
concentration of 7 wt % PVA, the NFs comprised a network of small
beads interconnected by thin NFs. The higher PVA concentration
was, the bigger fiber diameter became due to an increase in
viscoelastic force which counteracted the electric field force. The
fibers failed to form when the PVA concentration was too low or too
high. The α-Fe2O3 NFs had the belt-like morphology at 2 wt% ferric
salt and become quite round and uniform, and smooth surfaces with
9


Air

(b)

R (M)

(a)

o

H2S@ 1ppm & 300 C

5

o

H2S@ 1ppm & 350 C
o
o

R(M)

H2S@ 1ppm & 450 C

1ppm H2S

(c)

20
10

1

0

Resp. (Ra/Rg)

0

H2S@ 1ppm & 400 C

(d) 1000
Recovery time

100



Response time

(s)

o

H2S@ 1ppm & 250 C

resp-recov

10

100

Time (s)

1000

250

300
350
400
450
Operating Temp. (oC)

Figure 3.9. Sensing transients of α-Fe2O3 NF sensors to 1 ppm
H2S at various operating temperatures (a), sensor resistances (b),
sensor response (c), response time and recovery time (d) as a
function of operating temperatures.
4 wt% ferric salt. With further increased ferric salt of 8 wt%, NFs
diameters increased and the NFs surfaces became rough.
The FE-SEM images of on-chip α-Fe2O3 NFs with electrospun
time from 10 to 120 min were illustrated. When the electrospun time
went up, the number of NFs connecting two electrodes also increased,
especially the number of intersections among ZFO NFs significantly
got bigger.
The morphologies of NFs with different annealing temperatures
were shown in Fig. 3.7. The NFs were 50–100 nm in diameter. The
surface of the NFs became rough because the NFs were made up of
many nanograins. The higher the annealing temperature was, the
rougher the surface of the NFs was because of nanograin growth. At
high annealing temperature of 800°C, NFs had the same shape as a
bamboo due to coalescence and grain growth process.
TEM, HRTEM, and EDX analyses further examined the
morphologies, structures, and compositions of α-Fe2O3 NFs calcined
at 600°C. TEM images showed that the NFs were composed of many
nanograins; however, the NFs structure was quite tight. HRTEM
image and FFT inset image confirmed that the NFs had a good
crystal structure with parallel lattice fringes. The composition of αFe2O3 NFs with the presence of Fe and O elements was indicated in

EDX spectrum results.
10


3.2.2. H2S gas sensing properties of sensors based on αFe2O3 NFs
3.2.2.1.
Effects of operating temperature
The effect of working temperature on the gas sensing
performances of the sensor was shown in Fig. 3.9. The sensor
response also decreased sharply with the increased working
temperature because the gas desorption became stronger than gas
adsorption and the height of the potential barrier at the grain
boundaries decreased with increased working temperature.
Conversely, the recovery time also became too long with the
decreased working temperature because of the reduced reaction rate
and diffusion rate along the grain boundaries. Therefore, to optimize
the sensor response and recovery time, the working temperature of

Figure 3.12. H2S sensing transients of α-Fe2O3 NF sensors with
various annealing temperatures (400−800°C) (a–e) and
different electrospinning time (10−120 min) (f–i). Sensor
response to H2S gas as a function of annealing temperatures (k)
and electrospinning time (l).
11


350°C was selected for further investigating gas sensing properties of
the α-Fe2O3 NFs sensors.
3.2.2.2.
Effects of solution contents

