The relentless pursuit of higher energy density, enhanced safety, and longer cycle life continues to drive innovation in li ion battery technology. Among the critical internal components, the separator plays a pivotal yet often understated role. It is a porous membrane positioned between the cathode and anode, whose primary functions are to prevent physical contact of the electrodes (thus avoiding short circuits) while facilitating the unimpeded transport of lithium ions. The ideal li ion battery separator must therefore exhibit a delicate balance of properties: high porosity and electrolyte wettability for optimal ionic conductivity, excellent mechanical strength to withstand winding and assembly stresses, superior thermal and dimensional stability to endure operational heat, and appropriate pore size to block electrode particulates.
Conventional polyolefin separators, such as polyethylene (PE) and polypropylene (PP), dominate the market due to their chemical stability, mechanical properties, and low cost. However, their inherent hydrophobicity leads to poor electrolyte uptake and retention, which can increase interfacial resistance. Furthermore, their relatively low melting points pose a significant safety risk, as thermal shrinkage at elevated temperatures can lead to internal short circuits. To overcome these limitations, extensive research has focused on modifying or replacing these membranes. Electrospinning has emerged as a powerful technique for creating non-woven fibrous separators. These membranes, composed of interwoven submicron or nanoscale fibers, boast a highly porous, interconnected three-dimensional network. This structure offers exceptional porosity (often exceeding 80%), high electrolyte uptake, and a large surface area, all of which are highly beneficial for ion transport in a li ion battery. Polymers like Poly(vinylidene fluoride) (PVDF) and its copolymers are frequently chosen as the fiber matrix due to their strong dipole moment, high dielectric constant, good electrochemical stability, and affinity for common carbonate-based electrolytes.
Despite its advantages, the electrospinning process for separator fabrication faces challenges for large-scale industrialization, primarily related to low production yield and fiber brittleness when very thin membranes are required. Conversely, electro-spraying, a sibling technology to electrospinning, operates with lower viscosity solutions to generate fine particles or droplets rather than continuous fibers. It has been used to deposit functional coatings onto commercial separators to enhance performance. However, sequentially combining electrospinning and electro-spraying processes adds complexity and may lead to pore blockage or reduced porosity in the final composite. Inspired by the concept of coaxial electrospinning—where two different polymer solutions are ejected simultaneously from concentric needles to form core-shell fibers—we conceived and developed an integrated, one-step fabrication strategy: coaxial electro-spraying/electrospinning (CEES). This work details the design principle of the CEES apparatus, the optimization of its operational parameters, and its successful application in creating a novel, high-performance composite separator for li ion battery applications. This technique uniquely allows for the in-situ, simultaneous deposition of a nanofibrous network and functional particles, creating a synergistic hierarchical structure directly onto a substrate.
1. Principle and Apparatus Design of Coaxial Electro-Spraying/Electrospinning
The fundamental principle unifying both electrospinning and electro-spraying is the manipulation of a viscoelastic fluid jet using a high-voltage electrostatic field. When a sufficiently high voltage is applied to a polymer solution droplet at the tip of a capillary, the electrostatic repulsion forces overcome the surface tension, deforming the droplet into a conical shape known as a Taylor cone. From the apex of this cone, a charged jet is ejected. The subsequent behavior of this jet determines the output morphology. For electrospinning, a high polymer concentration and molecular weight result in sufficient chain entanglement, preventing the jet from breaking up and allowing it to thin and whip continuously until solid fibers are deposited on the collector. The governing forces can be described by a balance between electrostatic stretching and viscoelastic resistance. For electro-spraying, a lower concentration/viscosity solution leads to a jet that undergoes varicose instability and breaks into fine droplets, which solidify into particles.
The key distinction in our CEES approach is the controlled, simultaneous initiation of both phenomena from a single nozzle assembly. The core design, as illustrated schematically, features a concentric dual-capillary system:
- Inner Nozzle (Core): Dedicated to the electrospinning process. It carries a high-viscosity “spinning dope” designed to form continuous nanofibers.
