Designing Superabsorbent Polymers for Rapid Water Absorption through Microparticle Assembly after Inverse Suspension Polymerization: Examining Capillary Diffusion as a Driving Force

Abstract

Manufacturers employ inverse suspension polymerization to increase water absorption and absorption under the pressure of superabsorbent polymers (SAPs) used in personal hygiene products. Researchers have primarily focused on maximizing the water absorption capacity of SAPs when saturated, often by maximizing the swelling ratio. However, there is a scarcity of literature addressing the fast absorption properties necessary for user comfort and dryness in real-world applications. In this study, we assembled PAA-based SAP microparticles in two different sizes to produce a novel capillary diffusion effect, aimed at enhancing water retention capacity after 1 min of absorption (RCW-1), capable of absorbing more water before swelling. We prepared 30–40 μm SAP microparticles (SAP-1), showing high RCW-1 but very low absorption against a pressure of 0.3 psi (AAP-0.3). We removed the oil phase and added fresh reaction medium to continuously aggregate and grow SAP-1 into larger particles (SAP-2) with a high AAP-0.3 and low RCW-1. Using fluorescence microscopy analysis, we were surprised to find that the small gaps between different-sized SAP microparticles could form water channels, allowing water to diffuse among the internal SAP particles through capillary force, providing rapid water absorption. By mixing SAP-1 and SAP-2 in different proportions, we were able to assemble the SAPs into grape-like clusters, achieving an RCW-1 of 197.97 ± 8.08 g/g and an AAP-0.3 of 21.51 ± 1.85 g/g. The ratio of SAP-1 to SAP-2 could be adjusted to produce clusters with different properties, allowing us to customize them for a wider variety of applications.

Introduction

Superabsorbent polymers (SAPs) are defined by Prof. Matsuda and Ueda as “cross-linked hydrophilic polymers that can retain absorbed water under pressure due to an equilibrium between dissolution and the thermodynamically favored expansion of polymer chains constrained by the cross-linked structure.¹ Commercial SAPs are widely used in various applications, including personal hygiene products for the absorption of water, urine, and blood; in agricultural products to enhance water retention and gas permeability in soils; and in food packaging products to eliminate excess moisture. Frequently used SAPs made of poly(acrylic acid) (PAA) powders are typically produced via polymerization of partially neutralized AA monomer in aqueous solution. The resulting product is a solid bulk or a gel with high density and viscosity, which is difficult to produce with a specific micro/nano morphology for the enhancement of water absorption. In contrast, the alternative inverse suspension polymerization method uses an oil continuous phase to disperse water droplets containing water-soluble monomers in a surfactant with an appropriate hydrophilic lipophilic balance (HLB) value, forming a water-in-oil (w/o) suspension. These droplets can be uniformly polymerized to form high-molecular-weight polymer droplets. The droplets can then be further processed by adding dispersants, coagulants, crosslinkers, and modifiers to form composite particles with high water absorbency and hydraulic conductivity.²⁻⁸

There are many advantages of inverse suspension polymerization for the manufacturing of SAPs.^2,9 The reaction rate can be easily stabilized by controlling the various parameters of the polymerization process. Additionally, the oil continuous phase provides a good heat transfer medium, improving the efficiency of polymerization and allowing for even distribution of the product, which can be further processed to form desired particle structures. The particle size can be finely controlled, unlike in the solution polymerization process, which requires the SAPs to be ground into fine particles.

Unlike certain academic studies that emphasize water absorption at equilibrium, the diaper market in East Asia prioritizes rapid absorption. The indicators “retention capacity in water for 1 min” (RCW-1) and “absorption against pressure” (AAP) are very important for sanitary product manufacturers to evaluate the absorbency speed of SAPs. The microscale SAP particles can be agglomerated and assembled to form clusters in multiple stages of inverse suspension polymerization, producing SAPs with relatively high RCW-1, high swelling ability, and fast absorption kinetics. The inverse suspension polymerization process was originally patented by Sumitomo Co.^10,11 Sumitomo’s product can exhibit high RCW-1 and AAP-0.3 (AAP at 0.3 psi) properties compared to other preparation methods, including the common and convenient solution polymerization method. As a result, Sumitomo has a market share of approximately 20% and is widely used in baby diapers and sanitary napkin products.

