1.Introduction
Lithium-ion conducting glasses have been widely studied due to their potential
application as solid-state amorphous electrolytes in secondary batteries.
Since amorphous electrolytes could increase lithium ion conductivity and
have more contact area with electrode. LiPO3 stands out due to high Li-ion
conductivity. Furthermore, adding other lithium metal oxide could increase
ion conductivity. Meanwhile, other lithium metal oxide could decrease the
glass transition temperature thus electrolyte has larger contact area with
electrode during working. Hence Li3BO3 and Li4SiO4 are two candidates for
improving LiPO3 battery characters.
Radio Frequency (RF) thermal plasma can be characterized as: a high enthalpy
flame with extremely high temperature fields, low velocity (10~20 m/s)
resulting long reaction time, absence of electrodes thereby avoids contamination,
and rapid quenching. Therefore, RF plasma has been widely applied in production
of nanoparticles which are difficult to synthesized by other methods. In
present study, LiPO3, Li3BO3 and Li4SiO4 are tired to be synthesized by
RF thermal plasma respectively.
2.Experimental
The setup mainly consists of a powder feeder, a plasma torch, a reaction
chamber, a quenching tube, and a particle collection filter. Quenching
tube was set under the torch with a distance of 15 cm. Raw materials were
fed into thermal plasma and instantaneously evaporated due to the high
enthalpy and homogeneous nucleation, condensation, coagulation will happen
in the chamber. Synthesized nanoparticles were collected from the inner
wall of the reaction chamber and the filter.
Experiments were operated at a condition of 4MHz, 20 kW, atmospheric pressure. Ar was introduced as carrier gas (3 L/min), inner gas (5 L/min). Sheath gas was set at 60 L/min with 0~5 L/min comes from O2, 55~60 L/min comes from Ar. Quenching gas was injected with different flow rate from 20~40 L/min. Raw material was mixed with stoichiometric ratio and the feed rate changes from 200~1000 mg/min.
Prepared nanoparticles were characterized for phase identification by X-Ray
Powder Diffraction (XRD). Amorphization degree was calculated semi-quantitatively
by JADE 7.0.
Morphology of the particles, diffraction patterns and size distributions
were observed by Transmission Electron Microscopy (TEM) and Scanning Electron
Microscope (SEM) observation. RAMAN was used to identify the structure
of the constituent molecules. Element mapping and relative abundance were
observed by STEM-EDS.
3.Results and Discussion
(a) Li-P-O system
In Li-P-O system, amorphous LiPO3 nanoparticles were synthesized successfully
regardless of the oxygen gas flow rate and feed rate change. Amorphization
degrees are almost 100wt%. Since necessary quenching rate for LiPO3 is
smaller (16.6 K/s) than quenching rate estimated in plasma (10^4 K/s),
therefore, amorphous LiPO3 nanoparticles were synthesized effortlessly.
However, mean diameter of the nanoparticles increases as the feed rate
increases due to the longer growth time and larger growth rate.
(b) Li-B-O system
In Li-B-O system, amorphous high-temperature phase Li3BO3, β-Li3BO3, was
synthesized. This unstable component was preserved because of the high
quenching rate in RF thermal plasma. Amorphization degree of β-Li3BO3 nanoparticles
increased notably from 60wt% to 71wt% due to the increase of quenching
gas. Since quenching rate is estimated to increase from 104 K/s to above
105 K/s due to existence of quenching gas, thus amorphization degree was
increased.
(c) Li-Si-O system
Diffraction peaks of XRD correspond to Li4SiO4 were observed. All the peaks
become broader when quenching gas was injected. Li2CO3 and Li2SiO3 peaks
were also detected. Li2CO3 peaks come from synthesized Li2O particles reacted
with CO2 in atmosphere after the experiments. Li2SiO3 was considered as
by-products.
The amorphization degree increased from 52wt% to 70wt% after quenching gas was injected. The reason is the same with Li-B-O system. Since plasma plume becomes compressed in axial direction when quenching gas was injected. In compressed area, quenching rate was estimated to be higher than 10^5 K/s even 10^6 K/s according to numerical analysis while quenching rate is 10^4 K/s without quenching gas. Therefore, amorphization degree was increased.
Halo diffraction patterns in TEM images suggest amorphous state exists
in all the products. Spherical particles were observed in no quenching
gas condition with a mean diameter at 93 nm. Spherical and irregular nanoparticles
were observed when quenching gas at 20 L/min and 40 L/min. Since the less
long-range order in material, the higher amorphization degree would be.
Therefore, the particles become irregular and hard to clarify.
Formation mechanism of Li4SiO4 nanoparticles was investigated. Si has the
highest nucleation temperature then nucleates firstly. Li oxide and silica
vapor co-condense on Si then Li4SiO4 was synthesized.
4. Conclusion
In present work, amorphous nanoparticles of LiPO3, Li3BO3, and Li4SiO4
were synthesized by RF thermal plasma successfully and all the formation
mechanisms were investigated. LiPO3 could be synthesized even at 1000 mg/min.
Quenching gas could increase the amorphization degree of Li3BO3, Li4SiO4
nanoparticles. Future work is to decrease the Li2CO3 content in products
and mix the Li3BO3 or Li4SiO4 with LiPO3.
国内学会