- Outer Nozzle (Shell): Dedicated to the electro-spraying process. It carries a low-viscosity “spraying solution” designed to atomize into micro- or nanoparticles.
The two nozzles are arranged co-axially with a precise gap between them. A critical design parameter is the protrusion length of the inner needle relative to the outer one. To minimize interaction between the two solutions before jet formation and to prevent the spraying solution from wetting and diluting the spinning dope at the tip, the inner nozzle is typically extended 1-2 mm beyond the outer nozzle. This configuration allows two independent, concentric Taylor cones to form under a shared high-voltage electric field. The inner cone emits a fiber-forming jet, while the surrounding annular cone emits a droplet-forming jet. The trajectories of the fibers and particles are influenced by the mutual electrostatic interactions and the converging electric field lines, leading to their co-deposition and intermingling on the grounded collector. This results in a composite membrane where micro/nano-particles are intimately integrated within or attached to a supporting nanofibrous matrix, creating a unique 3D porous architecture ideal for li ion battery separator function.
The process is governed by a complex interplay of parameters from both regimes. Key variables include:
- Solution Properties: Concentration, viscosity ($\eta$), surface tension ($\gamma$), electrical conductivity ($\sigma$), and solvent volatility for both the spinning and spraying solutions.
- Process Parameters: Applied voltage ($V$), flow rates for the inner and outer solutions ($Q_i$, $Q_o$), the inner diameter of both nozzles ($D_i$, $D_o$), and the tip-to-collector distance ($H$).
- Ambient Conditions: Temperature and relative humidity.
2. Experimental Methodology
2.1. Materials and Precursor Solutions
The core materials were selected based on their established relevance to li ion battery components. Poly(vinylidene fluoride) (PVDF, Mw ~534,000 g/mol) was chosen as the base polymer for both fibers and particles due to its excellent electrochemical stability, good mechanical properties, and high affinity for electrolyte solvents. Fumed silicon dioxide (SiO2, ~12 nm primary particle size) was selected as a functional filler to enhance thermal stability, mechanical strength, and electrolyte retention. N,N-Dimethylformamide (DMF) and acetone were used as solvents.
Two distinct precursor solutions were prepared:
- Electrospinning Solution (Inner Core): A 17 wt% PVDF solution was prepared in a binary solvent mixture of acetone and DMF (volume ratio 1:2). This relatively high concentration ensures sufficient chain entanglement for stable fiber formation without beads. The solution was magnetically stirred at 50°C for 4 hours and subsequently sonicated for 30 minutes to remove air bubbles.
- Electro-Spraying Solution (Outer Shell): A 4 wt% PVDF solution was prepared in pure DMF. To this, 0.1 wt% (relative to the total solution) of nano-SiO2 was added. This low concentration favors droplet break-up. The mixture was stirred vigorously for 2 hours and sonicated for 1 hour to achieve a homogeneous dispersion of SiO2 nanoparticles.
The physical properties of the spraying solutions at different PVDF concentrations were systematically characterized, as summarized in Table 1. The optimal concentration (4 wt%) was identified as it provided a balance of viscosity and conductivity that yielded the smallest, most uniform particles.
| PVDF Concentration (wt%) | Viscosity, $\eta$ (mPa·s) | Surface Tension, $\gamma$ (mN/m) | Electrical Conductivity, $\sigma$ ($\mu$S/cm) | Dominant Outcome |
|---|---|---|---|---|
| 1 | 15.8 | 26.1 | 4.24 | Very fine mist, poor film formation |
| 2 | 44.3 | 26.7 | 4.26 | Fine particles, some coalescence |
| 3 | 79.9 | 27.0 | 4.45 | Uniform particle formation |
| 4 | 272.0 | 27.6 | 6.47 | Optimal, smallest & most stable particles |
| 5 | 368.5 | 28.7 | 9.05 | Irregular droplets/particles, occasional clogging |
2.2. CEES Apparatus and Membrane Fabrication
A custom CEES setup was assembled. It consisted of two syringe pumps for independent control of the inner and outer solution flow rates. The coaxial nozzle was fabricated by fitting a stainless-steel blunt-tip needle (22-gauge, inner diameter $D_i$ = 0.41 mm) inside a larger needle (18-gauge, inner diameter $D_o$ = 0.84 mm), ensuring the inner needle protruded by approximately 1.5 mm. This assembly was connected to a high-voltage DC power supply (positive polarity). A flat aluminum plate covered with aluminum foil (or a commercial separator substrate) served as the grounded collector.