The sensor response decreased with the increased PVA
concentration from 7 to 15 wt% PVA because of increased NFs
diameters. However, the NFs comprised a network of small beads
interconnected by thin fibers with 7 wt% PVA concentration.
Whereas, the NFs prepared from precursor solution with 11 wt%
PVA showed the typical spider-net morphology with many round
and uniform NFs fabricated by electrospinning as reported in many
works. The sensor response to 1 ppm H2S gas was 2 at 2 wt%, and
reached a maximum of 6.2 at 4 wt% g and then went down to 4.9
with 8 wt% ferric salt. Therefore, to optimize the NFs morphology
and gas response, the sensor prepared from precursor solution with
11 wt% PVA and 4 wt% ferric salt was chosen for further study.
3.2.2.3.
Effects of
annealing temperature and
electrospinning time
As shown in Fig. 3.12, the response of α-Fe2O3 NFs sensors
fluctuated with changed annealing temperature, which could possibly
be explained by the change in crystallinity and grain size of NFs with
different annealing temperatures. When the temperature went down
from 600 to 500°C, the response decreased due to the strong
influence of decreased crystallinity caused by decreased annealing
temperature. When the temperature further decreased from 500 to
400°C, the sensor response increased because of the strong effect of
decreased nanograin size. Meanwhile, when the temperature
increased from 600 to 800°C, the sensor response decreased
remarkably because of grain growth.
The densities of the NFs on the microelectrode chip, which could
be controlled by electrospinning time, strongly affected gas-sensing
performance. The gas response showed a bell-shape relation with

electrospun time at working temperature of 350°C and the response
peak was obtained at the electrospun time of 30 min. The NFs sensor
response increased with increased electrospun time due to an
increase in the NFs-NFs junctions between Pt electrodes. Such
junctions improved the sensor sensitivity when the sensors were
exposed to H2S gas. The response decreased with the further
12


increased electrospun time because of the increased thickness of the
sensing layer, resulting in the increased gas diffusion length.
In short, the sensor based on α-Fe2O3 NFs was calcined at 600°C
and electrospun for 30 min with the precursor solution of 11 wt%
PVA and 4 wt% Fe (NO3)3 salt for optimizing among structures,
morphologies and sensor response.
3.2.2.4.
Selectivity and stability
The sensor also had good selectivity to reducing gases like H2
and NH3 but its selectivity to oxidizing gas SO2 was still limited.
The sensor also showed good repeatability, which highlighted
practical applicability of the α-Fe2O3 NFs sensor.
3.3.
H2S gas sensors based on α-Fe2O3 NFs loaded with
RGO
3.3.1. Morphologies and structures of α-Fe2O3 NFs loaded
with RGO
Morphologies of RGO-loaded α-Fe2O3 NFs were not significantly
affected by the changed RGO contents (0–1.5 wt%). RGO could not
be found in FESEM images of RGO-loaded α-Fe2O3 NFs since RGO
(c)

(a)

[441]

(104)
-1

2 nm
(104)
0.27 nm

200 nm

(b)

5 nm

(d)

RGO

(1310)
(223)
(404)

α- Fe2O3

(2110)
50 nm

-1

10 nm

Figure 3.18. TEM images at different magnifications (a-b),
SAED pattern (c), and HRTEM image (d) with corresponding
fast Fourier transform (FFT) inset image of 1%wt RGO loaded
α-Fe2O3 annealed at 600°C for 3 hours in air.
13


amount in NFs was relatively little. The effect of annealing
temperatures on morphologies of 1.0 wt% RGO-loaded α-Fe2O3 NFs
was similar to that of pure α-Fe2O3 NFs. The rhombohedral structure
of α-Fe2O3 of the NFs was confirmed by the XRD results (JCPDS
33–0664). The EDX spectrum showed the presence of Fe, O, and C
elements from the RGO-loaded α-Fe2O3 NFs.
Fig. 3.18a showed a low-magnification TEM image of the 1.0
wt% RGO-loaded α-Fe2O3 NFs. The NFs with the diameter of 50–
100 nm consisted of many nanograins. The presence of RGO NSs on
the NF surface was shown in Fig. 3.18b. Parallel lattice fringes were
clearly visible in HRTEM images in Fig. 3.18c, which indicated a
good crystalline structure. The SAED result confirmed the
polycrystalline nature of the single-phase rhombohedral structure of
hematite α-Fe2O3 (JCPDS 33–0664). All observed results proved that
well-crystalline RGO-loaded α-Fe2O3 NFs were successfully
fabricated on chip by electrospinning.
3.3.2. H2S gas sensing properties of RGO-loaded α-Fe2O3
NFs sensors