Current strategies primarily involve utilizing AA with cross-linking networks or surface strengthening structures to achieve high water absorption through both physical and chemical adsorption. Recent research has focused on how to improve the absorption rate by adjusting the “chemical strategy” related to reaction process conditions, such as copolymer type and monomer ratio, oil phase type and oil-to-water ratio, cross-linker type and concentration, and dispersant type and concentration.¹² Other topics of focus involve the “physical strategy” of structural control of SAPs in the formation of porous or spherical agglomerated particles.¹³ Two different chemical strategies involve using either ionic monomers or surfactants with sulfonate groups to enhance water retention capacity and reswelling ability.^14,15 However, the high cost of ionic monomers limits their use outside of the laboratory. Some have reported that utilization of AA and acrylamide (AM) for inverse suspension copolymerization allowed for synthesis of highly absorbent hydrogel cores.^{1−3,12,14,15} After the hydrogel core is formed, a surface reaction is carried out with a cross-linker to form a shell around it, preventing gel blocking after swelling.¹²

SAPs with only ionic structures tend to undergo localized swelling without subsequent conduction, causing gel blocking.^18,19 Insufficient cross-linking networks and hydrophobic groups can also result in limited and slow physical absorption of water. Therefore, further enhancement of SAPs requires the combination of multiple factors, such as hydrophilicity, ionic forces, hydrophobic interactions, and physical structure. Furthermore, the common theory for the driving force of SAP absorption is the swelling of SAPs due to very high osmotic pressure inside the polymer network, as the ions in SAPs are forced close together. The osmotic pressure is reduced by diluting the charges via absorption of water until all forces are in equilibrium.²⁰ However, we found this explanation to imply a long process for achieving thermodynamic equilibrium. Considering the rapid absorption of water, one of the water absorption mechanisms—capillary forces in the macro-pores or gap structures—might dominate the kinetics of the initial absorption.^17,21

In this study, we prepared two kinds of SAP microparticles with different particle sizes via first- and second-stage polymerization. We then uniquely assembled them to fabricate SAP clusters with gap structures beneficial to capillary diffusion. We also pioneered a fluorescence experiment for observing the speed of water diffusion through capillary force to find the key factors for rapid water adsorption.

Experimental Section

Materials

AA (99%) was the main component of SAPs, and its process surfactants sorbitan monostearate (Span 60, TG > 90%), sorbitan oleate (Span 80, TG > 90%), and polysorbate 80 (Tween 80, TG > 90%) were purchased from Emperor Chemical Co., Ltd. (Taiwan). Cyclohexane, n-heptane, methanol sodium hydroxide, potassium hydroxide, and potassium persulfate (KPS, 98%) were purchased from Echo Chemical Co., Ltd. (Taiwan). 2,2′-Azobis(2-amidinopropane) dihydrochloride (AIBA, 97%), as an initiator in water phase, and the fluorescent dye rhodamine B (>95%) were purchased from Sigma-Aldrich Co. Ltd. The cross-linker polyethylene glycol (400) diacrylate was granted from Eternal Co. Ltd. Surface modifiers. Aliphatic epoxy waterborne cross-linker (W1) and polycarbodiimide (PC1) were from An-Fong Development Co., Ltd., and the ethylene glycol diglycidyl ether (EGDE, 99.7%) was from Weng Jiang Reagent Co., Ltd.

Synthesis of First PAA Microparticles (SAP-1)

The synthesis process was similar to general inverse suspension polymerization shown in the literature.^12,22 Briefly, adequate amounts of sodium hydroxide (or potassium hydroxide) were weighed and put into a flask before RO water was poured in to dissolve the base. AA was dropped into the flask and partially neutralized with the base up to 70% neutralization at 0 °C in an ice bath. Afterward, the initiator, cross-linker, and other additives were added while stirring to obtain an aqueous solution. In another three-neck flask, the surfactant and n-heptane were added as the oil phase with a weight equal to that of the AA aqueous solution. The AA aqueous solution was added dropwise into the oil phase through an attached feeding funnel with vigorous stirring and then stirred for 20 min under N₂ blowing to obtain homogeneous dispersion. The reaction temperature was then raised to 45 °C for 30 min to generate prepolymers and then raised to 65 °C for another 3 h to prepare the first PAA microparticles (SAP-1) (Figure 1a).

