MINISTRY OF EDUCATION AND TRAINING

MINISTRY OF NATIONAL DEFENCE

ACADEMY OF MILITARY SCIENCE AND TECHNOLOGY

NGUYEN VAN THANG

STUDY ON SYNTHESIS PROCESS OF SILICON NANOMATERIALS
TO FABRICATE ANODE ORIENT APPLICATION FOR Li-ION BATTERIES

Specialization: Theoretical chemistry and Physical chemistry
Code: 9 44 01 19

SUMMARY OF DOCTORAL THESIS

Hanoi - 2019


This thesis has been completed at:
Academy of Military Science and Technology, Ministry of Defence

Scientific supervisors:
Dr. Nguyen Tran Hung
Assoc. Prof. Dr. Nguyen Manh Tuong

Reviewer 1:

Prof. Dr. Vu Thi Thu Ha
Vietnam Institute of Industrial Chemistry


environment. Therefore, energy storage issues are becoming increasingly
important today.
Li-ion battery (LIB) is an advanced battery generation with many
advantages such as high specific energy density, fairly compact size, the
high number of discharging cycles (about 400 - 600 times) should be
applied widely used.
However, due to the increasing demands of practice, there are still many
studies to increase the capacity and durability of LIB. One of the studies
concerned is the battery anode material. Graphite materials are widely used
because they are cheap and easy to manufacture but have low capacity, only
in the range of 130 - 270 mAh /g in actual operation.
In materials that can be applied to the anode of LIB, silicon is a potential
material, thanks to its very high specific capacity (up to 4200 mAh/g),
corresponding to Li22Si5 compound. In operation, the reaction of Si with Li
can cause material cracking due to an increase in the volume of about
400%, reducing the capacity of the electrode as well as the battery capacity
and reducing the durability of the electrode.
The situation of nano-silicon synthesis for the application of anode Liion battery: Complex silicon nanoparticles technology, requiring modern
and expensive equipment; Sources of silicon in nature: in rice husk, SiO2
content accounts for a high proportion (over 20% of mass). Rice husk has a
large yield, hardly used effectively. Studies on the synthesis of silicon
nanoparticles from rice husk with simple and easy-to-implement
technology have not been paid much attention.
Stemming from the practical problems, the PhD student has proposed
and implemented the thesis topic: “Study on synthesis process of silicon
nanomaterials to fabricate anode orient application for Li-ion batteries”.


2


The research results of the thesis contribute to the basic research
direction of thermodynamic and kinetic properties of the synthesis process
of silicon nanoparticles, used to manufacture anode of LIB with the aim of
increasing the capacity and performance of anode electrode. This is a
meaningful research direction, if successful will contribute to solving the
current problems of anode material sources, thereby improving the battery
quality of devices using power from batteries.


3

6. The layout of the thesis:
The thesis consists of 130 pages divided into the following sections:
Introduction; Chapter 1: Overview; Chapter 2: Subjects and research
methods; Chapter 3: Results and discussion; Conclusion; List of published
scientific works; References.
Chapter 1. OVERVIEW
Overview of LIB, on anode electrode materials of LIB, on silicon
nanoparticle synthesis methods and the current status of thermodynamics,
kinetics of synthesis of silicon from rice husk, overview of graphene
synthesis methods. Since then, set the scientific basis and orientation for the
implementation of the research content of the thesis.
Chapter 2. SUBJECTS AND METHODS OF RESEARCH
2.1. Research subjects
Rice husk, nano silica, silicon nanoparticles and experimental conditions
for the synthesis of silicon nanoparticles from rice husk; rGO, nano Si@rGO.
The anode of LIB is made on the basis of silicon nanoparticles, rGO,
nano Si@rGO.
2.2. Methods of analysis
TG / DTA, DSC thermal analysis method; Method of scanning electron

crush it, wash with a solution of HCl and HF to remove impurities, then
wash the mixture with distilled water to neutral, centrifuge and decant the
solids. The resulting solids bring vacuum drying to dryness, grinding, and
obtaining silicon nanopowder.