3.3.2.1.
Effects of RGO contents
As shown in Fig. 3.19a–d, all the sensors presented a typical ntype sensing behaviour, which confirmed that the conducting channel
in RGO-loaded α-Fe2O3 NFs mainly went through α-Fe2O3 NFs
nanograins. The sensor response increased with increased RGO
contents up to 1.0 wt%, and then the response decreased with further
increased RGO contents. The similar results were obtained with the
effects of RGO content on the sensor DL in Fig. 3.19f. The enhanced
response of the RGO-loaded α-Fe2O3 NFs sensors was possibly
explained by the formation of a heterojunction between RGO and αFe2O3 and a homojunctions among α-Fe2O3 grain boundaries.
Furthermore, the presence of RGO and RGO/Fe2O3 interfaces in
RGO-loaded α-Fe2O3 NFs caused additional active gas-adsorption
sites like vacancies, defects, and oxygen functional groups; this
consequently enhanced the sensor response. However, when the
RGO content went up to 1.5 wt%, the sensor response declined
because RGO sheets connected together to form an individual
conducting path, which decreased overall sensor resistance (Fig.
14


1

0.25 ppm

H2S@350C &0.5 ppm

9.2

1 ppm

10
8

H2S@350C &0.1 ppm

6.1

7.3

6

(b)

0.5 wt.% RGO

(e)

H2S@350C &0.25 ppm

0.1 ppm

0.5 ppm

Resistance (M)

H2S@350C &1 ppm

(a)

DL (ppb) Gas Response (Ra/Rg)

-Fe2O3@ H2S&350oC

3.1

10

4
2

(f)

(c)

1 wt.% RGO

1
10

0
o

1.5 wt.% RGO

@Air&350 C

(d)

(g) 30

R (M)

10

20

5

10
0
0

500

1000

1500

2000

Time (sec)

0.0

0.5

1.0
RGO Conc. (wt.%)

1.5

Figure 3.19. H2S sensing transients of α-Fe2O3 NFs sensors
loaded with different RGO concentrations: 0 (a), 0.5 (b) 1.0 (c)
and 1.5 wt% (d). Sensor resistance (e), gas response (f), and
response time and recovery time (g) as a function of RGO
concentrations at working temperature of 350°C.
3.19g). As a result, exposure of the sensor to H2S gas also decreased
the resistance modulation and led to a weaker sensor response.
3.3.2.2.
Effects of working temperature
The effects of working temperature on gas sensing properties of
RGO-loaded sensor were similar to those of pure α-Fe2O3 NF sensors,
which indicated that the loading of RGO in the NFs did not affect the
working temperature.
3.3.2.3.
Effects of annealing temperatures
The effect of the annealing temperature on the response of the
sensor based on 1.0 wt% RGO-loaded α-Fe2O3 NFs was similar to
that of pure α-Fe2O3 NFs. RGO enhanced the sensor response at low
annealing temperatures but decreased the response at high annealing
temperatures, compared to pure α-Fe2O3 NFs. This was similar to the
effect of the annealing temperature on DL of the sensors of pure αFe2O3 NFs and 1.0 wt% RGO-loaded α-Fe2O3 NFs.
3.3.2.4.
Selectivity and stability
The RGO-loaded sensors showed high selectivity and short-term
stability. Regarding the selectivity to above test gases of the pure α15


