Carbon-based material aqueous polyurethane dispersant, and preparation method and application thereof

Abstract

The application belongs to the field of new energy battery materials and high molecular dispersants, and particularly relates to a carbon-based material water-based polyurethane dispersant and a preparation method and application thereof. The method comprises the following steps: 1) under the action of a catalyst, isophorone diisocyanate is reacted with a hydroxyl-containing reactant in a protective atmosphere to synthesize a prepolymer, a reaction solution is added to the prepolymer to obtain a solution containing a highly branched chain-extending intermediate; and 2) an end-capping agent is added to the solution containing the highly branched chain-extending intermediate to obtain a water-based polyurethane dispersant. The application adopts a highly branched polyurethane skeleton containing a pyrrolidone structure unit, and through the synergistic effect of interfacial anchoring, highly branched steric hindrance and a polyethylene glycol hydration segment, the dispersion stability of a carbon-based conductive agent in a water-based negative electrode slurry is improved; when a hydroxyl-terminated polybutadiene soft segment is preferably introduced, the stress buffering capacity of a conductive network in the cycle process of a silicon-carbon negative electrode can also be improved, so that the sheet resistance of a pole piece is reduced and the cycle capacity retention rate is improved.

Description

Technical Field

[0001] This invention belongs to the field of new energy battery materials and polymeric dispersants, specifically relating to carbon-based waterborne polyurethane dispersants, their preparation methods, and applications. Background Technology

[0002] With the continuous improvement of the energy density of lithium-ion batteries, silicon-carbon composite anodes have attracted widespread attention due to their high theoretical capacity (approximately 4200 mAh / g). However, silicon-based materials undergo massive volume expansion of over 300% during charging and discharging, which makes the conductive network inside the anode prone to breakage and collapse under cyclic stress, leading to rapid capacity decay.

[0003] Single-walled carbon nanotubes (SWCNTs) are considered ideal materials for constructing long-lasting conductive networks in silicon-carbon anodes due to their excellent one-dimensional electronic conductivity, high aspect ratio, and mechanical flexibility. Furthermore, in graphite anodes, SWCNTs can significantly improve the electronic conductivity and structural integrity of the electrode.

[0004] However, there is a strong van der Waals attraction between SWCNTs, which makes them very prone to re-aggregation, forming CNT bundles ranging from hundreds of nanometers to micrometers in size.

[0005] To suppress this tendency, existing technologies mainly employ the following dispersant systems, but all of them have unresolved drawbacks: For example, in the CMC / SBR system, the compound system of sodium carboxymethyl cellulose (CMC) and styrene-butadiene rubber (SBR) is the mainstream bonding scheme for graphite anodes. However, CMC itself has high thickening properties, and the viscosity of the slurry often increases sharply when coexisting with SWCNTs, which leads to a narrowing of the coating process window. At the same time, the dispersion effect of CMC on SWCNTs mainly relies on physical encapsulation and lacks interfacial chemical anchoring, resulting in significant re-agglomeration of CNTs after long-term storage. For example, in the physical blending system of polyvinylpyrrolidone (PVP), the pyrrolidone ring in the PVP molecule has a certain affinity for the surface of carbon materials, which can improve the dispersibility of SWCNT to a certain extent. However, PVP exists in the slurry system in a free state, and it is easy to migrate, wash off or redistribute during electrode drying and subsequent electrolyte wetting, resulting in loss of conductive network uniformity and irreversible increase in electrode resistance during cycling. In addition, the residue of free PVP may affect the electrochemical stability of the electrode / electrolyte interface. For example, linear polyurethane dispersants are used in some existing technologies. They provide steric hindrance through polyethylene glycol segments. However, the steric radius of linear topology is limited. For SWCNTs with high aspect ratio (>1000), their ability to suppress entanglement and collision is insufficient. At the same time, they lack specific interfacial anchoring groups with CNT surface. The dispersion stability mainly depends on weak electrostatic interaction, making it difficult to maintain long-term stability in slurries with high solids content (above 50%).

[0006] In summary, existing dispersant systems fail to simultaneously address the following technical issues: Firstly, while physically blended dispersants (such as PVP) can provide interfacial affinity, they lack chemical bonding and have poor long-term stability. Chemically bonded dispersants (such as conventional polyurethane) often lack specific anchoring groups for the surface of carbon nanotubes. Secondly, linear dispersants have insufficient steric hindrance. Although highly branched or cross-linked dispersants increase steric hindrance, they are prone to causing uncontrolled slurry viscosity or gelation due to excessively dense topological structures. Thirdly, the volume expansion during silicon-carbon anode cycling requires the conductive network to have flexible buffering capabilities, but the existing rigid dispersant framework cannot effectively absorb stress, causing the electrical contact between CNTs and active particles to be gradually lost during cycling. Finally, in complex aqueous anode slurries containing CMC, SBR, and lithium carboxylate groups, the dispersion mechanism relying solely on electrostatic repulsion is easily affected by the combined influence of the slurry's ionic environment, polymer chain conformation, and solid content, resulting in insufficient long-term stability. Therefore, there is an urgent need in this field for a dispersant system that combines chemically bonded interface anchoring, highly branched steric hindrance, flexible stress buffering, and hydration stability to achieve long-term stable dispersion of SWCNTs in silicon-carbon / graphite anode slurries and to construct a conductive network resistant to cyclic stress. Summary of the Invention

[0007] To address the challenges in coordinating the interfacial anchoring effect and structural stability of dispersants in existing technologies, the contradiction between the dispersion effect caused by steric hindrance and the rheological properties of the slurry, the relatively limited mechanical properties, and the tendency to cause dispersion failure in complex aqueous negative electrode slurries containing CMC, SBR, active materials, and neutralizing salt groups, this invention provides a novel carbon-based aqueous polyurethane dispersant, along with its preparation method and applications.

[0008] The main objective of this invention is: I. A highly branched polyurethane dispersant with pyrrolidone structural units chemically bonded in its molecular structure is provided, which fixes the pyrrolidone ring to the polyurethane skeleton through covalent bonds. This is different from the free PVP in physical blends and significantly reduces the risk of migration or elution of free pyrrolidone dispersion components during electrode drying and long-term cycling. II. A polyurethane skeleton with a highly branched topology is provided. Through the synergistic construction of a trifunctional branching agent and a carboxyl-containing chain extender, a compact random branched conformation is formed, which significantly increases the steric radius and suppresses close-range collisions and entanglement of carbon-based materials. Third, through the synergy of polyethylene glycol hydration segments and carboxyl / carboxylate groups, a dispersion mechanism dominated by hydration layer stability is provided in complex aqueous negative electrode slurries containing CMC, SBR, active materials and neutralizing salt groups, making up for the stability gap after electrostatic repulsion is compressed.

[0009] To achieve the above objectives, the present invention adopts the following technical solution.

[0010] Preparation method of waterborne polyurethane dispersant for carbon-based materials. The method includes: 1) In a protective atmosphere, isophorone diisocyanate is reacted with hydroxyl-containing reactants to synthesize NCO-terminated prepolymers under the action of a catalyst. Chain extension / branching reaction solution is added to the NCO-terminated prepolymers to carry out chain extension and branching reactions to obtain a solution containing highly branched chain extension intermediates. 2) Add an end-capping agent to the solution containing highly branched chain-extending intermediates to perform end-capping treatment, thereby obtaining an aqueous polyurethane dispersant.

[0011] As a preferred option Step 1) The hydroxyl-containing reactants include dimethylolpropionic acid and / or hydroxyl-terminated polybutadiene; Step 1) The catalyst comprises dibutyltin dilaurate; Step 1) The synthesized NCO-terminated prepolymer includes: The hydroxyl-containing reactant was dissolved in DMF, followed by the addition of isophorone diisocyanate and a catalyst. The mixture was then placed in a protective atmosphere and reacted at 75–85 °C for 2–3 h to obtain the NCO-terminated prepolymer.

[0012] As a preferred option Step 1) The amount of the hydroxyl-containing reactant used is 0.075–2.500 g / g isophorone diisocyanate; Step 1) The amount of catalyst used is 1.0 to 2.0 mg / g isophorone diisocyanate.

[0013] As a preferred option When using the dimethylolpropionic acid, the dosage is 0.075–0.225 g / g isophorone diisocyanate; The hydroxyl-terminated polybutadiene has a hydroxyl value of 115–135 mgKOH / g, and the dosage used is 2.0–2.5 g / g isophorone diisocyanate. The amount of dibutyltin dilaurate used is 1.0 to 2.0 mg / g isophorone diisocyanate.

[0014] As a preferred option The chain extension / branching reaction solution includes trimethylolpropane, dimethylolpropionic acid, and hydroxyl reactive monomers containing pyrrolidone structures.

[0015] As a preferred option The hydroxyl reactive monomers containing pyrrolidone structures include N-(2-hydroxyethyl)pyrrolidone and / or hydroxyl-terminated polyvinylpyrrolidone; in: When using the N-(2-hydroxyethyl)pyrrolidone, the dosage is 0.1–0.3 g / g isophorone diisocyanate; When using the hydroxyl-terminated polyvinylpyrrolidone, the dosage is 0.1–0.3 g / g isophorone diisocyanate.

[0016] As a preferred option The amount of dimethylolpropionic acid used is 0.075–0.225 g / g isophorone diisocyanate; The amount of trimethylolpropane used is 0.05–0.15 g / g isophorone diisocyanate.

[0017] As a preferred option Step 1) involves adding the chain extension / branching reaction solution to prepare a solution by dissolving the chain extension / branching reaction solution in DMF. Then, the chain extender solution is slowly added at a rate of 0.75–2.00 % vol / min. After the addition is complete, the reaction is carried out at 73–78 ℃ for 2.5–3.5 h to obtain a solution containing highly branched chain extension intermediates.

[0018] As a preferred option Step 2) The capping agent is methoxy-terminated polyethylene glycol; The methoxy-terminated polyethylene glycol has an average molecular weight of 1600–2200 and a hydroxyl value of 25–35 mgKOH / g; The amount of methoxy-terminated polyethylene glycol used is 0.2–0.3 g / g isophorone diisocyanate.