The optimized process parameters for fabricating the PVDF&PVDF/SiO2 composite membrane were determined empirically:
- Applied Voltage ($V$): +22 kV
- Tip-to-Collector Distance ($H$): 10 cm
- Inner Solution Flow Rate ($Q_i$): 0.8 mL/h
- Outer Solution Flow Rate ($Q_o$): 1.5 mL/h
- Ambient Humidity: 30 ± 5%
Under these conditions, a stable compound Taylor cone was observed, emitting a central fibrous jet surrounded by a fine spray of droplets. The as-prepared free-standing composite membrane was designated as “PVDF&PVDF/SiO2“.
For li ion battery testing, the CEES process was used to directly modify a commercial ultra-high molecular weight polyethylene (UHMWPE) separator (Celgard 2400). The UHMWPE membrane was fixed onto the aluminum collector, and the CEES process was performed for a defined duration (2-8 minutes) to deposit a thin, uniform layer of the PVDF&PVDF/SiO2 composite onto one side. This created a hierarchical composite separator (UHMWPE@PVDF&PVDF/SiO2).
2.3. Material Characterization and Electrochemical Testing
The morphology of the fabricated membranes was examined using scanning electron microscopy (SEM). Wettability was assessed by measuring the contact angle of a standard liquid electrolyte (1 M LiPF6 in EC/DMC/EMC, 1:1:1 by volume) using a sessile drop method. Porosity ($P$) was determined by a gravimetric method using n-butanol as the wetting liquid, calculated according to:
$$P (\%) = \frac{(m_2 – m_1) / \rho}{V} \times 100\%$$
where $m_1$ and $m_2$ are the weights of the dry and wet membrane, respectively, $\rho$ is the density of n-butanol, and $V$ is the geometric volume of the dry membrane.
Thermal dimensional stability was evaluated by measuring the shrinkage of membrane samples after being held at 90°C for 1 hour in an oven. The ionic conductivity ($\sigma_i$) of the separator saturated with electrolyte was calculated from electrochemical impedance spectroscopy (EIS) measurements using a symmetric stainless-steel (SS) cell configuration:
$$\sigma_i = \frac{d}{R_b \times A}$$
where $d$ is the membrane thickness, $R_b$ is the bulk resistance obtained from the high-frequency intercept on the real axis of the Nyquist plot, and $A$ is the contact area between the separator and the SS electrode.
The electrochemical performance was evaluated by assembling CR2032 coin cells. The cathode was composed of LiFePO4, conductive carbon, and PVDF binder (8:1:1 by weight) coated on an Al foil. Lithium metal was used as the anode. The cells were assembled in an argon-filled glovebox using the modified or pristine UHMWPE separator soaked with 80 µL of electrolyte. Galvanostatic charge-discharge cycling was performed between 2.5 V and 4.0 V vs. Li/Li+ at various C-rates (where 1C = 170 mA/g) at room temperature.
3. Results and Discussion: Optimization and Performance
3.1. Morphological Control via CEES Parameters
The concentric nozzle design is critical. Using an outer nozzle with an excessively large inner diameter ($D_o$ > 1.2 mm) resulted in poor atomization of the spraying solution, leading to large, flattened droplets that pooled on the collector. With an outer nozzle that was too small ($D_o$ < 0.7 mm), the gap between the inner and outer needles became too narrow, causing the two solutions to interact prematurely and often resulting in bead-on-string fiber morphology rather than distinct particles. An outer nozzle inner diameter ($D_o$) of 0.84 mm provided the optimal annular space for stable, independent jet formation from both channels.