Fe2O3 NFs sensor and the
10

Fe O NFs
9.2
Fe O NFs loaded 1 wt.% RGO
1.0 wt% RGO-loaded α8
SO @350 C& 10 ppm
Fe2O3 NFs sensor, the latter
6.1
H @350 C&1000 ppm
6 5.6
102
NH @350 C&1000 ppm
sensor had better selectivity
4
H S@350 C& 1ppm
to H2S gas (Fig. 3.24).
2.2
1.6 1.6 1.7 1.6
2 3.8
Conclusion of chapter 3
0
This chapter studied the
NH
HS
SO
H
effects
of
annealing
Figure 3.24. Comparative selectivity
temperature,

electrospun
of sensors based on α-Fe2O3 NFs
time and precursor solution
and 1.0 wt% RGO loaded α-Fe2O3
contents
(i.e.
PVA
NFs to various gases at 350°C.
concentration
and
salt
concentration) on morphology and structure of α-Fe2O3 NFs
fabricated on chip by electrospinning. The optimal results showed
that the α-Fe2O3 NFs sensor calcined at 600°C and electrospun for 30
min with the precursor solution of 11 wt% PVA and 4 wt% Fe
(NO3)3 gave a response of 6.1 to 1 ppm H2S gas at 350°C. In addition,
RGO enhanced the sensing properties of RGO-loaded α-Fe2O3 NFs
sensor compared to that of pure α-Fe2O3 NFs. The response of 1.0
wt% RGO-loaded α-Fe2O3 NFs sensors reached 9.2 to 1 ppm H2S at
350°C (1.5 times higher than that of pure α-Fe2O3 NFs at the same
conditions).
However, the response and selectivity of the sensors based on αFe2O3 NFs and their incorporation with RGO were not high.
Therefore, improving the sensor selectivity and response is essential,
which will be studied in the next chapter.
S (Ra/Rg or Rg/Ra)

2

3

2

3

o

2

o

2

o

3

o

2

2

2

3

2

CHAPTER 4. ZFO NFs AND THEIR LOADING WITH RGO
FOR H2S GAS SENSING APPLICATION

4.1.
Introduction
Sensors based on binary α-Fe2O3 have low selectivity because
they are sensitive to many different gases and their sensor response is
also quite low. Many methods including doping binary α-Fe2O3 with
noble metals and combining binary α-Fe2O3 with other metal oxides
to form composites or ternary compounds have been used to improve
16


the sensor selectivity and response. Particularly, ternary ZFO, a
typical normal spinel with cubic crystal structure, is a promising
material for detecting gases because of its good chemical and thermal
stability, low toxicity, high specific surface area and excellent
selectivity. The gas sensing properties of ZFO, especially to H2S,
have been investigated in many works. However, researches on H2S
gas sensitivity, especially at sub-ppm concentrations, of NFs ZFO
sensors have not been published. In addition, the effects of heat
treatment parameters such as annealing temperature, annealing time
(a)

(c)

(440)
(511)
(422)
(400)
(311)
(220)
-1

200 nm

(b)

5 nm

(d)

[101]

(020)

(000)

(1ī ī)

0.49 nm

0.42 nm
100 nm

5 nm

Figure 4.7. TEM images at different magnifications (a-b), SAED
pattern (c), and HRTEM image (d) with corresponding fast
Fourier transform (FFT) inset image of ZFO-NFs calcined at
600°C for 3 h in air.
and annealing rate on the sensor morphology, structure and H2S gas
sensing properties of NFs ZFO sensors have not been investigated.

Furthermore, the H2S gas-sensing performance of ZFO NFs
loaded with RGO has not been also studied. Therefore, in this
chapter, ZFO-NFs sensors and their incorporation with RGO were
fabricated by facile on-chip electrospinning. Then, the effects of heat
treatment conditions on morphology, structure and H2S gas sensing
performances of the ZFO NFs sensors were investigated.
17