[0019] As a preferred option Step 2) The end-capping treatment is carried out at 68-72 °C for 1-2 h.

[0020] As a preferred option After the end-sealing process in step 2) is completed, a neutralization process is performed; The neutralization treatment involves neutralizing the carboxyl group to a carboxylate, and determining the degree of neutralization by back titration and / or potentiometric titration until the degree of neutralization is 80-90%.

[0021] As a preferred option The neutralization treatment is performed by titration with a 5-10 wt% lithium hydroxide aqueous solution, with continuous stirring during the titration process.

[0022] Aqueous polyurethane dispersant for carbon-based materials.

[0023] Waterborne polyurethane dispersants for carbon-based materials The molecular structure of the carbon-based aqueous polyurethane dispersant comprises the following chain segment units: End groups, side groups, or grafted segments containing pyrrolidone structures; And, a highly branched polyurethane skeleton constructed from diisocyanate, trifunctional branching agent and carboxyl chain extender; And, covalently bonded to the polyethylene glycol segments at the ends of the polyurethane chains.

[0024] As a preferred option The carbon-based waterborne polyurethane dispersant also contains hydroxyl-terminated polybutadiene soft segments.

[0025] Application of waterborne polyurethane dispersants for carbon-based materials The carbon-based waterborne polyurethane dispersant is used to formulate carbon-based conductive agent slurry. The carbon-based conductive agent includes single-walled carbon nanotubes and / or multi-walled carbon nanotubes and / or graphene and / or conductive carbon black and / or carbon nanofibers.

[0026] It should be noted that the research in this invention was initially based on single-walled carbon nanotubes (SWCNTs), but the research process showed that it has similar effectiveness for other carbon-based materials such as multi-walled carbon nanotubes (MWCNTs) and graphite. Therefore, the following discussion will focus on carbon-based materials, primarily SWCNTs, but the scope of protection claimed by this invention should not be limited by the following content.

[0027] The stabilizing effect of this invention on the conductive network of carbon nanotubes does not rely on a simple amplification of a single mechanism. Instead, it employs a hierarchical design of the molecular structure to decouple and recouple four physicochemical mechanisms at the nanoscale, forming a synergistic system of anchoring, steric hindrance, buffering, and hydration. From the surface of the carbon nanotube outwards, each layer sequentially undertakes a different stabilizing function: the pyrrolidone structural unit is responsible for interface anchoring, the highly branched polyurethane framework is responsible for steric hindrance, the optional hydroxyl-terminated polybutadiene soft segments are responsible for stress buffering, and polyethylene glycol and optional carboxyl groups are responsible for hydration stabilization. The mechanisms of each layer are described in detail below.

[0028] First, the surface of single-walled carbon nanotubes is composed of sp 2 It consists of a graphite lattice composed of hybrid carbon atoms, with a highly delocalized π-electron cloud and a hydrophobic surface.

[0029] In an aqueous environment, carbon nanotubes spontaneously aggregate due to strong van der Waals attraction, reducing their contact area with water molecules.

[0030] The pyrrolidone ring is a five-membered nitrogen-containing heterocycle with a molecular dipole moment significantly higher than that of common organic functional groups. The synergistic effect between the amide carbonyl group and the lone pair electrons of the nitrogen atom in the ring gives it strong polarity.

[0031] When the pyrrolidone ring approaches the surface of the carbon nanotube, two types of interactions can occur: First, a directional dipole-π interaction is formed between the strong dipole and the delocalized π electron cloud on the surface of the carbon nanotube. Although this interaction is weaker than the covalent bond, it is significantly stronger than the ordinary van der Waals force and has a certain degree of directional selectivity. It can competitively replace water molecules in the aqueous phase and achieve interface anchoring. Secondly, the methylene skeleton of the pyrrolidone ring is hydrophobic, and can form a thermodynamically favorable hydrophobic association with the hydrophobic surface of carbon nanotubes.

[0032] The synergistic effect of the two aforementioned actions results in a much higher affinity of the pyrrolidone ring for the carbon nanotube surface compared to conventional polyether or polyester segments.

[0033] In the prior art, although physically blended polyvinylpyrrolidone (PVP) can interact with carbon nanotubes through the pyrrolidone ring, PVP exists in a free state in the continuous phase. Its adsorption and desorption are in dynamic equilibrium and are significantly affected by concentration, temperature, shear and electrolyte composition.

[0034] During electrode drying, the concentration gradient caused by solvent evaporation can drive PVP to migrate to the electrode surface or the interface of active particles; during long-term cycling, the swelling and extraction of free polymers by the electrolyte further exacerbates their elution. This dynamic instability is the fundamental cause of conductive network collapse.

[0035] The present invention uses N-(2-hydroxyethyl)pyrrolidone (HEP) or hydroxyl-terminated polyvinylpyrrolidone (PNVP-OH) to attach pyrrolidone structural units to the polyurethane backbone or chain end.

[0036] Compared to physical blending, the above-mentioned access method reduces the risk of free migration of pyrrolidone structural units during electrode drying and long-term cycling. Physically blended PVP exhibits dynamic and reversible adsorption, and can migrate or redistribute driven by concentration gradients. In contrast, in this invention, pyrrolidone structural units are fixed in polyurethane molecules through covalent linkages such as urethane bonds, which is more conducive to forming stable interfacial anchorage.

[0037] Furthermore, HEP and PNVP-OH provide two different pathways for introducing pyrrolidone structures. The HEP route provides a single pyrrolidone ring structure, while the PNVP-OH route provides a short-linked branched structure containing multiple pyrrolidone rings. Under similar conditions, multi-point adsorption structures generally have a higher desorption energy barrier, which is beneficial for improving long-term dispersion stability.

[0038] On the other hand, the re-aggregation process of single-walled carbon nanotubes is controlled by two time scales: In the rapid phase from milliseconds to seconds, carbon nanotubes undergo end-to-end or side-to-side contact due to van der Waals forces, forming primary aggregates; In the slow phase, which lasts from minutes to hours, primary aggregates grow into larger flocs through collisions and coalescence.

[0039] The steric hindrance of dispersants mainly acts in the rapid phase, inhibiting the formation of primary aggregates by increasing the energy barrier when carbon nanotubes approach. For single-walled carbon nanotubes with high aspect ratios, their effective collision cross-section is extremely large, and the steric hindrance layer thickness of linear dispersants is often insufficient to cover the entire collision cross-section.

[0040] Therefore, this invention employs a trifunctional branching agent and a difunctional carboxyl-containing chain extender to synergistically construct a polyurethane backbone. According to branching theory, the polycondensation system of trifunctional and difunctional monomers can form polymers with random branched topology, achieving a high degree of branching, but not yet reaching the threshold of typical hyperbranched polymers. This choice balances steric hindrance and process feasibility; sufficient branching ensures that the polymer chains exhibit a compact random coil conformation in solution, rather than the extended chain conformation of nonlinear polymers, while avoiding the risks of intramolecular crosslinking and gelation caused by excessive branching.

[0041] At the same molecular weight, highly branched polymers typically have smaller hydrodynamic radii than linear polymers, but occupy a larger volume. This is because branching topology causes the polymer chain ends to extend outward, forming a radial structure. When multiple dispersant molecules of this conformation adsorb onto the surface of carbon nanotubes, their terminal polyethylene glycol segments, together with the branched framework, form a steric hindrance layer tens of nanometers thick, significantly increasing the nearest approach distance between carbon nanotubes.

[0042] The steric hindrance mechanism of this invention is primarily kinetic steric hindrance. When two carbon nanotubes adsorbed with dispersant approach each other, the chain segments of the steric hindrance layer first overlap. The polymer chain segments within the overlap region exhibit osmotic repulsion due to a decrease in conformational entropy. The rigid nodes of the highly branched framework further restrict the compressive deformation of the chain segments, giving the steric hindrance layer elastic recovery capability. Compared to the flexible brush steric hindrance of linear dispersants, this elastic steric hindrance maintains a more stable layer thickness during shearing or collision processes.

[0043] Linear polyurethane dispersants exhibit random coil or extended chain conformations in solution, and their steric hindrance layer thickness is strongly dependent on solvent mass and ionic strength. In high-solids slurries, linear chains are prone to intermolecular entanglement, leading to steric hindrance layer compression or even failure. In contrast, the compact conformation of highly branched topologies is less dependent on solvent mass and ionic strength, and its steric hindrance layer thickness is mainly determined by molecular topology, thus exhibiting more stable dispersion performance in complex slurry environments.

[0044] In addition, it is well known that the volume change rate of silicon-carbon composite anodes during lithium insertion and delithiation can reach 250-300%, leading to cracking of active particles, breakage of conductive networks and interface peeling.

[0045] In the prior art, carbon nanotube conductive networks are usually fixed to the surface of active particles by rigid adsorption or physical winding. When silicon particles expand, carbon nanotubes are pulled apart or peeled off the surface due to lack of deformation margin, resulting in a continuous increase in electrode resistance during cycling.

[0046] The present invention introduces hydroxyl-terminated polybutadiene soft segments. The glass transition temperature of the polybutadiene backbone is about tens of degrees below zero. It remains in a highly elastic state within the range of room temperature and battery operating temperature. When hydroxyl-terminated polybutadiene is copolymerized with isophorone diisocyanate, dimethylolpropionic acid and trimethylolpropane hard segments, microphase separation occurs in the system due to thermodynamic incompatibility, forming nanoscale soft segment phase regions and hard segment phase regions.

[0047] In traditional rigid dispersant systems, the expansion stress of silicon particles is directly transmitted to the interface between carbon nanotubes and active particles, leading to carbon nanotube fracture. In the system of this invention, the stress transmission path is redesigned. The expansion stress of silicon particles first acts on the soft segment phase region of hydroxyl-terminated polybutadiene. This low-modulus phase region absorbs stress through high elastic deformation, which helps reduce the local stress transmitted to the pyrrolidone anchoring layer and the carbon nanotube conductive network, thereby improving the cycle integrity of the conductive network.

[0048] When the hydrophobic backbone of hydroxyl-terminated polybutadiene comes into contact with the electrolyte, a transition interface can be formed on the surface of the carbon nanotubes.