Figure 1.

(a) Preparation schemes for SAP-1 and SAP-2 and (b) surface modification of assembled SAP clusters.

Synthesis of Second PAA Microparticles (SAP-2)

The solution containing SAP-1 was placed in a reaction bottle after decantation of the water phase from the top. Cyclohexane, n-hexane, or n-heptane and SAP-1 were mixed in a 1:1 ratio by weight to obtain the oil phase. 70% neutralized AA and a set of surfactants with an HLB value of 8 were added to obtain the water phase. The oil/water ratio was 1:1 (or 1:1.1), and the two phases were mixed. The reaction was initiated by adding initiator and cross-linker solutions, and the temperature was raised to 65 °C. The reaction was then allowed to proceed for 3 h. After polymerization, the surface cross-linker (EGDE) was added, and the stirring speed was increased. The cross-linking reaction was allowed to proceed for an additional 30 min. The product was dried at 120 °C for 2 h to eliminate the presence of n-heptane, obtaining SAP-2 (Figure 1a). The evaporated solvent was collected and recycled for future use.

Analysis of RCW-1 of SAPs

We used a tea bag method to investigate the effects of different parameter adjustments on water absorption speed within 1 min. The testing method was as follows: weigh 0.1 g of SAPs in a PP nonwoven tea bag and measure its weight. Record the original weight of SAPs as Q and the weight of SAPs plus PP nonwoven tea bag as W₁. Pour RO water into a container and place the tea bag containing SAPs in the water for 1 min. After 1 min, remove the tea bag from the water surface and allow it to stand for 1 min to drain excess water. Then, record the weight after absorption as W₂. Finally, the 1 min absorption speed of SAPs can be calculated by the formula RCW-1 = (W₂ – W₁)/Q.

AAP at 0.3 psi (AAP-0.3)

To investigate the effects of different parameter adjustments on AAP-0.3, we also tested the absorption of 0.9% salt solution under 0.3 psi of pressure. The testing method was as follows: place a plastic outer cylinder on an analytical balance and weigh 0.90 ± 0.01 g of SAPs. Spread the SAPs evenly over the surface of the cylinder and then record the SAPs’ weight as S₁. Place a plastic inner cylinder and weights inside the outer cylinder and record the total weight as W₁. Place a filter plate in a Petri dish, pour 0.9% sodium chloride aqueous solution up to the height of the filter plate, and then place filter paper on the filter plate. Place the experimental device in the Petri dish and let it stand for 1 h. Afterward, the device was removed and its weight was recorded as W₂. Finally, the AAP-0.3 of SAPs can be calculated by the formula AAP-0.3 = (W₂ – W₁)/S₁.

Size Analysis of Particles

The particle sizes were analyzed using a synchronizing size and shape analyzer (Microtrac-SYNC) with both dry powder measurements and wet suspension/emulsion measurement. The instrument provides laser diffraction analysis when particles pass the sensor tube and generates dynamic images to additionally analyze the size distribution. In addition to displaying particle size distribution for wet nano/micro-particles and dry powders, graphical reports depicting particle size and shape can be generated through image analysis. Real-time visual observation of particles is also facilitated. Through the analysis, we followed the particle growth process during polymerization and identified dispersion and aggregation via the particle size distribution and particle images.

Fluorescence Microscopic Analysis

To observe the water absorption and gel-blocking phenomena that occur in a very short time, we developed an exploratory method. We adhered SAP powder to one side of a piece of double-sided tape, which was fixed on a glass slide. Fluorescent water dyed with rhodamine B was applied to the slide, allowing for the observation of its diffusion and transportation among SAP particles. The fluorescence distribution in localized areas was examined using an ApoTome subconfocal fluorescence microscopy system.