Simultaneously, the effects of RGO concentration and annealing
temperature on the H2S gas sensing properties of the RGO-loaded
ZFO NFs sensors were also discussed in detail.
4.2.
H2S gas sensors based on ZFO NFs
In this section, the morphology and structure of the ZFO NFs as
well as the influence of heat treatment conditions (i.e. annealing
temperature, annealing time, and heating rate) on morphology,
structure and H2S gas sensing characteristics of ZFO NFs were
systematically investigated.
4.2.1. Microstructure characterization
The cubic spinel structure of ZFO NFs at different calcination
conditions was confirmed in XRD pattern. The nanograins and
crystallinity of the ZFO NFs increased with increased annealing
temperature from 400 to 700°C and with the increased annealing
time from 0.5 to 48 hours. Whereas, the grain size and crystallinity
of ZFO NFs declined with the increased heating rate between 0.5 and
2°C/min because of a dramatic decrease in calcination duration.
However, with a further increase in the heating rate, the grain size
and crystallinity also rose. FESEM confirmed the effect of annealing
temperature and annealing time on the NFs morphologies. Whereas,

the heating rate changed from 0.5 to 5°C/min, the ZFO NFs were
still spider-net-like and continuous, however, when the heating rate
went up to 20°C/min, almost NFs with thinner diameters were
fractured. Only NFs with larger diameters were still continuous. The
EDX detected four elements (Fe, O, Si and Zn).
The morphology and microstructure of ZFO NFs were further
examined by TEM and HRTEM images (Fig. 4.7). Obviously, the
synthesized ZFO sample was the multi-porous NFs composed of
many nanograins with the average grain size of about 5–25 nm (Fig.
4.7a–b). The SAED pattern of the ZFO NFs in Fig. 4.7c revealed that
the diffraction rings combined with the spots of polycrystalline
nature of the cubic spinel ferrite phase. The HRTEM image and
corresponding FFT inset image in Fig. 4.7d further confirmed the
crystalline nature of the synthesized ZFO NFs. The HRTEM image
exhibited parallel lattice fringes with spacing approximately 4.9 and
4.2 Å, corresponding to lattice planes (020) and (1 1 1 ), which was
proved by the FFT inset.
18


80

H2S@ 1 ppm & 350oC

Calcinated @0.5 h
Calcinated @3 h
Calicnated @12 h
Calcinated @48 h

(b)

(a)

(s)

60

100
80
60
40
20

Recovery time 100

40

(c)

20

resp./recov.

Response (Ra/Rg)

100

Resp. (Ra/Rg)

4.2.2. Gas sensing properties

4.2.2.1.
Effects of the operating temperature
The sensor response and recovery time strongly increased when
the operating temperature decreased from 450 to 250°C. The
working temperature of 350°C was selected to further investigate gas
sensing properties by compensation between the sensor response and
recovery time.

10



Response time
1
0.50

0.75

1.00

0

12
24
36
48
Calcinated time (h)

H2S conc. (ppm)

60

(e)

100
80
60
40
20

(d)

Recovery time

40

(f)
Response time

100
10



20

(s)

80

H2S@ 1 ppm & 350oC
o

Heating rate @0.5 C/min
o
Heating rate @2 C/min
o
Heating rate @5 C/min
o
Heating rate @20 C/min

resp./recov.

Response (Ra/Rg)

100

Resp. (Ra/Rg)

0.25

0.25

0.50

0.75

H2S conc. (ppm)

1.00

0

5
10
15
20
Heating rate (oC/min)

Figure 4.10. Response at working temperature of 350°C as a
function of H2S concentration for different annealing time (a) and
heating rate (d). Response and response-recovery time as a
function of annealing time (b, c) and heating rate (e, f).
4.2.2.2.
Effects of the annealing temperature
When the annealing temperature increased from 400 to 600°C, the
sensor response also went up. The sensor response fell down with the
further increased annealing temperature. This was explained by the
as-mentioned effects of grain size and crystallinity on gas sensing
properties of the sensor. The DL calculation was 0.048 ppb
corresponding to the sensor calcined at 600°C. Therefore, 600°C was
19


selected as the optimal annealing temperature for ZFO NFs.
Response time and recovery time also decreased with the increased
annealing temperature from 400 to 700°C.
4.2.2.3.
Effects of annealing time and heating rate
According to Fig. 4.10, the sensor response increased with the