[0049] This interface does not hinder lithium-ion transport and reduces irreversible decomposition of the electrolyte on the carbon nanotube surface, maintaining flexible electrical contact between the carbon nanotubes and the active particles. However, it is important to note that the introduction of hydroxyl-terminated polybutadiene increases the hydrophobicity of the system; therefore, its dosage needs to be optimized in conjunction with the hydration segments of polyethylene glycol to ensure the overall water solubility of the dispersant.

[0050] In addition to the above aspects, lithium-ion battery anode slurry is not a simple carbon nanotube aqueous dispersion system, but a complex aqueous system containing carbon-based conductive agents, binders, thickeners, active materials and neutralizing salt groups.

[0051] In this complex aqueous system, electrostatic repulsion may be weakened by the ionic environment and high solids content. Therefore, dispersants that rely solely on electrostatic repulsion are unlikely to maintain a stable dispersion state during long-term storage, coating, and cycling.

[0052] This invention provides outer hydration stability through methoxy-terminated polyethylene glycol (PEG) segments. The ether oxygen atoms of the PEG segments possess lone pairs of electrons, which can act as hydrogen bond acceptors to form hydrogen bonds with water molecules. When using methoxy-terminated PEG with an average molecular weight of approximately 1600–2200, the PEG segments can bind water molecules through hydrogen bonding to form a hydration shell. This hydration shell helps reduce the probability of direct contact between carbon nanotubes and improves the stability of the dispersion system in complex aqueous slurries.

[0053] The water molecules in this hydration shell are not free water, but structural water that forms a dynamic hydrogen bond network with the polyethylene glycol chains. When two carbon nanotubes adsorbed with polyethylene glycol approach each other, the overlap of the hydration shells requires the disruption of this hydrogen bond network, a process that is thermodynamically unfavorable and kinetically slow.

[0054] Therefore, the polyethylene glycol hydration layer provides a kinetic barrier, rather than a simple thermodynamic repulsion barrier. Unlike electrostatic repulsion, the stability of the polyethylene glycol hydration layer mainly depends on the hydrogen bond strength and the segment conformational entropy, and is less dependent on the ionic strength of the solution. Even in high ionic strength environments, the hydration shell can still be maintained through local hydrogen bonding.

[0055] This invention introduces a carboxyl group via dimethylolpropionic acid, which can optionally be neutralized to form a carboxylate. In a high ionic strength environment, the conventional electrostatic interaction of the carboxyl group is significantly suppressed, but its strong hydration becomes the main stabilizing mechanism.

[0056] Furthermore, the high charge density of carboxylate ions allows them to form a primary hydration sheath with 4 to 6 water molecules through ion-dipole interactions. This hydration sheath, along with the outer hydration layer of polyethylene glycol, forms a synergistic hydration gradient. The inner layer is a strongly hydration sheath of carboxylate ions with oriented water molecules, the middle layer is a weakly hydration layer of polyurethane framework and pyrrolidone rings, and the outer layer is a hydrogen-bonded hydration network of polyethylene glycol. This gradient hydration structure creates a continuous hydration barrier on the carbon nanotube surface, from strong to weak and from oriented to dynamic, effectively preventing direct contact during collisions.

[0057] It is particularly important to note that the present invention employs a post-neutralization method, preferably with a neutralization degree of 80-90% rather than 100%. This is mainly to retain 10-20% of the carboxyl groups, which can form intramolecular and intermolecular hydrogen bonds with the amino or carbonyl oxygen in the polyurethane skeleton, thereby enhancing the cohesive force and mechanical strength of the skeleton. This avoids the ion shielding effect that may be caused by excessive lithium ions in the electrolyte during 100% neutralization, i.e., excessive counterions compressing the hydration layer. The presence of some carboxyl groups can regulate the acid-base responsiveness of the dispersant in the slurry, maintaining a suitable dissociation balance in a weakly acidic environment.

[0058] The core of this invention lies not only in the simultaneous introduction of four dispersion mechanisms, but also in the hierarchical covalent design of molecular structures, which enables each mechanism to work synergistically at its optimal scale, producing a nonlinear overall enhancement effect.

[0059] Through hierarchical molecular structure design, four mechanisms—pyrrolidone interface anchoring, highly branched steric hindrance, hydroxyl-terminated polybutadiene flexible buffering, and synergistic hydration of polyethylene glycol and carboxylate—are integrated into a single dispersant molecule. These mechanisms are connected by a covalent framework, forming a synergy across four dimensions: thermodynamic anchoring, kinetic steric hindrance, mechanical decoupling, and kinetic stability. This effectively solves the technical problems of insufficient long-term stability, failure in high-ionic-strength environments, and cyclic stress fracture of silicon-carbon dispersants in carbon nanotube conductive slurries. This synergistic system enables carbon nanotubes to maintain a micro-nano-scale dispersion in anode slurries with over 50% solids content, and maintains the integrity of the conductive network under the cyclic volume expansion stress of silicon-carbon anodes.

[0060] The beneficial effects of this invention are as follows: 1) In this invention, pyrrolidone structural units are chemically bonded to a polyurethane skeleton through covalent bonds to form a well-defined anchoring layer. Compared with physical blended PVP, this structure is less prone to free migration or elution during electrode drying and long-term cycling, and the electrode resistance remains stable during cycling, thus avoiding irreversible collapse of the conductive network. 2) The highly branched polyurethane skeleton provides significantly better steric hindrance than linear topology due to its compact random branched conformation. 3) When HTPB soft segments are introduced, its low-modulus, high-elasticity interface phase can effectively absorb the volume expansion stress of silicon-carbon particles, preventing the loss of electrical contact between carbon nanotubes and active particles during cycling. 4) Through the synergistic hydration of PEG hydration segments and optional carboxyl / carboxylate groups, good dispersion stability is maintained in complex aqueous negative electrode slurries containing CMC, SBR, active materials and neutralizing salt groups. Attached Figure Description

[0061] Figure 1 The FTIR spectra of samples with different degrees of neutralization in Example 1; Figure 2 The GPC molecular weight distribution curves for the 0% neutralization sample and the 85% neutralization sample in Example 1 are shown. Figure 3 DLS particle size distribution curves of CNT predispersants prepared from samples with 0%, 85%, and 95% neutralization in Example 1; Figure 4 The CNT adsorption isotherms and Langmuir fitting curves are for the 0%, 85%, and 95% neutralization samples in Example 1. Figure 5 The 7-day viscosity tracking curves are for the slurries prepared from the 0%, 85%, and 95% neutralization samples in Example 1. Figure 6 The graph shows a comparison of sheet resistance and peel strength of the electrodes made from the 0%, 85%, and 95% neutralization samples in Example 1. Figure 7 The cycle capacity retention curves of coin half-cells prepared from samples with different degrees of neutralization in Example 1; Figure 8 The image shows the FTIR spectrum of the sample from Example 2. Detailed Implementation

[0062] The present invention will be further described clearly and in detail below with reference to specific embodiments and the accompanying drawings. Those skilled in the art will be able to implement the present invention based on these descriptions. Furthermore, the embodiments of the present invention described below are generally only some, not all, of the embodiments of the present invention. Therefore, all other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort should fall within the scope of protection of the present invention.

[0063] Unless otherwise specified, in actual industrial production, the end-capping step in Step 3 of the embodiments and comparative examples of this invention can be performed by using the di-n-butylamine method to determine the residual NCO content of the system, and adding methoxy-terminated polyethylene glycol according to the residual NCO equivalent, so that the molar ratio of the hydroxyl groups in the methoxy-terminated polyethylene glycol to the residual NCO is controlled at 0.95 to 1.10:1 to achieve residual NCO end-capping and reduce material waste. The embodiments and comparative examples of this invention only show the amount added in specific experiments, and the results all meet the requirement of reducing the NCO content to below 0.3 wt%.

[0064] Unless otherwise specified, all raw materials used in the embodiments of the present invention are commercially available or obtainable by those skilled in the art; unless otherwise specified, all methods used in the embodiments of the present invention are methods mastered by those skilled in the art, and some characterization diagrams highlight key and effective information, and are displayed in segments using software during export.

[0065] Example 1

[0066] A carbon-based waterborne polyurethane dispersant is prepared by the following method: Step 1: Prepolymer Formation Dissolve 4.50 g DMPA in 40 mL anhydrous DMF, add 20.0 g IPDI and 0.03 g DBTDL, purge with nitrogen, and react at 80 °C for 2.5 h to form a prepolymer.

[0067] Step 2: High-branching chain extension Dissolve 3.50 g DMPA and 2.5 g TMP in 40 mL anhydrous DMF to prepare a mixed solution. Add the solution slowly and uniformly dropwise to the prepolymer prepared in Step 1 over 2 h. After the addition is complete, react at 75 °C for 3.0 h.

[0068] Step 3: PEG end-capping 5.0 g of MPEG was added to the Step 2 product, and the reaction was carried out at 70 °C for 90 min. The residual NCO content was titrated using the di-n-butylamine method, and the results showed that the NCO content decreased to below 0.3 wt%. The sample was retained as a 0% neutralization sample (i.e., unneutralized).

[0069] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution, prepare and retain samples with neutralization degrees of 75%, 80%, 85%, 90%, 95% and 100% respectively (determined by potentiometric titration), stir continuously for 2 h, discharge, remove DMF by rotary evaporation, dry under vacuum at 60 °C and pulverize to obtain the product HB-PU-PEG.

[0070] The unneutralized sample, the sample with 85% neutralization, and the sample with 100% neutralization obtained in this example were characterized by FTIR. The characterization results are as follows: Figure 1 As shown.

[0071] In the carbonyl region, Figure 1 All three spectra show a height of 1725 cm. -1 The peak (urethane C=O internal standard peak) is attributed to the carbonyl stretching vibration of the urethane bond formed by the reaction of IPDI with DMPA / TMP / MPEG in the sample. This peak position is basically stable at 0%, 85%, and 100% neutralization, with slight differences in intensity, mainly affected by sample preparation thickness and baseline correction. It can be used as an internal standard for comparing the relative intensities of other peaks.