Figure 2.2. Process of synthesizing silicon nano from silica nano
2.3.3. rGO synthesis process from graphite
2.3.3.1. Oxidizing graphite into graphene oxide (GO)
Add the mixture of graphite and KMnO4 (mass ratio 1: 6) to the mixture
of concentrated H2SO4 and H3PO4 (volume ratio of 9: 1) in the flask, placed
on the ice cube tray (the mixture temperature does not exceed 10 oC). The
ratio of acid mixture: graphite volume is 100 mL: 1 g graphite. The mixture
is continued to stir for 10 minutes, remove the ice tray, start heating, keep
the mixture reacted at 80 oC for 3 hours. The reaction mixture is cooled to


5

room temperature, diluted with distilled water in a 2 L volumetric flask,
centrifuged, filtered, washed and obtained GO in a dark brown gel.
2.3.3.2. Reduce GO into rGO by thermal shock method

Figure 2.3. rGO synthesis process from graphite
Freeze dried GO in the furnace, quickly raise the temperature to 800 oC,
keep this temperature for 20 minutes in an argon stream. The product
obtained after heat reduction is rGO.
2.3.4. The synthesis process of nano Si@rGO

Figure 2.4. The synthesis process of nano Si@rGO
2.3.5. The anode fabrication process of LIB

TG/%

10.14

d TG/% /min
Peak :54.34 °C

60

Peak :475.19 °C

40
-3
20

Peak :303.45 °C

0

-6

Mass variation: -7.62 %

-20
Mass variation: -51.48 %

-40

-9


The TGA curve of RHs in the air has an obvious mass loss process
where percentage of mass loss increases gradually with the increasing of
ramp rates and TGA curve seems to move toward to the right. In details, the
TGA plots of leached RHs show a typical three-stage mass loss in air: (i)
mass loss below 100 °C, which corresponds to humidity loss; (ii) mass loss
around 300 °C, which corresponds to cellulose/hemicellulose/lignin
degradation; and (iii) mass loss between 350-550 °C, corresponding to the
burning of carbonous residues. Around 12 % of the mass remains as the
SiO2 product.
3.1.1. Investigate the effects of acid treatment
3.1.1.1. The effect of the acid treatment time

Figure 3.2. SiO2 content in rice husk ash dependence on acid treatment time


7

The SiO2 content increases when the HCl acid wash time increases,
helping to remove metal impurities in rice husk. During acid washing time
from 2 and 3 hours, SiO2 content did not change much. Because metal
impurities combine organic and inorganic components in rice husk, the
treatment process, acid cannot penetrate and completely dissolve impurity
metals. Maybe this is the limit of the process of dissolving metal impurities
in rice husk, so we choose an acid treatment time of 2 hours.
3.1.1.2. The effect of the acid treatment temperature
With an acid treatment time of 2 hours, the SiO2 content in rice husk ash
increases with increasing treatment temperature. This is explained when the
temperature increases, the rate of dissolution of metal impurities in rice husk
increases, making the treatment of fast metal impurities balanced. However,
at 100 oC, the vapor evaporates strongly, leading to strong HCl vapor.

Content of SiO2 in rice
husk ash, % weight
96,79
96,64
96,73
96,81
96,65

Remaining acid
concentration, C %
8,7
9,1
8,9
9,0
8,8

Thus, the proper condition of the treatment of RHs by acid are: HCl acid
10 %, temperature 90 oC, time 2 hours, ratio of RHs/acid: 3.0 g/40 mL.
3.1.2. Investigate the effects of calcination mode
3.1.2.1. The effect of calcination temperature:
Temperatures below 600 oC, low SiO2 content in rice husk ash, organic
matter has not completely burned, ash is tarnish black. At temperatures of
600 °C or more, other substances have almost burned, the ash has turned
white, the SiO2 content has also increased. When the temperature reaches
650 oC, SiO2 content in rice husk ash reaches over 97 %. When increasing
the calcination temperature to 700 oC, the SiO2 content in rice husk ash
increased insignificantly, indicating that the burning process took place
quite thoroughly. Thus, an appropriate husk burning temperature is 650 oC.