Results and Discussion

Adjusting the Polymerization Formulation

In the development of the inverse suspension polymerization, we initially tried three reaction parameters based on previous literature: addition of ionic monomers, surfactants, and oil/water ratios.^2−4,16,17 The ionic monomers added to copolymerize with AA were certain amounts of cationic monomer 2-(dimethylamino)ethyl methacrylate (DMAEMA) and zwitterionic monomer 2-(methacryloyloxy)ethyl)dimethyl-(3-sulfopropyl)ammonium hydroxide (SBMA). The preliminary results indicated that the maximum absorption capacity for the AA-co-DMAEMA sample was 250 g/g in water (denoted as RCW-1) and 59 g/g in a NaCl solution (0.9% saline solution) after 5 min of absorption. For the AA-co-SBMA sample, the absorption capacities were 200 g/g for RCW-1 and 51 g/g for the NaCl solution. In comparison, the pure PAA sample exhibited absorption capacities of 220 g/g for RCW-1 and 59 g/g in the NaCl solution. Those sample surfaces were not modified using cross-linking agent EGDE for the evaluation of monomer effects. The results show that the ionic monomer could contribute to RCW-1 values, even when the copolymer ratios and related parameters were not optimized. However, the cost of ionic monomers was too high. For example, the price of DMAEMA is 9 times that of AA, and that of SBMA is even up to 77 times higher. After consideration of cost, practical value, and the structural effect via the physical assembly method, we adopted only AA as the monomer in the investigation. Regarding the cross-linkers, once the amount added was greater than 0.5%, cross-linking density became too high, resulting in RCW-1 under 100 g/g; therefore, we used no more than 0.5% cross-linker in the following experiments. The amount of initiator was also checked and adjusted to be between 1% and 3%, due to the moderate control of molecular weights and polymerization rates affecting water absorption. Surfactant dosage varied depending on the surfactant owing to the different effects of emulsification. Some preliminary studies showed that the optimal dosage of the surfactant was between 1% and 5% for good emulsification and maintenance of stable reactions. As for the neutralization of AA and the reaction temperature, we consulted various literature sources, which suggested that neutralization could be between 65% and 75% and the temperature could range from 65 to 75 °C.²⁴ When the neutralization degree of AA was higher than 70%, the excessive Na⁺ in the polymer chains would have a shielding effect on –COO^–, reducing the electrostatic repulsion between carboxylate groups and hindering the expansion of the molecular chain, thus decreasing water absorption capacity.²⁵ When the degree of neutralization was relatively low, a large amount of COOH groups were present, and the molecular chain conformation between the cross-linking points was curly, which was not conducive to water swelling. Another reason is that AA reactivity was relatively high and the polymerization reaction was difficult to control when the neutralization degree of AA was low.²⁶ According to previous research for the partial neutralization of AA, 70% neutralization can achieve an optimal water absorption value.²⁵⁻³⁰ Therefore, we directly adopted the results of the references to prepare SAP-1 using a neutralization level of 70% AA and set the reaction temperature at 65 °C.

Choosing an Effective Surfactant

In order to choose an effective surfactant, we checked some surfactants with different HLB values to observe the stability of the suspension and its effects on water absorption. We investigated the following surfactants: sorbitan sesquioleate (HLB = 3.7), sorbitan monostearate (SPAN 60) (HLB = 4.7), ethylene glycol distearate (HLB = 5–6), cetearyl olivate (HLB = 8–9), and lecithin (HLB = 10). In the selection of surfactants, ethylene glycol distearate and lecithin were excluded due to the instability of the emulsion system, which often resulted in block formation after polymerization. Although cetearyl olivate provided the emulsion system stability, it was not easily soluble. Therefore, the investigation was mainly focused on sorbitan sesquioleate and sorbitan monostearate. Surfactants with lower HLB values generally contribute to a more stable emulsion system with a lower required dosage. On the other hand, surfactants with higher HLB values may require an increased dosage to achieve emulsion system stability. The evaluation results can be seen in Table S1. The stirring speed was also evaluated in the preliminary tests, and it was found that the control of 300 rpm was suitable for obtaining stable suspension systems without solid precipitation because the surfactant could be redispersed well and sufficiently to stabilize the polymer droplets.²⁴ As a result, sorbitan sesquioleate and sorbitan monostearate were selected for use in further tests.