[0072] And for 1705 cm -1 The peak is strong in the unneutralized sample, and is located at 1725 cm⁻¹. -1The peaks exhibit a distinct "double carbonyl" characteristic, with a peak spacing of approximately 20 cm. -1 The peak is attributed to the C=O stretching vibration of the carboxylic acid dimer formed by the carboxyl group of DMPA, which is the most typical structural fingerprint of unneutralized samples. In the 85% neutralization sample, this peak is significantly weakened but still identifiable, indicating that there is still a residual amount of carboxyl groups in the system that have not been converted into carboxylate, consistent with the 85% neutralization determined by potentiometric titration. In the 100% neutralization sample, this peak almost disappears, and the residual intensity is less than 5% of the baseline noise, proving that the carboxyl groups have been almost completely converted into carboxylate groups.

[0073] As for COO - Antisymmetric stretching peak (1600 cm⁻¹) -1 For unneutralized samples, this position is at baseline level with no absorption, indicating the absence of carboxylates in the sample. For samples with 85% neutralization, a faint shoulder peak or bulge appears at this position, indicating the presence of some COO. - It was generated, but the concentration was still low, and the peak shape was not fully developed. However, in the 100% neutralized sample, this peak became an independent and sharp strong peak, with a peak at 1540 cm⁻¹. -1 The amide bands are clearly separated. Due to the higher lithium ion field, the lithium carboxylate peak is typically located at 1600 cm⁻¹. -1 Nearby, the peaks are wider and stronger.

[0074] In the hydrogen-bonded region, the focus is on observing the 3400 cm⁻¹ region. -1 The unneutralized sample exhibits a typical broad, dome-shaped peak with the largest half-peak width. This peak is a superposition of NH stretching and OH stretching of the carboxylic acid dimer. Due to the strong hydrogen-bonded dimer network formed by the DMPA carboxyl groups, OH stretching occurs between 2500 and 3300 cm⁻¹. -1 The extremely wide absorption plateau in the region contributes to the overall peak broadening and baseline elevation. However, the peak shape of the 85% neutralized sample is in a transitional state, with a narrowing at the base and a relatively sharp peak tip, labeled as NH stretching + residual COOH. At this point, most of the carboxylic acid dimers have disintegrated, but the remaining 15% carboxyl groups still contribute some background broadening. For the completely neutralized 100% neutralized sample, the peak shape is the sharpest, and the half-maximum width at half-maximum is the narrowest. At this point, 3400 cm⁻¹... -1 The main evidence is the stretching of NH in urethane and the complete disappearance of the broadening platform of the carboxylic acid dimer, proving that the hydrogen bond network has been completely broken down.

[0075] Furthermore, 1461 cm can be observed during the neutralization process. -1 Peak and 1240 cm -1 The complete retention of these two characteristic peaks proves that the hydrocarbon framework was fully preserved during neutralization, and the polyurethane framework was effectively preserved. However, one characteristic peak requires special attention: the 1108 cm⁻¹ peak present in the unneutralized sample. -1The peak, and the samples with 85% and 100% neutralization after neutralization, are both located at 10¹⁸ cm⁻¹. -1 The location of this displacement may be related to the increased hydration of PEG segments after neutralization and changes in the local dielectric environment, or it may be affected by COO. - The effect of intramolecular hydrogen bond rearrangement after generation.

[0076] Based on the above product verification, this invention characterizes samples with different degrees of neutralization.

[0077] The sample preparation process is as follows: Take each sample powder to prepare a 2.0 wt% aqueous mother liquor, and record the appearance of the mother liquor (transparent / opaque / turbid / layered). Mix the mother liquor, SWCNT, and deionized water in a ratio of 50.0 g, 1.0 g, and 49.0 g to prepare a CNT pre-dispersion. Mix 48.0 g of silicon-carbon composite powder (Si / C, Si content 10%), 50.0 g of CNT pre-dispersion, 0.5 g of CMC, and 1.0 g of SBR (styrene-butadiene rubber latex, solid content 50%), add deionized water to make up to 100.0 g, stir under vacuum at 1500 rpm for 60 min, and filter with a 200-mesh stainless steel sieve to obtain the negative electrode slurry (i.e., the slurry to be tested).

[0078] The slurry to be tested was coated onto copper foil (10 μm, rough side up) using a doctor blade coater, with a wet film thickness of 200 μm. It was then dried in a forced-air dryer at 80 ℃ for 10 min, followed by vacuum drying at 120 ℃ for 2 h (vacuum degree ≤ -0.09 MPa). The slurry was then cut into φ14 mm circular samples (for battery assembly testing) and 50 mm × 100 mm rectangular samples (for sheet resistance testing and peel testing). The samples were then rolled once at 2 MPa using a roller press.

[0079] The battery assembly test was conducted in a glove box with water and oxygen concentrations << 1 ppm. The assembled battery was a CR2032 (Li counter electrode). The assembly sequence was as follows: negative electrode shell → lithium sheet (φ15.6 mm) → separator (Celgard 2400, φ19 mm) → electrolyte (60 μL, 1M LiPF6 / EC:DMC:EMC = 1:1:1, 2% VC) → test electrode (φ14 mm) → gasket → spring contact → positive electrode shell. After assembly, the glove box was left to stand for 2 hours before sealing.

[0080]

[0081] Note: The first-effect ICE and cycle capacity retention were tested at 0.2 C and 1.0 C rates, respectively; all samples used in the tests were standardized samples, such as CNT pre-dispersion for dispersibility testing.

[0082] In the characterization results above, GPC testing showed that as the degree of neutralization increased from 0% to 95%, the number-average molecular weight exhibited a monotonically increasing trend, gradually rising from approximately 31,000 to approximately 37,000. The polydispersity index also increased monotonically, from approximately 2.1 to approximately 2.6. This change was not due to the growth of the skeletal molecular chains, but rather because after neutralization, the formation of carboxylate groups caused the polymer chains to extend in solution due to like charge repulsion, resulting in a significant increase in hydrodynamic volume. This led to a change in the apparent retention behavior in size-exclusion columns, manifested as a simultaneous increase in both number-average and weight-average molecular weight. Simultaneously, weak secondary interactions may occur between the ionized segments and the column packing material, further exacerbating the broadening of the distribution.

[0083] Figure 2 The GPC molecular weight distribution curves of two representative samples with a neutralization degree of 0% (S1-0%) and 85% (S1-85%) are presented. The peak of 0% neutralization is located at a retention time of approximately 19 min, with a narrow peak shape and a short tail on the right. The peak of 85% neutralization shifts towards a shorter retention time, to approximately 18 min, with a significantly wider peak shape and a longer tail on the right. Both curves exhibit a broad and asymmetrical single-peak morphology, unlike the typical narrow and symmetrical distribution of linear polyurethanes. This morphological difference indicates that trimethylolpropane and dimethylolpropionic acid synergistically construct a highly branched topology, rather than a simple linear chain extension. Furthermore, the polydispersity index of all samples in the table is greater than 2.0. In the field of polyurethane synthesis, the theoretical limit of the polydispersity index (PDI) of linear condensation polymers is approximately 2.0. However, after introducing the trifunctional branching agent trimethylolpropane in this invention, the PDI consistently remains above 2.1 and further increases with increasing neutralization degree. This phenomenon is consistent with the formation of branched topology and can serve as supplementary evidence for the existence of a highly branched framework. The highly branched topology causes the polymer chains to exhibit a compact, random coil conformation in solution, with their ends extending outward to form a radial structure. This provides a structural basis for the subsequent formation of an effective steric hindrance layer on the surface of carbon nanotubes.

[0084] Regarding dispersion performance, DLS tests showed that as the degree of neutralization increased from 0% to 95%, the particle size of the CNT predispersant exhibited a monotonically decreasing trend (see Table 2 for details). Figure 3 The particle size distribution curves of three representative samples with neutralization degrees of 0% (S1-0%), 85% (S1-85%), and 95% (S1-95%) are shown. The peak position of the 0% neutralization sample is approximately 420 nm with a half-maximum width at half-maximum (HWHM) of approximately 150 nm; the peak position of the 85% neutralization sample is approximately 260 nm with an HWHM of approximately 80 nm; and the peak position of the 95% neutralization sample is approximately 200 nm with an HWHM of approximately 60 nm. The decrease in particle size and the narrowing of the distribution demonstrate that the introduction of carboxylate ions causes the dispersant molecules to extend, effectively stripping away the CNT bundles.

[0085] The zeta potential also exhibits a monotonically increasing absolute value: approximately -15 mV for the 0% neutralized sample, rising to -28 mV for the 75% neutralized sample, -33 mV for the 80% neutralized sample, -38 mV for the 85% neutralized sample, -43 mV for the 90% neutralized sample, and reaching -48 mV for the 95% neutralized sample. This increase in absolute potential demonstrates that the carboxylate group indeed imparts a stronger electrostatic repulsion to the particle surface.

[0086] The CNT adsorption test results are shown in the table below. Figure 4 As shown, the saturated adsorption capacity increases slowly as the degree of neutralization increases from 0% to 95%. Figure 4 The adsorption isotherms of three representative samples with neutralization rates of 0% (S1-0%), 85% (S1-85%), and 95% (S1-95%) are shown. The Langmuir fitted Qmax for the 0% neutralization sample is approximately 18.0 mg / g CNT, for the 85% neutralization sample it is approximately 26.0 mg / g CNT, and for the 95% neutralization sample it is approximately 30.0 mg / g CNT. The saturation adsorption capacities of the three isotherms are all relatively low (18–30 mg / g CNT) and the spacing between them is small. This is because the sample in Example 1 does not contain pyrrolidone structural units, and there is a lack of specific interfacial anchoring between the dispersant and carbon nanotubes. The adsorption mainly relies on weak hydrogen bonds and van der Waals forces, thus limiting the upper limit of the adsorption capacity.