Figure 3.5. Influence of calcination temperature on SiO2 content in rice husk ash

temperature: 90 oC; Processing time: 2 hours; rice husk/acid ratio: 3 g rice
husk/40 mL 10 % HCl; Rice husk calcination temperature: 650 oC for 3
hours at a heating rate of 3 oC / min.


10

3.1.3. Thermodynamics, the kinetics of the synthesis of nano silica from rice husk
3.1.3.1. Kinetic characteristics of the synthesis of nano silica from rice husk.
* DSC thermal analysis: rice husk is treated with 10 % HCl acid
solution at 90 oC. After 2 hours of treatment, rice husk is washed several
times with water to a neutral environment. Rice husk samples after acid
treatment are heated in air to a temperature of 650 oC at different heating
rates. The DSC thermal analysis curve of rice husk burning process is
shown in the figure:

Figure 3.8. DSC curves of rice husk burning process
with heating rates of 3, 6, 9, 12, 15 oC / min

From the DSC thermal analysis curve, we determined the basic
kinematic parameters according to the FWO model and Kissinger model in
the process of burning rice husk in air.
Basic kinematic parameters according to the FWO model and Kissinger model
in rice husk burning process
β
(K/min)

Tp
(K)



611,1

1,64

0,95

-10,63

12

620,2

1,61

1,08

-10,38

15

622,7

1,60

1,18

-10,16

*

119,5

143,1

573

-66,9

117,8

156,1

723

-68,9

116,6

166,4

923

-70,9

114,9

180,3

ΔG˚> 0: not self-evolving; ΔS˚ 0: endothermic reaction.

The heating rate significantly affects the morphology and size of silicon
nanoparticles. At a heating speed of 5 oC, the size of Si-R5 silicon
nanoparticles is uniform and smaller than that of Si-R15 particle heating
rate of 15 oC/min. The different characteristics of Si are generated at
different heating rates possibly by local heat accumulation. Since the
reduction of SiO2 reaction by Mg is an exothermic reaction, a large amount
of heat is released at the locations where SiO2 is in direct contact with Mg.
With a fast heating rate, there is not enough time for heat or transfer to
occur, so the local temperature rises at nearby reaction centers, adding to the


13

local temperature. These high-temperature reaction centers cause the
synthesis of Si products and the disappearance of small pores.

Figure 3.14. DSC curve of SiO2
reduction by Mg

Figure 3.15. XRD pattern of RH-5
silicon nano sample

DSC shows the reaction occurring around 330-350 oC, and the greater
the heating rate, the greater the heat build up (Figure 3.14), leading to a
strong reaction and pushing the reactants. Therefore, to obtain nano-Si, a
slow heating rate is required, about 5 °C/min is appropriate.
The obtained nano-Si material has a crystal structure. The mechanism of
Si formation is the crystalline structure from amorphous SiO2 material due
to the chemical reaction with Mg in liquid Mg-Si-O alloy state. After
removing MgO, the structure of the remaining Si will be crystallized from


1/Tp103
(K-1)
1,66
1,64
1,63
1,62

Figure 3.17. Plots of lg and 1/Tp of the
reduction of SiO2 of the F-W-O model

FWO,
lgβ
0,70
0,95
1,10
1,18

Kissinger,
ln (β/Tp2)
-11,20
-10,63
-10,35
-10,13

Figure 3.18. Plots of ln(β/Tp2) and 1/Tp of the
reduction of SiO2 of Kissinger model

- Model FWO: Activated energy, E* = 308.34 (kJ/mol).
- Kissinger model: Activated energy, E* = 314. 1 (kJ / mol), A = 8.19.1026.

Graphite clearly shows the structure of the crystallized material, GO has
an unclear crystalline structure. This result is due to the functional groups
containing oxygen on the surface and boundary of GO plate.