increased annealing time from 0.5 to 3 h; however, the sensor
response decreased with the further increased annealing time up to
48 h. The recovery time also relatively decreased with increased
annealing time. Regarding the heating rate, two peak responses were
observed at 0.5 and 5°C/min. The sensor calcined at the heating rate
of 0.5°C/min showed the highest response, however, the heating rate
could not further decrease because of the limitation of furnace
temperature controller. As observed in Fig. 4.10h, response-recovery
time as a function of the
heating rate decreased
100
when
heating
rate
increased from 0.5 to
20
5°C/min, but slightly
increased when heating
10
rate was further increased
to 20°C/min. The effects
0
NH
HS
SO
H
of crystallinity, grain size,
Figure 4.12. Comparative selectivity
and
high

porosity
of ZFO NFs-based sensors and αstructure were employed
Fe2O3-based NFs sensors to various
to explain these gas
gases at 350°C.
sensing results.
4.2.2.4.
Selectivity and stability
The results showed that the ZFO NFs sensors displayed an
excellent selectivity to H2S gas among other gases. Moreover, as
shown in Fig. 4.12, the ZFO NFs sensor exhibited the higher
selectivity and response than the sensors based on α-Fe2O3 NFs and
their loading with RGO, which may be due to unique spinel crystal
structure of ZFO as reported in many works. The sensors also had a
good stability throughout the cycle test.
4.3.
H2S gas sensors based on ZFO NFs loaded with
RGO
4.3.1. Microstructure characterization
Fe2O3 NFs

S (Ra/Rg or Rg/Ra)

Fe2O3 NFs loaded 1 wt.% RGO
ZFO NFs

SO2@350 oC& 10 ppm

H2@350 oC&1000 ppm

NH3@350 oC&1000 ppm
H2S@350 oC& 1ppm

2

20

2

3

2


The FE-SEM images of RGO-loaded ZFO NFs with different
amounts of RGO confirmed that the NFs morphologies were not
significantly affected by the change of the RGO content (0–1.5 wt%).
and RGO could not be found from FESEM of RGO-loaded ZFO NFs.
FESEM images also showed that NF surfaces became rough due to
the increased nanograin size with the increased annealing
temperature. The XRD patterns proved the cubic spinel structure of
ZFO (JCPDS 89–7412). The EDX spectrum indicated the presence
of C, Fe, Zn and O elements from the NFs. The TEM, HRTEM,
SAED, and FFT results further determined that well-crystalline
RGO-loaded ZFO NFs had been successfully fabricated by
electrospinning.
4.3.2. Gas-sensing properties
4.3.2.1.
Effects of RGO contents
The sensor response reached a maximum at 1.0 wt% RGO when

the RGO content changed from 0 to 1.5 wt%. The explanation for
this result was similar to that of the RGO-loaded α-Fe2O3 NFs sensor
as mentioned in Section 3.3.2.1. The recovery time increased when
the RGO weight percentage rose between 0 and 1.0% due to the
increase in the amount of absorbed H2S gas. However, when the
RGO concentration further increased to 1.5 wt%, the individual
conducting path was formed, leading to the increased electron
mobility and decreased sensor recovery time. Meanwhile, the
response time of the sensor did not change much when RGO content
was varied. The response time of all RGO-load ZFO NF sensors was
quite short, below 10 s.
4.3.2.2.
Effects of operating temperature
The response decreased with the increased working temperature
from 250 to 450°C, which was as similar as that of pure ZFO NF
sensors in Section 4.2.2.1. This confirmed that the loading of RGO in
the NFs did not affect the working temperature as mentioned in
Section 3.3.2.2.
4.3.2.3.
Effects of annealing temperatures
The response reached a maximum at 600°C for all H2S
concentrations with the increased annealing temperature from 400 to
21


(a)

147

100

61

ZFO NFs

(d)

21.8
8.2

15.8

10
Response time

0
600

700

1000
100

Recovery time

500

0.2
0.0

ZFO NFs lai 1 wt.% RGO

43.4

400

0.4

(c)

77.6

50

0.6

1 wt% RGO+ZFO
ZFO

102

6
4
2
0

(s)