[0087] The slurry performance test results are shown in the table. As the degree of neutralization increased from 0% to 95%, the initial viscosity monotonically increased from approximately 4800 mPa·s to approximately 6600 mPa·s, while the 7-day viscosity growth rate monotonically decreased from approximately 40% to approximately 14%. Similarly, the 7-day settling rate also monotonically decreased from approximately 10% to approximately 1.8%. Figure 5 The 7-day viscosity tracking curves of three representative samples—S1-0% (0% neutralization), S1-85% (85% neutralization), and S1-95% (95% neutralization)—are presented. The increase in viscosity is attributed to the extension of polymer chains due to electrostatic repulsion caused by carboxylate groups, resulting in an increase in hydrodynamic volume. These results indicate that increasing the degree of neutralization improves the short-term stability of the slurry, but the improvement in viscosity growth rate and sedimentation rate diminishes marginally with increasing degree of neutralization.

[0088] The electrode performance test results are shown in the table. As the neutralization degree increased from 0% to 95%, the sheet resistance monotonically decreased from about 25 Ω per square meter to about 15 Ω per square meter, while the peel strength monotonically increased from about 38 N per meter to about 52 N per meter. Figure 6The sheet resistance and peel strength of three representative samples—S1-0% (0% neutralization), S1-85% (85% neutralization), and S1-95% (95% neutralization)—are compared. The decrease in sheet resistance and the increase in peel strength indicate that more carboxylate groups improve the interfacial interaction between the dispersant, current collector, and active particles. However, it is noteworthy that while the static performance of the electrode improves monotonically with increasing neutralization, subsequent electrochemical cycling tests show that the long-term capacity retention reaches an inflection point after exceeding 90% neutralization, indicating that dispersant design needs to strike a balance between optimizing interfacial contact and interfacial electrochemical stability.

[0089] Electrochemical performance testing was conducted using coin half-cells with a lithium electrode as the counter electrode. The electrolyte consisted of 1 M LiPF6 dissolved in a 1:1:1 volume ratio of EC, DMC, and EMC, with 2% VC added. Initial efficiency testing was performed at a 0.2 C rate, and cycling testing at a 1.0 C rate, with a voltage range of 0.01 V to 1.5 V. Regarding initial efficiency, the results were approximately 84.5% for 0% neutralized samples, 85.5% for 75% neutralized samples, 86.1% for 80% neutralized samples, 86.5% for 85% neutralized samples, 86.6% for 90% neutralized samples, and 87.2% for 95% neutralized samples. Initial efficiency increased slowly with increasing neutralization degree, but the increase was limited.

[0090] In terms of cycling performance, a key and non-monotonic phenomenon emerges. At 100 laps, the capacity retention is approximately 92.7% for the 0% neutralized sample, 94.9% for the 75% neutralized sample, and 95.6% for the 80% neutralized sample. The 85% neutralized sample peaks at approximately 95.9%, then drops to approximately 95.3% for the 90% neutralized sample, and further to approximately 94.7% for the 95% neutralized sample. This divergence becomes more pronounced at 200 laps: approximately 79.9% for the 0% neutralized sample, 81.9% for the 75% neutralized sample, 85.6% for the 80% neutralized sample, 87.1% for the 85% neutralized sample, and further decreases to approximately 85.1% for the 90% neutralized sample and approximately 82.7% for the 95% neutralized sample. At 300 laps, the 0% neutralized sample retained approximately 69.7%, the 75% neutralized sample approximately 71.8%, the 80% neutralized sample approximately 78.3%, and the 85% neutralized sample remained at its highest at approximately 79.5%. The 90% neutralized sample decreased to approximately 76.4%, and the 95% neutralized sample further decreased to approximately 71.7%. The above cycle capacity retention curves are shown below. Figure 7 As shown.

[0091] The above cycling data show that there is a clear optimal performance window between 80% and 90% neutralization, with the 85% neutralized sample maintaining the highest capacity retention at 100, 200, and 300 cycles. When the neutralization degree is below 80%, electrostatic repulsion and insufficient hydration layer lead to poor slurry stability and rapid capacity decay. When the neutralization degree is above 90%, although the short-term settling rate of the slurry is lower and the Zeta potential is higher, the excessive carboxylate ions trigger a shielding effect of lithium-ion counterions, resulting in increased local ionic strength, compressed effective range of the double layer, and promotion of SEI film overgrowth, which in turn accelerates capacity decay after 200 cycles. Notably, at 300 cycles, the capacity retention of the 95% neutralized sample has fallen back to a level close to that of the 75% sample, further demonstrating the negative effects of overneutralization.

[0092] Example 2

[0093] A carbon-based waterborne polyurethane dispersant is prepared by the following method: Step 1: Prepolymer Formation Dissolve 4.50 g DMPA in 40 mL anhydrous DMF, add 20.0 g IPDI and 0.03 g DBTDL, purge with nitrogen, and react at 80 °C for 2.5 h to form a prepolymer.

[0094] Step 2: High-branching chain extension 1.50 g DMPA, 2.5 g TMP and 4.0 g HEP were dissolved in 40 mL anhydrous DMF to prepare a mixed solution. The solution was then slowly and uniformly added dropwise to the prepolymer prepared in Step 1 over 2 h. After the addition was complete, the mixture was reacted at 75 °C for 3.0 h.

[0095] Step 3: PEG end-capping 5.0 g of MPEG was added to the product of Step 2 and reacted at 70 °C for 90 min. The residual NCO content was titrated using the di-n-butylamine method, and the results showed that the NCO content was reduced to below 0.3 wt%.

[0096] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with a neutralization degree of 85% (determined by potentiometric titration), stir continuously for 2 h, discharge, remove DMF by rotary evaporation, dry under vacuum at 60 °C to constant weight and pulverize to obtain the powder product HB-Pyr(HEP)-PU-PEG.

[0097] The product in this example was characterized by FTIR, and the characterization results are as follows: Figure 8 As shown. In the carbonyl region, from 1500 to 1750 cm⁻¹ -1Within this range, the spectrum exhibits a four-peak system, at 1725 cm⁻¹. -1 The peak at approximately 1670 cm⁻¹ represents the C=O stretching vibration of urethane, originating from all the urethane bonds formed by the reaction of IPDI with DMPA, TMP, MPEG, and HEP. This peak is the baseline strong peak in the full spectrum. -1 A new peak appears at 1705 cm⁻¹, which is the C=O stretching vibration of the lactam ring in HEP ​​pyrrolidone. This peak is completely absent in the corresponding range of Example 1 (S1-85%), proving that the pyrrolidone structural unit has been successfully introduced into the polyurethane framework through chemical bonding. -1 The weak peak at C=O of the residual carboxylic acid dimer indicates a neutralization degree of approximately 85%, 1600 cm⁻¹. -1 COO - Antisymmetric stretching vibration. In the hydrogen-bonded region at 3400 cm⁻¹. -1 Nearby, the peak shape exhibits a transitional state with a narrowing bottom resembling a steamed bun, labeled as a superposition of NH stretching and residual carboxyl OH stretching, similar to the 85% neutralized sample of Example 1. 2925 cm⁻¹ -1 With 2855 cm -1 The methylene CH stretching vibration is located at 1540 cm⁻¹, originating from the IPDI cyclohexane, TMP, HEP methylene segments and the MPEG backbone. -1 The vibration is in-plane bending at NH, 1461 cm. -1 The section is a methylene scissor bend, 1240 cm. -1 The area is the amide III band, 1100 cm. -1 The peak at 1018 cm⁻¹ represents a strong peak for PEG ether bonds. -1 The point is a CO stretching vibration, 873 cm. -1 With 724 cm -1 The positions represent the IPDI backbone and the long-chain methylene vibration, respectively.

[0098] The above results indicate that, while retaining the highly branched polyurethane framework, PEG hydration shell, and lithium carboxylate functional groups, Example 2 successfully added pyrrolidone structural units, providing a structural basis for interfacial anchoring with carbon nanotubes.

[0099] Based on the above product verification, the present invention systematically characterized the sample of Example 2. The sample preparation process was the same as that of Example 1. 85% neutralized sample powder was taken to prepare a 2.0 wt% aqueous mother liquor. The appearance of the mother liquor was recorded as transparent, which was significantly better than the opalescent state of Example 1, proving that the introduction of HEP side groups enhanced the water solubility and dispersibility of the dispersant.

[0100] GPC testing showed that the number-average molecular weight of the sample in Example 2 was approximately 36,000, and the polydispersity index was approximately 2.5. Compared with the sample in Example 1 with a neutralization degree of 85%, the molecular weight and distribution width increased slightly. This is because HEP participated in the construction of the polyurethane backbone as an additional chain extender, but the increase was limited, proving that the introduction of HEP did not destroy the original highly branched topology. The polydispersity index remained above 2.0, which is within the highly branched category.

[0101] Regarding dispersion performance, DLS testing showed that the CNT pre-dispersion liquid of the Example 2 sample had a hydrodynamic particle size of approximately 120 nm and a half-peak width of approximately 50 nm, which was significantly smaller than the 260 nm of the sample of Example 1 with a neutralization degree of 85%. The Zeta potential was approximately -42 mV, which was higher than the -38 mV of Example 1.

[0102] The significant reduction in particle size and the increase in potential indicate that the introduction of the pyrrolidone ring significantly enhances the interfacial interaction between the dispersant and carbon nanotubes, effectively exfoliating the CNT bundles.

[0103] The CNT adsorption capacity was determined using the UV-Vis adsorption isotherm method. The results showed that the Langmuir fitted saturation adsorption capacity Qmax of the sample in Example 2 was approximately 95 mg / g CNT, and the adsorption equilibrium constant KL was approximately 8.5 mL / mg. This adsorption capacity is approximately 3.6 times that of the sample in Example 1 with a neutralization degree of 85%, demonstrating that the pyrrolidone structural units, through dipole-π interactions and hydrophobic association, provide strong interfacial anchoring for carbon nanotubes, significantly improving the limitations of the S1 sample, which relies solely on weak hydrogen bonds and van der Waals forces for adsorption.

[0104] Regarding slurry properties, the initial viscosity of the sample in Example 2 was approximately 4200 mPa·s, significantly lower than the 5800 mPa·s of Example 1. The 7-day viscosity increase was approximately 6%, much lower than the 20% of Example 1. The 7-day settling rate was approximately 0.6%, much lower than the 3.0% of Example 1.

[0105] The above results indicate that although the molecular weight of Example 2 is similar to that of Example 1, the slurry exhibits lower initial viscosity and better long-term stability due to the more uniform dispersion and smaller particle size of CNTs caused by the pyrrolidone anchoring layer.