Adding different ratios of surfactants and adjusting the oil/water ratios revealed that a higher oil-to-water ratio and insufficient surfactant concentration both raised the risk of destabilizing the suspension systems, leading to significant solid precipitation or fast agglomeration. When the surfactant concentration was excessively high, although the emulsion system remained stable with reduced solid precipitation, there was a decrease in the water absorption speed, possibly caused by the increased hydrophobic clusters. Therefore, suitable oil-to-water ratios were determined to be 1:1 and 2:1, with a surfactant concentration of 2%. The different surfactant ratios and oil/water ratios with water and saline absorption are shown in Table S2. From the results, it was evident that at low oil/water ratios and a surfactant concentration of 2%, the RCW-1 concentration exceeded 200 g/g. Specifically, using sorbitan monostearate (2%) as the surfactant with an oil/water ratio of 1:1 achieved a high water absorption speed, reaching 233.06 ± 1.52 g/g within 1 min, without any precipitation observed. Figure 2 shows the SEM images of the SAP-1 synthesized with the conditions of SAPN60 surfactant addition and control of the oil/water ratio of 1:1, evidencing the excellent dispersion in the inverse suspension polymerization to form approximately 38 μm microparticles.

Figure 2.

(a) SEM Image (1.0 kV, 35×) and (b) its magnified view of SAP-1 via 2% SPAN addition with oil/water ratio of 1:1 (1.0 kV, 400×). (c) Size distribution of the microparticles with the majority measured at ∼38 μm.

Adjusting the Amounts of Cross-Linkers and Initiators

To optimize the SAP-1 preparation for increasing both the RCW-1 and AAP-0.3, and for safety considerations, we modified some other parameters in the aqueous solution—the bases for partial neutralization of AA, the initiators, and the cross-linker concentrations—so that it could be suspended in the oil phase of n-heptane. The modified formulations were designed to increase the swelling capacity for enhancing RCW-1, but with longer PAA main chains and cross-linking density for enhancing AAP-0.3. It was observed that most suspensions remained stable under this main formulation. When the initiator dosage was less than 0.5%, the reaction rate decreased and less heat was generated, leading to an unstable polymerization process and then precipitation condensed in the solution. Additionally, when the concentration of AA was too high, the polymerization rate got excessively fast, leading to uncontrollable rapid expansion and agglomeration. In contrast, when the concentration of AA was too low, the polymerization rate was reduced and the monomer remained in the reactor. In both cases, we were unable to obtain a usable product. However, we adjusted the oil/water ratios, which dominates the formation of polymer droplets, to control the AA concentration in the whole system. Therefore, the subsequent addition of additives or secondary cross-linking reactions was carried out based on the main formulation. In the optimization of the first-stage polymerization process, we encountered the problem that the continuous phase often counter-flowed into the condensation tube. This altered the ratio between the continuous and dispersed phases, destabilizing the emulsification process and causing partial agglomeration. Hence, the main formulation was adjusted to achieve higher stability and enhance water absorption rate. The main and modified formulations are listed in Table 1, along with their RCW-1 results. The base formulation was as follows: the aqueous phase was 70% AA neutralized by NaOH solution, 1.5% (w/w based on AA) KPS, and 0.2% (w/w based on AA) polyethylene glycol diacrylate (as a cross-linker). The oil phase was n-heptane with 2% (w/w based on AA) SPAN60, and the oil/water phase ratio was 1:1 by weight.

Table 1. Parameters of the First Stage of Polymerization (SAP-1).

sample	main parameters of first inverse suspension polymerization				physical property
entry	basic solutiona	initiatorb	cross-linkerc (%)	oil/water ratiosd	RCW-1 (g/g)
SAP1–1	KOH	KPS	0.2	1:1 (w/w)	NDe
SAP1–2	KOH	AIBA	0.2	1:1 (w/w)	NDe
SAP1–3	NaOH	AIBA	0.2	1:1 (w/w)	75.32 ± 2.26
SAP1–4	NaOH	AIBA/KPS (2/3)	0.2	1:1 (w/w)	70.54 ± 6.33
SAP1–5	NaOH	KPS	0.2	1:1.1 (w/w)	231.45 ± 6.71
SAP1–6	NaOH	KPS	0.35	1:1.1 (w/w)	236.56 ± 6.53

^a70% of AA neutralized by basic solution.

^b1.5% initiator (w/w based on AA).

^cCertain % (w/w based on AA) polyethylene glycol diacrylate (as a cross-linker).

^dThe oil phase included n-heptane with 2% (w/w based on AA), SPAN60, and the oil/water ratio was 1/1 by weight.

^eSample was agglomerated to a solid precipitation in the suspension.