150

(b)

DL (ppb) SZFO+RGO/SZFO

1wt.% RGO-loaded ZnFe2O4@ 1ppm H2S &350 oC

400

500

600

700

resp./recov.

o

@ 1 ppm H2S & 350 C

ZnFe2O4@ 1ppm H2S&350 oC



Response (Ra/Rg)

200

Annealing temp. (oC)

Figure 4.20. Comparison of response (a), change level of response

(b), detection limit (DL) (c) and response-recovery time (d) of
bare-ZFO and 1%wt RGO loaded ZFO NFs sensors to 1 ppm H2S
gas at 350°C as a function of annealing temperatures.
700°C because of the inverse effects of nanograin size and
crystallinity. In addition, the weight loss of RGO also declined with
the decreased annealing temperature, making the rest of the RGO
amount be enlarged, which caused the sensor response to change.
The response time remained almost unchanged when the annealing
temperatures varied while the recovery time fluctuated with
increased annealing temperatures because of the effects of nanograin
size and RGO weight loss.
Fig. 4.20ab presented the comparative effects of annealing
temperature on the sensor response of the pure ZFO and RGO-loaded
ZFO NF sensor. The magnitude of the effect on the response of the
RGO-loaded sensor significantly decreased with the increased
annealing temperature due to the increased weight loss of RGO. It
was the same as the effects of annealing temperature on the DL of
the sensors of pure ZFO NFs and RGO-loaded ZFO NFs in Fig.
4.20c. The response and recovery time of the pure ZFO-sensor were
longer than those of the RGO-loaded sensor because of the larger
electron mobility in RGO-loaded sensors.
4.3.2.4.
Selectivity, stability and RH effects
The sensor had good reproducibility and short-term stability. The
RGO-loaded ZFO NFs sensor also had excellent selectivity to H2S
gas. The RGO-loaded ZFO NFs sensor also presented better
selectivity to H2S than pure ZFO sensor. Besides, the effects of RH
on the response of the sensor were also investigated.
22

The H2S sensors in this thesis had much higher response than
sensors based on other materials or nanostructures due to RGO
influence, morphologies and structures of the as-synthesized NFs.
Obviously, the H2S response of ZFO sensor was higher than that of
the α-Fe2O3 sensor, which was because ZFO NFs had a multi-porous
structure while α-Fe2O3 NFs had a quite dense structure as seen in
TEM images. In addition, the unique crystal structure and high
surface activity with many defects of ZFO made it become a superior
gas sensing material, especially to reducing gases. Moreover, the
sensor loaded with RGO had a higher H2S response than the pure
sensor in same working conditions because of the heterojunction
between RGO and α-Fe2O3 or ZFO.
Conclusion of chapter 4
The chapter optimized heat treatment conditions (temperature
annealing, time annealing and heating rate) which strongly affected
morphologies, microstructures and gas sensing properties of ZFO
NFs sensors. The optimal condition showed that the ZFO NFs
calcined at 600°C for 3 h with heating rate of 0.5°C/min gave the
best response of 102 to 1 ppm H2S gas at the working temperature of
350°C. The results also confirmed that RGO enhanced sensitivity
and selectivity. The responses of 1 wt% RGO-ZFO NFs were 1.5
times higher than that of pure ZFO NFs at the same conditions,
which was significantly affected by formation of heterojunctions
between RGO and ZFO.
CONCLUSIONS AND RECOMMENDATIONS
Conclusion
Based on the above-mentioned research results, some conclusions
were drawn and listed hereafter.
 The α-Fe2O3 NFs, ZFO NFs and their loading with RGO were

successfully on-chip fabricated by electrospinning method.
 The precursor solution contents, electrospun time, and annealing
temperature,
which
strongly
affected
morphologies,
microstructures and gas sensing performances of α-Fe2O3 NFs,
were optimized. The optimized α-Fe2O3 NFs get response of 6.1
to 1 ppm H2S at 350ºC, which corresponding to PVA
23