[0106] Regarding electrode performance, the sheet resistance of the sample in Example 2 was approximately 10 Ω per square meter, and the peel strength was approximately 75 N per meter. Compared to Example 1, the sheet resistance decreased by approximately 45%, and the peel strength increased by approximately 56%. This indicates that the pyrrolidone anchoring layer not only improved the dispersion uniformity of CNTs but also enhanced the interfacial adhesion between the dispersant, current collector, and active particles.

[0107] Electrochemical performance testing was conducted using coin half-cells under the same conditions as in Example 1. In terms of first-cycle efficiency, the Example 2 sample achieved approximately 88.5%, slightly higher than the 86.5% of Example 1. Regarding cycle performance, the capacity retention was approximately 97% after 100 cycles, approximately 92% after 200 cycles, and approximately 85% after 300 cycles. Compared to Example 1, the 300-cycle retention improved by approximately 5.5 percentage points, indicating that the introduction of pyrrolidone structural units suppressed the migration and aggregation of CNTs during cycling through strong interfacial anchoring, maintaining the long-term integrity of the conductive network.

[0108] Example 3

[0109] A carbon-based waterborne polyurethane dispersant is prepared by the following method: Step 1: Prepolymer Formation Dissolve 4.50 g DMPA in 40 mL anhydrous DMF, add 20.0 g IPDI and 0.03 g DBTDL, purge with nitrogen, and react at 80 °C for 2.5 h to form a prepolymer.

[0110] Step 2: High-branching chain extension Dissolve 3.50 g DMPA, 2.5 g TMP and 5.0 g PNVP-OH (Mw≈4000) in 40 mL anhydrous DMF to prepare a mixed solution. Because the PNVP-OH chain is relatively long, it is slowly added to the prepolymer in two batches over 2 h with an interval of 30 min. After the addition is completed, react at 75 ℃ for 3.0 h.

[0111] Step 3: PEG end-capping Add 5.0 g of MPEG to the product of Step 2, react at 70 °C for 90 min, and confirm that NCO has dropped to below 0.3 wt% by the di-n-butylamine method.

[0112] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with 85% neutralization degree (potentiometric titration method), stir continuously for 2 h, discharge, remove DMF by rotary evaporation, vacuum dry at 60 °C to constant weight and pulverize to obtain the product HB-Pyr(PNVP)-PU-PEG.

[0113] The product in this example was characterized by FTIR, and the spectrum showed a value consistent with that of Example 2 at 1670 cm⁻¹. -1 The characteristic peak of pyrrolidone lactam C=O was observed, and because the short chain of PNVP-OH contains multiple pyrrolidone rings, the intensity of this peak was significantly enhanced compared to Example 2, proving that the multi-point adsorption structure was successfully constructed.

[0114] The sample in this example was systematically characterized, and the sample preparation process was the same as in Example 1. The sample mother liquor was transparent with a pale blue sheen.

[0115] GPC testing showed a number-average molecular weight of approximately 38,000, a polydispersity index of approximately 2.6, and the highly branched topology remained intact.

[0116] DLS testing showed that the CNT pre-dispersion had a particle size of approximately 90 nm, a half-peak width of approximately 40 nm (the smallest in the entire series), and a zeta potential of approximately -45 mV.

[0117] CNT adsorption capacity tests showed that Langmuir fitted Qmax was approximately 177 mg / g CNT, which was about 6.9 times that of Example 1 and 1.9 times that of Example 2. The desorption energy barrier of multi-point adsorption was significantly higher than that of single-ring anchoring.

[0118] The initial viscosity of the slurry was approximately 3800 mPa·s, the viscosity increase rate over 7 days was approximately 4%, and the sedimentation rate over 7 days was approximately 0.3%.

[0119] The sheet resistance of the electrode is approximately 8 Ω per square meter, and the peel strength is approximately 90 N per meter.

[0120] Electrochemical tests: initial efficiency approximately 89.5%, retention rate approximately 97.6% after 100 cycles, approximately 93.9% after 200 cycles, and approximately 89.8% after 300 cycles.

[0121] Example 4

[0122] A carbon-based waterborne polyurethane dispersant is prepared by the following method: Step 1: Prepolymer Formation 45.0 g of HTPB was dissolved in 40 mL of anhydrous DMF, and 20.0 g of IPDI and 0.03 g of DBTDL were added. The mixture was then purged with nitrogen and reacted at 80 °C for 3 h to obtain the HTPB prepolymer.

[0123] Step 2: High-branching chain extension Dissolve 2.0 g DMPA, 1.0 g TMP and 3.0 g HEP in 40 mL anhydrous DMF and slowly add the solution dropwise to the prepolymer over 2 h. React at 75 °C for 3 h.

[0124] Step 3: PEG end-capping Add 4.0 g MPEG and react at 70 °C until NCO drops below 0.3 wt%.

[0125] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with 85% neutralization degree, stir continuously for 2 h, discharge, remove DMF by rotary evaporation, vacuum dry at 60 °C to constant weight and pulverize to obtain the product HTPB-HB-Pyr(HEP)-PU-PEG.

[0126] The sample in this example was systematically characterized, and the sample preparation process was the same as in Example 1. The mother liquor had a microemulsified appearance.

[0127] GPC testing showed a number-average molecular weight of approximately 37,000 and a polydispersity index of approximately 2.4.

[0128] DLS testing showed that the CNT pre-dispersion had a particle size of approximately 110 nm, a half-peak width of approximately 45 nm, and a zeta potential of approximately -40 mV.

[0129] The CNT adsorption capacity test showed that Qmax was approximately 103 mg / g CNT, which was slightly higher than that of Example 2 (95 mg / g), demonstrating that the flexible buffer of the HTPB soft segment helped the polymer chain adhere to the CNT surface.

[0130] The initial viscosity of the slurry was approximately 4500 mPa·s, the viscosity increase rate over 7 days was approximately 8%, and the sedimentation rate over 7 days was approximately 0.8%.

[0131] The sheet resistance of the electrode is approximately 9 Ω per square meter, and the peel strength is approximately 85 N per meter.

[0132] Electrochemical testing: initial efficiency approximately 88.0%, retention rate approximately 97.1% after 100 cycles, approximately 93.2% after 200 cycles, and approximately 87.9% after 300 cycles. Compared to Example 2 (approximately 85% after 300 cycles), the retention rate after 300 cycles is improved by approximately 3 percentage points, demonstrating that the HTPB soft segment effectively maintains the integrity of the conductive network by absorbing the volume expansion stress of silicon-carbon.

[0133] Example 5

[0134] A carbon-based waterborne polyurethane dispersant is prepared by the following method: Step 1: Prepolymer Formation Similar to Example 4, 45.0 g of HTPB, 20.0 g of IPDI, and 0.03 g of DBTDL were reacted at 80°C for 3 h to obtain HTPB prepolymer.

[0135] Step 2: High-branching chain extension Dissolve 2.75 g DMPA, 1.5 g TMP and 3.0 g PNVP-OH (Mw≈4000) in 40 mL anhydrous DMF, and add the solution dropwise to the prepolymer in two batches over 2 h. React at 75 °C for 3 h.

[0136] Step 3: PEG end-capping Add 4.0 g MPEG and react at 70 °C until NCO drops below 0.3 wt%.

[0137] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with 85% neutralization degree, stir continuously for 2 h, discharge, remove DMF by rotary evaporation, vacuum dry at 60 °C to constant weight and pulverize to obtain the product HTPB-HB-Pyr(PNVP)-PU-PEG.

[0138] The sample in this example was systematically characterized, and the sample preparation process was the same as in Example 1. The mother liquor was transparent with a pale blue sheen.

[0139] GPC testing showed a number-average molecular weight of approximately 39,000 and a polydispersity index of approximately 2.7.

[0140] DLS testing showed that the CNT pre-dispersion had a particle size of approximately 85 nm and a half-peak width of approximately 35 nm, the smallest in the entire series; the zeta potential was approximately -44 mV.

[0141] The CNT adsorption capacity test showed that Qmax was approximately 196 mg / g CNT, which is the highest in the entire series.

[0142] The initial viscosity of the slurry was approximately 4000 mPa·s, the viscosity increase rate over 7 days was approximately 5%, and the sedimentation rate over 7 days was approximately 0.4%.

[0143] The sheet resistance of the electrode is about 7 Ω per square meter, and the peel strength is about 100 N per meter, both of which are the best in the entire series.

[0144] Electrochemical testing: initial efficiency approximately 90.0%, retention rate approximately 98.0% after 100 cycles, approximately 93.7% after 200 cycles, and approximately 91.6% after 300 cycles. This represents the highest cycle performance in the entire series.

[0145] Comparative Example 1 Sample preparation: 8.5 g DMPA was dissolved in 40 mL anhydrous DMF in a single step, followed by the addition of 20.0 g IPDI and 0.03 g DBTDL. The mixture was then reacted under nitrogen protection at 80 °C for 3 h. Due to the absence of TMP, the system primarily underwent linear chain extension, without forming highly branched topologies constructed by trifunctional branching agents. Subsequently, 5.0 g MPEG was added and the reaction was carried out at 70 °C for 90 min. A small amount of anhydrous methanol was then added to quench residual NCO until the residual NCO content, as determined by the di-n-butylamine method, was below 0.5 wt%. The mixture was cooled to 32 °C, and a 10 wt% lithium hydroxide aqueous solution was added dropwise to prepare a sample with a neutralization degree of 85% (determined by potentiometric titration). The mixture was stirred continuously for 2 h, and the DMF was removed by rotary evaporation. The product was then dried and pulverized to obtain the linear polyurethane product L-PU-PEG.

[0146] The same systematic performance characterization as in Example 1 was subsequently performed.

[0147] Comparative Example 2 The HB-PU-PEG sample (85% neutralization) obtained in Example 1 was used to prepare a 2.0 wt% stock solution. 5.0 g of PVP K30 was added, and the solution was stirred at room temperature for 30 min to obtain a physical blending dispersant solution. Subsequent CNT dispersion, slurry preparation, electrode preparation, and battery testing conditions were completely consistent with those in Example 1.