Figure 3.25. XRD pattern of graphite

Figure 3.26. XRD pattern of GO

Figure 3.27. SEM image of graphite

Figure 3.28. SEM image of GO


16

GO products have a clear layer structure, made up of small GO pieces.
This result shows the ability to self-assemble when drying GO. This has
practical implications for the purpose of the thesis when using graphene
films wrapped around silicon nanoparticles.
3.3.1.2. Results of rGO synthesis
After drying, dehumidifying rGO retains graphite layer structure, the
thickness has decreased many times, due to the oxidation process, the size
of rGO plates has decreased compared to raw graphite.

Figure 3.29. SEM images of rGO at different resolutions

Figure 3.30. XRD pattern of rGO

Figure 3.31. EDX pattern of rGO


From the SEM image, there was no clear morphological difference
between the two silicon and rGO nanomaterials after GO reduction. GO
elimination does not affect the intrinsic properties of both materials.
From TEM images, the formed rGO plates were wrapped around
silicon nanoparticles, not affecting the size of silicon nanoparticles.
Therefore, the Si@rGO nanocomposite mixture is used as a slurry
mixture for lithium ion battery testing.
3.4. Study on the applicability of rGO and Si@rGO nanomaterials
used to fabricate anode of LIB.
3.4.1. Experimental fabrication of anode material combination

Figure 3.36. Anode of LIB before and after drying in vacuum


18

After the testing process, we selected the rate of 80 % electrode
materials, 10 % PVDF and 10 % SUP-P carbon to produce electrodes. With
this ratio, the fabricated electrodes have a smooth surface, good adhesion to
the surface of the copper plate. Observe the SEM image of the combination
of anode materials on the picture, the surface is quite smooth, the polymer
adhesive material layer has covered a thin layer on inorganic material
particles.
3.4.2. Electrochemical performances of LIB
3.4.2.1. Survey the electrochemical performances of LIB with anode
fabricated on rGO basis
* Cyclic Voltammetry
The sample of half cell of rGO (Li/rGO) was analyzed by CV method
with a scanning speed of 0.1mV/s in the range of 0 - 2 V. Electrochemical
performances by CV method of half cell:

The flow density was started at 0.1C (37.2 mA/g) with a sample of halfanalyzed CV cells, gradually increased after every 10 cycles, to 50C
(18600mA/g), as shown in Figure 3.38. The results show that anode can
work at very high current density, but the capacity drops very low. At low
current density (0.1C; 0.2C; 0.5C) the capacity of the anode is unstable,
indicating that the charging and discharging process is not effective.
However, after charging-discharging at high current density, the anode is
evaluated at 0.1C current density. The 10-cycle nap-discharge result shows
that the anode works fairly stable with unchanged capacity compared to
before the evaluation. This suggests that it is possible for rGO anode
materials to be charged - discharge at high current density before being able
to work stably at low current density.

Figure 3.38. Cycling performance of LIB Li/EC:DMC 1:1, LiPF6 1M/rGO under the different
current rates in the voltage range of 0 to 2.0 V.

3.4.2.2. Survey the electrochemical performances of LIB with anode
fabricated on nano Si basis
* Cyclic Voltammetry

Figure 3.39. Cyclic voltammetry curves of LIB Li/EC:DMC 1:1,


20
LiPF6 1M/Si at a scan rate of 0.1 mV/s in the voltage range of 0 to 2.0 V.

Electrochemical characteristics through CV analysis show that in the
first cycle the charge potential of the Si anode ~ 100 mV shows that Si
materials have a crystal structure, due to crystalline Si reacting with Li+ at
this voltage (while amorphous Si is ~ 200 mV). This result again
demonstrates that nano Si have a crystal structure. Finish the loading