The emulsions prepared with the base formulation were generally stable, with only a small amount of solid deposited on the reactor wall after the reaction. When the initiator concentration was <0.5%, the reaction rate was too slow, resulting in low reaction heat, unstable polymerization process, and the formation of solid particles in the solution. When the acrylic monomer concentration was too high, the reaction was too fast, leading to gelation. Therefore, the base formulation was used for the subsequent addition of additives or secondary cross-linking reactions.

The parameters of SAP1–5 and SAP1–6 showed the most stable suspension system, suitable for the following SAP-1 preparation (1st stage). However, the AAP-0.3 shown in Tables S1 and S2 indicate that all samples’ AAP-0.3 were less than 8 g/g, suggesting that surfactant species, surfactant ratios, and oil/water ratios in a certain range were unable to enhance AAP-0.3; that is, the water could not be retained in the SAP networks to resist the pressure of 0.3 psi. Therefore, in subsequent experimental strategies, the focus was on allowing particles to undergo secondary growth or incorporating surface cross-linking agents to enhance the performance of AAP-0.3.

To further adjust the polymerization formulation, potassium hydroxide was substituted for sodium hydroxide in the aqueous phase to partially neutralize AA. This change was aimed at enlarging the cross-linked units within SAPs, providing additional space for water absorption. This was due to the larger atomic radius of potassium ions compared to that of sodium ions, contributing to a looser internal structure in SAPs. However, in sample SAP1–1, it was observed that the addition of the initiator into the aqueous phase formed an undissolved white, cloudy suspension, which showed a nonhomogeneous aqueous solution. This condition hindered the polymerization reaction, prompting a change to AIBA as the initiator in the sample SAP1–2 according to the literature.¹⁰ However, that formulation exhibited similar issues irrespective of whether AIBA or AIBA/KPS co-initiator was used, or even raising the neutralization up to 85%. By using KPS as the initiator and slightly modifying the oil/water ratio from 1:1 to 1:1.1, the samples SAP1–5 and SAP1–6 were effectively improved to show stable polymerization. Similarly, changing the concentration of internal cross-linkers from 0.2% to 0.35% in SAP1–6 increased the RCW-1 of SAP-1.

Amount of the Selected Modifier

Due to low AAP-0.3 values for SAP-1, the surface modification of SAP-1 particles was tried using three different modifiers to increase the cross-linking density on the SAP particle surfaces. The selected surface modifiers were, respectively, aliphatic epoxy waterborne cross-linker (W1), EGDE, and polycarbodiimide (PC1), spray-coated onto SAP particles. It was found that with the increase in the modifier amount for the two epoxy modifiers, the RCW-1 showed no significant change, and there was not much variation in AAP-0.3 after using the modifiers (listed in Table S3). When SAP particles were coated with PC1, an increase in the modifier amount decreased the level of RCW-1, but there was not a significant change in AAP-0.3. The surface modifiers led to surface cross-linking and probable formation of more hydrophobic surfaces (longer aliphatic chains in W1 and PC1), so we selected the EGDE with epoxy functional groups and short ethoxy chains, which might retain their hydrophilicity. By applying the 20% EGDE (based on SAPs) to SAP1–6, we obtained the SAP1–6E with performance of slightly lower RCW-1 of 223.20 ± 6.81 g/g and much higher AAP-0.3 of 18.60 ± 0.11 g/g (compared to SAP1–5 × 10³ in Table S3), showing the effect of altering the internal cross-links and surface cross-links of SAP particles on enhancing RCW-1 and AAP-0.3. However, SAP1–6E still could not meet the commercialization requirements of AAP-0.3.

Moreover, through the RCW-1 test, we found that samples of SAP1–5 and SAP1–6 exhibited higher water absorption but some SAP-1 particles failed to absorb water, appearing as white clusters in the tea bag (Figure S1). This may have been due to portions of the SAP sample experiencing gel blocking. Based on this observation, further exploration was conducted to observe the short-term water absorption of SAPs. We applied a drop of water containing a trace amount of blue dye as the indicator to monitor the water diffusion. We observed that when the dye solution was dropped on a small area of dry SAPs, the blue color immediately diffused in the adjacent area and those SAP particles retained their sizes, showing their ability to absorb water without swelling. The water spread toward the gaps between particles (Figure S1c), indicating that capillary diffusion was probably one of the dominant factors in the initial water absorption. It inspired us to design further absorption experiments employing fluorescence probes to explore the mechanism of initial absorption and improve the SAP design.