SEM analysis confirmed the successful co-deposition of fibers and particles. The electrospun PVDF fibers from the inner nozzle formed a non-woven network with an average diameter of 423 ± 85 nm. Simultaneously, the electro-sprayed PVDF/SiO2 solution produced spherical and slightly elliptical particles with an average diameter of 2.01 ± 0.45 µm. The particles were not merely stacked on top of the fiber layer but were integrated within the fibrous matrix, often adhering to fiber junctions or lying within the pores. This intimate integration is a direct result of the in-situ deposition process, which promotes strong interfacial adhesion between the two phases—a significant advantage over post-fabrication coating methods. The hierarchical structure creates a multi-scale pore network: large inter-fiber pores (several micrometers) for electrolyte reservoir function and smaller intra-particle or fiber/particle interstitial spaces (sub-micrometer) that can enhance capillary action for electrolyte retention.
3.2. Properties of the Modified Li-Ion Battery Separator
The CEES modification profoundly altered the properties of the commercial UHMWPE separator. The performance metrics for different deposition times are consolidated in Table 2.
| CEES Time (min) | Thickness (µm) | Porosity, $P$ (%) | Electrolyte Contact Angle (°) | Longitudinal Shrinkage at 90°C (%) | Ionic Conductivity, $\sigma_i$ (mS/cm) |
|---|---|---|---|---|---|
| 0 (Pristine) | 22 | 37.85 | 48.55 | 7.28 | 0.51 |
| 2 | 24 | 67.46 | 28.12 | 6.32 | 0.89 |
| 4 | 25 | 85.96 | 14.93 | 5.26 | 1.24 |
| 6 | 27 | 84.40 | 10.45 | 4.21 | 1.18 |
| 8 | 31 | 52.39 | 8.04 | 3.68 | 0.95 |
Porosity and Wettability: The most striking improvement is in porosity. A 4-minute CEES treatment increased the porosity from 37.85% to 85.96%. This is attributed to the highly porous nature of the deposited electrospun nanofiber web. The accompanying dramatic drop in contact angle—from 48.55° (hydrophobic) to 14.93° (super-hydrophilic)—stems from two factors: the inherent polarity of the PVDF material and the drastically increased surface roughness and capillary forces introduced by the nano/micro structure. Enhanced wettability ensures rapid and complete electrolyte filling, which is crucial for reducing activation time and ensuring uniform current distribution in a li ion battery.
Thermal Dimensional Stability: The PVDF&PVDF/SiO2 composite layer acts as a thermally resistant scaffold. While the underlying UHMWPE substrate still tends to shrink upon heating, the deposited composite layer constrains this movement. The longitudinal shrinkage decreased from 7.28% to 5.26% after 4 minutes of modification. The incorporated SiO2 nanoparticles further enhance this thermal stability. This improvement directly addresses a key safety concern for li ion battery separators.
Ionic Conductivity: The high porosity and superior wettability synergistically lead to a significant increase in ionic conductivity ($\sigma_i$), a direct measure of a separator’s ability to facilitate ion transport. The $\sigma_i$ value more than doubled, from 0.51 mS/cm for the pristine separator to 1.24 mS/cm for the 4-minute modified sample. This can be modeled by considering the effective ion transport pathway, which is influenced by the membrane’s tortuosity ($\tau$) and porosity:
$$\sigma_i \propto \frac{\epsilon}{\tau}$$
where $\epsilon$ is the porosity. The CEES-derived structure creates a highly porous ($\epsilon \uparrow$) and interconnected network with low tortuosity ($\tau \downarrow$), maximizing $\sigma_i$.
It is important to note the non-monotonic trend with increasing CEES time. An optimal time (4 min) exists. Shorter times yield insufficient coating for full benefit. Longer times (8 min) lead to excessive deposition that begins to fill and block the pores of the underlying fibrous layer, causing a decrease in porosity and ionic conductivity, despite continued improvement in wettability and thermal shrinkage. This highlights the importance of process optimization to achieve a balanced structure.