21

DMC 1: 1, LiPF6 1M/Si reaches the specific capacity of the adjacent anode
3000 mAh/g for the first 10 cycles. From cycle 11 onwards the capacity
decreases, to 2250 mAh/g in the 35th cycle. After the 35th cycle, the capacity
decreases sharply, indicating the battery's ability to work is not guaranteed.
This result corresponds to other claims about nano-Si materials. It is
worth noting here that the Coulombic efficiency is not high, reaching in the
range of 93 – 97 %, indicating that the reversible charge-discharge process
of the Si anode is not satisfactory. This result can explain the working
durability (number of charging and discharging cycles) of only 35 cycles.
3.4.2.3. Survey the electrochemical performances of LIB with anode
fabricated on nano Si@rGO basis
* Cyclic Voltammetry
The CV curve for the first 2 cycles, showing the electrochemical
properties of Si@rGO nano anode. If compared with the CV characteristics
of nano-Si anode, there will not be many other points. With the first cycle,
due to the presence of rGO, the charging process starts around 170 mV.
From 100 mV is the charging process of crystalline Si nanomaterials. At the
end of the charging process, the potential drops to 0mV, corresponding to
the formation of Li15Si4 crystalline and C6Li compounds. At the discharging
process, 2 peaks at about 330 mV and 490 mV. This demonstrates the
simultaneous existence of Li15Si4 amorphous Si crystalline and nano form.
At the end of the first cycle, the remaining nano-Si anode is amorphous and
carbon (graphene). At the second cycle, the charging process starts earlier,
at 500 mV nearby. This is the formation of the SEI layer on the anode
surface (mainly amorphous nano-Si) in parallel with the process of creating
LixSiy compound. When the potential decreases to 0 mV, the amorphous Si
nanotubes and graphene end up as Li15Si4 crystalline and C6Li compounds,
similar to the first cycle. The discharging process in the second cycle is

Potential (V vs Li/Li )

Figure 3.41: Cyclic voltammetry curves of LIB Li/EC:DMC 1:1,
LiPF6 1M/Si@rGO at a scan rate of 0.1 mV/s in the voltage range of 0 to 2.0 V.


22

* The charge-discharge characteristics of LIB:
Parallel to the results of analysis by characteristic CV, the half cell is
evaluated by galvanostatic analysis for the first 2 cycles. The results show
that the basic difference of 2 cycles is at the charging process. At the first
cycle, the charge at the voltage is lower than that of the second cycle,
because the first silicon nanomaterial is in crystal form. At the end of the
first cycle discharge process only exists amorphous nano-Si form.
Therefore the charging process will be the same from the second cycle. The
point here is that for the discharge characteristic line, it is shown that the flat
section at the voltage is about 40 mV, corresponding to the CV
characteristic when the compound exists Li15Si4 amorphous Si crystal and
nano form. At the 0.05C current density, the anode capacity is > 3000
mAh/g. This is the common value for anode including nano-Si and rGO.
The first cycle, specific capacity reaches > 1800 mAh/g and Coulombic
performance reaches 96 %. These are very high values, comparable to Si
nanowire materials in published studies. Within the first 200 cycles, the
specific capacity of the anode stays at values > 1200 mAh/g and stays at
this value for 500 cycles. This value can be compared to the number of
discharge cycles of a commercial Li-ion battery using graphite anode
material. This result shows that the role of rGO material has high electrical
conductivity, which helps stabilize the charging and discharging of the
battery. It can be explained by the nanocomposite structure, the graphene

2. Determined thermodynamic parameters, kinetics of the synthesis of
silica nano from rice husk: activation energy of synthesis of silica nano
from rice husk: E* = 126.14 (kJ/mol) (according to the FWO model); E* =
122.6 (kJ/mol) and the pre-exponential factor in the Arrhenius equation is A
= 1.033.1010 (according to the Kissinger model), there by determining the
reaction rate constant according to the Arrhenius equation. Determination of
thermodynamic parameters of the synthesis of silica nano from rice husk:
∆G* = 138.5  180.4 kJ/mol, ∆H* = 114.9  120.1 kJ/mol and ∆S* = -70.9 
-61.5 J/mol.K.
3. Determined suitable conditions selected for the fabrication of silicon
nanoparticles from silica nano with reducing agent Mg, molar ratio Mg:
SiO2 is 2.1: 1, calcination temperature at 800 oC in 2 hours at a heating rate