Factors Affecting the Formation of SAP-2

To enhance the performance of AAP-0.3, we initiated a second stage of polymerization aimed at growing and/or agglomerating the first-stage microparticles (SAP-1) into larger particles (SAP-2). A surfactant with an appropriate HLB value added into the partially neutralized AA (70%) in the aqueous phase significantly improved monomer absorption and achieved particle growth. Another factor was the timing of the addition of the initiator and the cross-linker, either before or after mixing the water and oil phases. All the parameter adjustments were based on the competitive reactions between the PAA reforming to generate new small particles or getting absorbed into the original SAP-1 to form larger particles or clusters. The results in Table 2 reveal that an unstable reaction occurred when the initiator and cross-linker were added before mixing the oil and water phases: the second polymerization phase took place mostly in the aqueous solution and quickly produced a gel block (samples SAP2–1 and SAP2–2). The oil/water ratio was then fixed at 1:1, but we conducted the second polymerization via adjusting different surfactants with varying HLB values and initiator amounts to control the oil/water ratio and enhance the growth of microparticles. Sample SAP2–3 was prepared using sodium dodecyl sulfate, the surfactant with the highest HLB value of 40, increasing the generation of new small microparticles, similar to SAP-1. Sample SAP2–4 was designed to use a lower HLB surfactant, Tween 80, to increase the oil phase thickness in suspended droplets to adsorb on the original SAP-1 particle surfaces for the growth of larger particles. The size distribution of the sample showed that the particle sizes of SAP2–4 were slightly increased, with about 33% growing to approximately 80 μm, but the sample still displayed poor AAP-0.3 performance, as shown in Figures S2 and S3. We continued to adjust the HLB and found that an appropriate value of 8 using a mixture of surfactants could produce a faster aggregation of small particles. Compared to samples SAP2–5 with 1.5% initiator, SAP2–6, −7, and −8 with a lower concentration of 0.5% initiator were found to rapidly aggregate those original small SAP-1 particles to produce SAP clusters, as shown in Figure 3. Samples SAP2–7 and −8 consisted of an SAP2–6 preparation with surface cross-linking reaction by adding EGDE using SAP1–5 and SAP1–6 as the seeds, respectively. With the aid of particle size analysis and sampling microscopic analysis, the process of particle growth was controlled at different stages by adjusting parameters of the second stage of polymerization (shown in Figure 4). After decanting out the top aqueous layer, the suspended SAPs slightly aggregated (shown in Figure 4a series), redispersing after new AA aqueous solution in combination with a mixed surfactant with an HLB of 8 were added and stirred for 40 min to mix the oil–water phases. The critical operation led to AA being absorbed (or swelled) and dispersed between SAP-1 particles (shown in the Figure 4b series). A further 40 min reaction after addition of the initiator and cross-linker resulted in increasing small particle formation, and stable particle growth and increased particle sizes were found after 2 h of reaction time, as shown in Figure 4d series. The size distribution can be seen in Figure 4d4, showing that nearly 70% of particles had agglomerated to more than 100 μm while staying well-suspended. After surface cross-linking, the larger particles of SAP-2 achieved the aggregation of SAP microparticles resembling clusters of grapes. The D₅₀ size distributions of SAP-2 were between 29 and 45 μm, which can be seen in Figure S2. We employed SAP1–6 (with more internal cross-links than SAP1–5) as the base particles to prepare SAP2–8, achieving high RCW-1 of 121.38 ± 4.76 (g/g) and much improved AAP-0.3 of 28.17 ± 0.35 (g/g) with larger suspended clusters (shown in Figure 3).