[0148] Comparative Example 3 Sample preparation: Step 1: Prepolymer Formation Dissolve 4.50 g DMPA in 40 mL anhydrous DMF, add 20.0 g IPDI and 0.03 g DBTDL, purge with nitrogen, and react at 80 °C for 2.5 h to form a prepolymer.

[0149] Step 2: High-branching chain extension 1.50 g DMPA, 2.5 g TMP and 5.0 g HEP were dissolved in 40 mL anhydrous DMF to prepare a mixed solution. The solution was then slowly and uniformly added dropwise to the prepolymer prepared in Step 1 over 2 h. After the addition was complete, the mixture was reacted at 75 °C for 3.0 h.

[0150] Step 3: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with a neutralization degree of 85% (determined by potentiometric titration), stir continuously for 2 h, discharge, remove DMF by rotary evaporation, dry under vacuum at 60 °C to constant weight, and pulverize to obtain powder product D3.

[0151] The same systematic performance characterization as in Example 1 was subsequently performed.

[0152] Comparative Example 4 Sample preparation: Step 1: Prepolymer Formation 45.0 g of HTPB was dissolved in 40 mL of anhydrous DMF, and 20.0 g of IPDI and 0.03 g of DBTDL were added. The mixture was then purged with nitrogen and reacted at 80 °C for 3 h to obtain the HTPB prepolymer.

[0153] Step 2: Linear chain extension 5.5 g DMPA and 3.0 g HEP were dissolved in 40 mL anhydrous DMF to prepare a mixed solution, which was then slowly and uniformly added dropwise to the prepolymer over 2 h. After the addition was complete, the mixture was reacted at 75 °C for 3.0 h. Because no TMP was involved, the system did not form a highly branched topology constructed by the trifunctional branching agent, and mainly yielded linear or low-branched polyurethane structures.

[0154] Step 3: PEG end-capping 4.0 g of MPEG was added to the product and reacted at 70 °C for 90 min. The residual NCO content was titrated using the di-n-butylamine method, and the results showed that the NCO content decreased to below 0.3 wt%.

[0155] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with a neutralization degree of 85% (determined by potentiometric titration), stir continuously for 2 h, discharge, remove DMF by rotary evaporation, dry under vacuum at 60 °C to constant weight, and pulverize to obtain powder product D4.

[0156] The same systematic performance characterization as in Example 1 was subsequently performed.

[0157] Comparative Example 5 Sample preparation: Step 1: Prepolymer Formation 45.0 g of HTPB was dissolved in 40 mL of anhydrous DMF, and 20.0 g of IPDI and 0.03 g of DBTDL were added. The mixture was then purged with nitrogen and reacted at 80 °C for 3 h to obtain the HTPB prepolymer.

[0158] Step 2: High-branching chain extension 3.0 g DMPA and 1.5 g TMP were dissolved in 40 mL of anhydrous DMF to prepare a mixed solution, which was then slowly and uniformly added dropwise to the prepolymer over 2 h. After the addition was complete, the mixture was reacted at 75 °C for 3.0 h. In this comparative example, HEP was not added, and a highly branched framework was constructed solely through DMPA and TMP.

[0159] Step 3: PEG end-capping 4.0 g of MPEG was added to the product and reacted at 70 °C for 90 min. The residual NCO content was titrated using the di-n-butylamine method, and the results showed that the NCO content decreased to below 0.3 wt%.

[0160] Step 4: Post-neutralization Cool to 32 °C, add 10 wt% lithium hydroxide aqueous solution to prepare a sample with a neutralization degree of 85% (determined by potentiometric titration), stir continuously for 2 h, discharge, remove DMF by rotary evaporation, dry under vacuum at 60 °C to constant weight, and pulverize to obtain powder product D5.

[0161] The same systematic performance characterization as in Example 1 was subsequently performed.

[0162] Comparative Example 6 Sample preparation: Step 1: Prepolymer Formation 45.0 g of HTPB was dissolved in 40 mL of anhydrous DMF, and 20.0 g of IPDI and 0.03 g of DBTDL were added. The mixture was then purged with nitrogen and reacted at 80 °C for 3 h to obtain the HTPB prepolymer.

[0163] Step 2: High-branching chain extension 3.0 g DMPA, 1.5 g TMP, and 3.0 g NMP (N-methylpyrrolidone, a small molecule) were dissolved in 40 mL of anhydrous DMF to prepare a mixed solution. This solution was slowly and uniformly added dropwise to the prepolymer over 2 h. After the addition was complete, the mixture was reacted at 75 °C for 3.0 h. In this comparative example, NMP was used instead of HEP in Example 4. Although NMP contains a pyrrolidone ring, it has no reactive hydroxyl groups and does not participate in polyurethane polycondensation; it exists in the system only in a physically mixed form.

[0164] Step 3: PEG end-capping 4.0 g of MPEG was added to the product and reacted at 70 °C for 90 min. The residual NCO content was titrated using the di-n-butylamine method, and the results showed that the NCO content decreased to below 0.3 wt%.

[0165] Step 4: Post-neutralization The sample was cooled to 32 °C, and a 10 wt% lithium hydroxide aqueous solution was added dropwise to prepare a sample with a neutralization degree of 85% (determined by potentiometric titration). The mixture was stirred continuously for 2 h, discharged, and DMF was removed by rotary evaporation. Unreacted small molecules were then washed with deionized water and vacuum dried at 60 °C to constant weight. Gas chromatography or equivalent methods showed that the NMP residue was below the detection limit, yielding powder product D6.

[0166] The same systematic performance characterization as in Example 1 was subsequently performed.

[0167] The characterization results are shown in the table below.

[0168] Comparison table of core performance between the examples and comparative examples

[0169] Note: Example 1 uses a sample with 85% neutrality for comparison.

[0170] From Examples 1 to 3, it is evident that as the pyrrolidone structural unit progresses from deletion to monocyclic side groups and then to short polycyclic chains, the adsorption capacity of carbon nanotubes increases from 26 mg / g to 95 mg / g, and then to 177 mg / g. The dispersion particle size decreases from 260 nm to 120 nm, and then to 90 nm. The capacity retention after 300 cycles also increases from 79.5% to 85.0%, and then to 89.8%. This series of monotonically increasing trends demonstrates that after the pyrrolidone ring is chemically bonded to the polyurethane framework, it can form a strong interfacial anchorage through the interaction of dipoles with delocalized electrons on the carbon nanotube surface and hydrophobic association. Furthermore, the desorption energy barrier of the multi-ring synergistic adsorption is higher than that of the monocyclic ring.

[0171] In contrast, neither Comparative Example 5 nor Comparative Example 6 contained chemically bonded pyrrolidone structures in their final products. Comparative Example 5 did not add any hydroxyl-containing pyrrolidone monomers, while Comparative Example 6, although containing NMP, did not participate in polycondensation due to its lack of reactive hydroxyl groups and was completely volatilized during rotary evaporation. The carbon nanotube adsorption capacities of both were only 32 mg / g and 30 mg / g, respectively, with particle sizes remaining at 245 nm and 250 nm, and retention rates of 82.0% and 81.0% after 300 cycles, respectively, essentially the same as in Example 1. This indicates that simply containing a pyrrolidone ring structure at the molecular level, or simply physically adding small pyrrolidone-containing molecules, is insufficient to improve the dispersion and long-term stability of carbon nanotubes. Only by chemically bonding pyrrolidone units into the polyurethane framework can a quasi-irreversible interfacial anchoring effect be achieved.

[0172] In addition, Example 1 used trimethylolpropane as a trifunctional branching agent to synergistically construct a branched polyurethane framework with dimethylolpropionic acid. Comparative Example 1 removed trimethylolpropane and formed a linear polyurethane solely through linear chain extension, with the subsequent neutralization degree controlled at 85%. The results showed that the dispersion of Comparative Example 1 expanded to a particle size of 380 nm, exhibited a sedimentation rate as high as 9.5% after 7 days, and a capacity retention rate of only 68.5% after 300 cycles, the worst in the entire series.

[0173] This is because the steric hindrance layer thickness of linear topology is more susceptible to the influence of solvent environment and high solid content. Chain entanglement or steric hindrance layer compression may occur in the slurry, resulting in insufficient ability to suppress end-to-end and side-to-side entanglement of high aspect ratio carbon nanotubes.

[0174] Similarly, Example 4, while introducing hydroxyl-terminated polybutadiene and hydroxyethylpyrrolidone, maintained the highly branched framework constructed with trimethylolpropane, resulting in a dispersion particle size of 110 nm. Comparative Example 4, using linear chain extension, although also containing hydroxyl-terminated polybutadiene and hydroxyethylpyrrolidone, saw its particle size increase to 185 nm, and the 300-turn retention decreased from 87.9% in Example 4 to 80.5%. This further illustrates that even with the simultaneous presence of anchoring groups and stress-buffering soft segments in the system, the linear framework, due to insufficient steric hindrance radius, still cannot maintain an effective dispersion in complex slurries with high solids content and high ionic strength. The compact random coil conformation of the highly branched topology is a structural prerequisite for forming a stable, thick sterically hindered layer.

[0175] The main difference between Example 2 and Comparative Example 3 lies in whether methoxylated polyethylene glycol (PEG) was introduced during the end-capping step. The carbon nanotube adsorption capacities of Example 2 and Comparative Example 3 were 95 mg / g and 90 mg / g, respectively, and the dispersion particle sizes were 120 nm and 140 nm, respectively, showing little difference. This indicates that the anchoring effect of hydroxyethylpyrrolidone (HOP) and the steric hindrance effect of the highly branched framework were not significantly weakened by the absence of PEG. However, the 7-day sedimentation rate of Example 2 was only 0.6%, while that of Comparative Example 3 was as high as 6.8%, and the capacity retention rate after 300 cycles also decreased from 85.0% to 77.5%.

[0176] The reason is that the negative electrode slurry of lithium-ion batteries is not a simple aqueous dispersion of carbon nanotubes, but a complex aqueous system containing sodium carboxymethyl cellulose, styrene-butadiene rubber latex, active materials and neutralizing salt groups.