3.3. Electrochemical Performance in Li-Ion Battery Cells
The ultimate test of the CEES-modified separator is its performance in a functional li ion battery. Cells using LiFePO4 cathodes and lithium metal anodes were constructed. Figure 1a shows the initial charge-discharge profiles at 0.1C. The cell with the 4-minute modified separator exhibited the highest initial discharge specific capacity of 122.4 mAh/g, compared to 107.8 mAh/g for the cell with the pristine UHMWPE separator. This 13.5% increase is a direct consequence of the lower internal resistance and more efficient lithium-ion transport enabled by the modified separator.
The rate capability and cycling stability are critical for practical li ion battery applications. Cells were cycled at progressively higher C-rates (0.2C, 0.5C, 1C, 2C) before returning to 0.2C. The cell with the optimal composite separator demonstrated superior capacity retention at high rates, recovering nearly all of its initial capacity when the rate was lowered back to 0.2C. This indicates excellent reversibility and minimal polarization.
Long-term cycling performance at 0.5C is shown in Figure 1b. The cell with the pristine UHMWPE separator suffered from significant capacity fade, retaining only ~82% of its initial capacity after 50 cycles. In contrast, the cell with the 4-minute CEES-modified separator showed a much more stable cycling profile, with a capacity retention of ~91% after 50 cycles. The enhanced cycle life can be attributed to several factors: (i) the stable composite interface reduces side reactions between the electrolyte and the separator, (ii) the excellent electrolyte retention maintains effective ionic pathways throughout cycling, and (iii) the improved thermal/dimensional stability maintains electrode-separator integrity. Post-mortem SEM analysis of the cycled separator confirmed that the fibrous/particulate structure remained largely intact without significant collapse or pore blockage, unlike the pristine separator which showed signs of deformation.
| Separator Type | Initial Discharge Capacity at 0.1C (mAh/g) | Capacity Retention at 1C (vs. 0.1C) (%) | Capacity Retention after 50 cycles at 0.5C (%) | Charge Transfer Resistance, $R_{ct}$ (Ω)* |
|---|---|---|---|---|
| Pristine UHMWPE | 107.8 | 71.5 | 82.3 | 148 |
| UHMWPE@PVDF&PVDF/SiO2 (4 min) | 122.4 | 88.2 | 91.1 | 78 |
*$R_{ct}$ value obtained from fitting the medium-frequency semicircle in EIS spectra of Li|separator|LiFePO4 cells.
4. Conclusion and Perspectives
This work successfully demonstrates the design, implementation, and application of a novel coaxial electro-spraying/electrospinning (CEES) technique for fabricating advanced composite separators. By integrating the fiber-forming and particle-forming electrostatic processes into a single, simultaneous operation, we have developed a method to create hierarchical membranes featuring an interpenetrating network of nanofibers and functional micro/nanoparticles. Applied to li ion battery technology, this technique enables the direct, in-situ modification of commercial polyolefin separators with a layer of PVDF&PVDF/SiO2 composite.
The optimized CEES process yields a composite separator with a remarkable set of synergistic properties: exceptionally high porosity (>85%), super-hydrophilic surface (contact angle <15°), enhanced thermal dimensional stability, and consequently, a high ionic conductivity (>1.2 mS/cm). When deployed in a LiFePO4/Li li ion battery, this separator contributes to a significant increase in initial discharge capacity, superior rate capability, and markedly improved cycling stability compared to its unmodified counterpart.
The CEES strategy presented here is highly versatile and promising. The choice of materials for both the fiber and particle phases can be tailored extensively. For instance, using gel polymer electrolytes as the spinning dope could lead to quasi-solid-state battery separators. Incorporating flame-retardant particles (e.g., Al(OH)3, polyphosphazene) into the spray solution could address safety concerns even more directly. Furthermore, this technique is not limited to energy storage. The ability to create tailored porous structures with controlled surface chemistry and multi-scale features makes CEES a compelling platform for applications in high-efficiency filtration, catalytic substrates, tissue engineering scaffolds, and advanced protective textiles. Future work will focus on scaling up the CEES process through multi-nozzle array design and exploring a wider library of functional material combinations to push the boundaries of performance in li ion battery and beyond.