[0177] In this environment, simple electrostatic stabilization mechanisms are easily affected by the ionic environment of the slurry and the high solids content. Example 2 uses polyethylene glycol end-capping to form an outer hydration shell, relying on the hydrogen bond network between ether oxygen atoms and water molecules to provide hydration stabilization. This results in lower sensitivity to the ionic environment compared to simple electrostatic repulsion, thus maintaining better long-term stability. In Comparative Example 3, the absence of this hydration shell significantly reduces long-term stability.

[0178] Comparing Examples 2 and 4, both contain hydroxyethylpyrrolidone for anchoring, but Example 4 additionally introduces hydroxyl-terminated polybutadiene in the prepolymer stage. The 300-cycle capacity retention of Example 4 increased from 85.0% to 87.9%, and the peel strength increased from 75 N / m to 85 N / m. Meanwhile, the carbon nanotube adsorption capacity slightly increased from 95 mg / g to 103 mg / g, and the particle size decreased from 120 nm to 110 nm, with limited variation. This indicates that hydroxyl-terminated polybutadiene does not directly participate in the interfacial anchoring or peel dispersion of carbon nanotubes; its effect is manifested at the mechanical level.

[0179] When the silicon-carbon composite anode undergoes a volume expansion of 250% to 300% during cycling, the stress in the traditional rigid dispersant system is directly transmitted to the interface between the carbon nanotubes and the active particles, leading to carbon nanotube breakage or exfoliation. In Example 4, the hydroxyl-terminated polybutadiene undergoes microphase separation from the hard segment, forming a low-modulus soft segment phase region in a highly elastic state. This absorbs the expansion stress through high-elastic deformation, attenuating the stress transmitted to the anchoring layer below the carbon nanotube breakage threshold, thereby maintaining the integrity of the conductive network. Similarly, in Example 5, compared to Example 3, the retention rate increased from 89.8% to 91.6%, and the peel strength increased from 90 N / m to 100 N / m, further verifying the significant impact of hydroxyl-terminated polybutadiene on performance.

[0180] Comparative Example 2 used physically blended polyvinylpyrrolidone. Although its carbon nanotube adsorption capacity of 32 mg / g was slightly higher than that of Example 1 (26 mg / g), its capacity retention rate after 300 cycles was only 74.5%, which was not only lower than that of Example 1 (79.5%), but also much lower than that of Example 2 (85.0%). This is because physically blended polyvinylpyrrolidone exists in a free state in the continuous phase. Its adsorption and desorption are in dynamic equilibrium. Affected by concentration gradient, electrolyte wetting, and long-term cyclic shearing, it will migrate, elute, or redistribute, which will destroy the uniformity of the conductive network and cause the electrode resistance to increase irreversibly during cycling.

[0181] Comparative Example 6 used NMP instead of hydroxyethylpyrrolidone. Although NMP contains a pyrrolidone ring, it lacks reactive hydroxyl groups and does not participate in polyurethane polycondensation, existing only in a physically mixed form. After washing or dialysis purification and vacuum drying, the NMP residue in the final product was below the detection limit. Its properties were similar to Comparative Example 5, with a 300-cycle retention rate of 81.0% and a carbon nanotube adsorption capacity of 30 mg / g. This indicates that small molecules containing pyrrolidone rings, if they cannot chemically bond to polyurethane molecules, are unlikely to maintain stable interfacial anchorage during electrode drying and long-term cycling.

[0182] Example 5 integrates polycyclic pyrrolidone anchoring, highly branched polyurethane topology, hydroxyl-terminated polybutadiene soft segment buffering, polyethylene glycol hydration shell, and moderately neutralized carboxyl groups, achieving optimal performance across the entire series. The carbon nanotube adsorption capacity reaches 196 mg / g, the minimum dispersion particle size is 85 nm, the electrode sheet resistance is 7 Ω / sq, the peel strength is 100 N / m, and the capacity retention after 300 cycles is 91.6%.

[0183] This result is not a simple superposition of four dispersion mechanisms, but rather a decoupling and recoupling of anchoring, steric hindrance, buffering, and hydration at the nanoscale through hierarchical covalent design of the molecular structure. The pyrrolidone ring provides thermodynamic anchoring on the carbon nanotube surface, the highly branched framework radiates outward to form a kinetic steric hindrance layer, the hydroxyl-terminated polybutadiene soft segments provide mechanical decoupling under cyclic stress, and polyethylene glycol and carboxylates construct a synergistic hydration gradient in the outer layer. Each mechanism is connected by a covalent framework, functioning at its optimal scale, supporting rather than restricting each other, ultimately producing a nonlinear overall enhancement effect. This allows carbon nanotubes to maintain micro-nano-scale dispersion in anode slurries with over 50% solids content, and to maintain the long-term integrity of the conductive network under the cyclic volume expansion stress of the silicon-carbon anode.

Claims

1. A method for preparing a waterborne polyurethane dispersant for carbon-based materials, characterized in that, The method includes: 1) In a protective atmosphere, isophorone diisocyanate is reacted with hydroxyl-containing reactants to synthesize NCO-terminated prepolymers under the action of a catalyst. Chain extension / branching reaction solution is added to the NCO-terminated prepolymers to carry out chain extension and branching reactions to obtain a solution containing highly branched chain extension intermediates. 2) Add an end-capping agent to the solution containing highly branched chain-extending intermediates to perform end-capping treatment, thereby obtaining an aqueous polyurethane dispersant.

2. The method for preparing the carbon-based waterborne polyurethane dispersant according to claim 1, characterized in that, Step 1) The hydroxyl-containing reactants include dimethylolpropionic acid and / or hydroxyl-terminated polybutadiene; Step 1) The catalyst comprises dibutyltin dilaurate; Step 1) The synthesized NCO-terminated prepolymer includes: The hydroxyl-containing reactant was dissolved in DMF, followed by the addition of isophorone diisocyanate and a catalyst. The mixture was then placed in a protective atmosphere and reacted at 75–85 °C for 2–3 h to obtain the NCO-terminated prepolymer.

3. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 1 or 2, characterized in that, Step 1) The amount of the hydroxyl-containing reactant used is 0.075–2.500 g / g isophorone diisocyanate; Step 1) The amount of catalyst used is 1.0 to 2.0 mg / g isophorone diisocyanate.

4. The method for preparing the carbon-based waterborne polyurethane dispersant according to claim 2, characterized in that, When using the dimethylolpropionic acid, the dosage is 0.075–0.225 g / g isophorone diisocyanate; The hydroxyl-terminated polybutadiene has a hydroxyl value of 115–135 mgKOH / g, and the dosage used is 2.0–2.5 g / g isophorone diisocyanate. The amount of dibutyltin dilaurate used is 1.0 to 2.0 mg / g isophorone diisocyanate.

5. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 1, characterized in that, The chain extension / branching reaction solution includes trimethylolpropane, dimethylolpropionic acid, and hydroxyl reactive monomers containing pyrrolidone structures.

6. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 5, characterized in that, The hydroxyl reactive monomers containing pyrrolidone structures include N-(2-hydroxyethyl)pyrrolidone and / or hydroxyl-terminated polyvinylpyrrolidone; in: When using the N-(2-hydroxyethyl)pyrrolidone, the dosage is 0.1–0.3 g / g isophorone diisocyanate; When using the hydroxyl-terminated polyvinylpyrrolidone, the dosage is 0.1–0.3 g / g isophorone diisocyanate.

7. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 5 or 6, characterized in that, The amount of dimethylolpropionic acid used is 0.075–0.225 g / g isophorone diisocyanate; The amount of trimethylolpropane used is 0.05–0.15 g / g isophorone diisocyanate.

8. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 1, characterized in that, Step 1) involves adding the chain extension / branching reaction solution to prepare a solution by dissolving the chain extension / branching reaction solution in DMF. Then, the chain extender solution is slowly added at a rate of 0.75–2.00 % vol / min. After the addition is complete, the reaction is carried out at 73–78 ℃ for 2.5–3.5 h to obtain a solution containing highly branched chain extension intermediates.

9. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 1, characterized in that, Step 2) The capping agent is methoxy-terminated polyethylene glycol; The methoxy-terminated polyethylene glycol has an average molecular weight of 1600–2200 and a hydroxyl value of 25–35 mgKOH / g; The amount of methoxy-terminated polyethylene glycol used is 0.2–0.3 g / g isophorone diisocyanate.

10. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 1 or 9, characterized in that, Step 2) The end-capping treatment is carried out at 68-72 °C for 1-2 h.

11. The method for preparing the carbon-based waterborne polyurethane dispersant according to claim 1, characterized in that, After the end-sealing process in step 2) is completed, a neutralization process is performed; The neutralization treatment involves neutralizing the carboxyl group to a carboxylate, and determining the degree of neutralization by back titration and / or potentiometric titration until the degree of neutralization is 80-90%.

12. The method for preparing the carbon-based aqueous polyurethane dispersant according to claim 11, characterized in that, The neutralization treatment is performed by titration with a 5-10 wt% lithium hydroxide aqueous solution, with continuous stirring during the titration process.

13. A waterborne polyurethane dispersant for carbon-based materials prepared by any one of claims 1 to 12.

14. A waterborne polyurethane dispersant for carbon-based materials, characterized in that, The molecular structure of the carbon-based aqueous polyurethane dispersant comprises the following chain segment units: End groups, side groups, or grafted segments containing pyrrolidone structures; And, a highly branched polyurethane skeleton constructed from diisocyanate, trifunctional branching agent and carboxyl chain extender; And, covalently bonded to the polyethylene glycol segments at the ends of the polyurethane chains.

15. The waterborne polyurethane dispersant for carbon-based materials according to claim 14, characterized in that, The carbon-based waterborne polyurethane dispersant also contains hydroxyl-terminated polybutadiene soft segments.

16. The application of the carbon-based aqueous polyurethane dispersant as described in claim 13, 14, or 15, characterized in that, The carbon-based waterborne polyurethane dispersant is used to formulate carbon-based conductive agent slurry. The carbon-based conductive agent includes single-walled carbon nanotubes and / or multi-walled carbon nanotubes and / or graphene and / or conductive carbon black and / or carbon nanofibers.

Images