Table 2. Parameters of the Second Stage of Polymerization (SAP-2) Based on SAP1-5 and SAP1-6.

sample	parameters of second polymerization							water absorption
entry	oil/water ratio	surfactant (based on AA) added in aqueous phase	addition of the initiator/cross-linker before or after oil/water mixing	amount of the initiator (based on AA) (%)	addition of the surface modifier	stirring rate (rpm)	results	RCW-1 (g/g)	saline absorption for 5 min (g/g)	AAP-0.3 (g/g)
SAP2–1	1:1		before	1.5	EGDE	300	agglomerate to a block	144.86 ± 4.75	43.13 ± 1.65	17.39 ± 0.66
SAP2–2	1:1.5		before	1.5	EGDE	300	agglomerate to a block	83.84 ± 1.35	35.12 ± 1.03	18.81 ± 0.85
SAP2–3	1:1	1% SDS (HLB 40)	before	1.5	EGDE	500	grow to SAP-1	145.28 ± 2.54	38.56 ± 0.64	12.57 ± 0.32
SAP2–4	1:1	1.25% Tween80 (HLB 15)	after	1.5		500	grow to larger particles	128.55 ± 21.31	49.25 ± 1.16	5.27 ± 0.12
SAP2–5	1:1	1% HLB-8	after	1.5		500	grow to SAP-1	145.93 ± 20.21	46.52 ± 2.64	4.95 ± 0.02
SAP2–6	1:1	1% HLB-8	after	0.5		500	grow to larger particles	42.63 ± 5.59	24.86 ± 1.40	3.75 ± 0.85
SAP2–7	1:1	1% HLB-8	after	0.5	EGDE	500	grow to larger particles	81.44 ± 2.54	42.64 ± 1.00	25.12 ± 0.53
SAP2–8a	1:1	1% HLB-8	after	0.5	EGDE	500	grow to larger particles	121.38 ± 4.76	44.52 ± 0.06	28.17 ± 0.35

^aSAP2–8 is based on SAP1–6 to continue increasing particle size in the second polymerization, while others are based on SAP1–5.

Figure 3.

SEM Images of SAP2–5, SAP2–6, and SAP2–8. (a,d,g) Low magnification views at 40×, 40×, and 100×, respectively; (b,e,h) Closer looks at particle agglomeration at 400×, 120×, and 400× magnification, respectively; and (c,f,i) Cluster surfaces in a local spot at 1000×, 1000×, and 2500× magnification, respectively. The accelerating voltage is 3.0 kV for all images.

Figure 4.

Observation of SAP particle growth and related size distributions of those wet suspension samples at a, b, c, d different stages via (a1–d1) camera and (a1–d2) optical microscope. The scale bars represent 100 μm. Size evaluation via (a3–d3) laser detection and (a4–d4) dynamic image analysis of particles. The small tables show their related statistical grouping of particle sizes.

According to the results, we contemplated whether it would be possible to blend SAP-1and SAP-2 in different proportions to impart distinct material properties. For instance, SAP-1 could contribute to a high RCW-1, while SAP-2 could provide elevated AAP-0.3. Hence, SAP1–6 with RCW-1 of 236.56 ± 6.53 g/g and AAP-0.3 of 8.74 ± 0.61 g/g was chosen as the representative of SAP-1 for the following blending. This design allowed for a further exploration of the gap formation between particles of varying sizes and whether they play a significant role in water absorption rate, which can be seen in Figure 5. Water absorption tests were conducted using different proportions of SAP-1 and SAP-2, as listed in Table 3. It was evidenced that, in comparison with the group with no surface modification, as the amount of SAP-2 increased, so did AAP-0.3. Moreover, the larger particles of SAP-2 after surface modification significantly enhanced AAP-0.3 compared to those with no surface cross-linking. With a relatively small amount of surface modification (3% modifier based on SAPs with 3% EGDE), an increase in the level of AAP-0.3 and an accompanying rise in the level of RCW-1 were observed (7:3 and 5:5 groups in Table 3, respectively). It was inferred that a cross-linking reaction occurred between SAP-1 and SAP-2, thereby increasing the gap ratio between smaller SAP-1 particles and enhancing water absorption speed. In intragroup comparisons, an increase in EGDE from 5% modifier to 15% slightly decreased the RCW-1, indicating that some EGDE reacting with outer surfaces of the clusters may cause a decrease in water absorption, but the formation of capillary diffusion gaps remains crucial for rapid water absorption. Conversely, an increase in the proportion of SAP-2 led to a reduction in RCW-1. This flexible design provided a good compromise between RCW-1 and AAP-0.3 that could be adjusted for various requirements.

Figure 5.

SEM images for the blends of SAP-1 with SAP-2 (1/1). (a–c) Before EGDE modification and (d–f) after the modification. (a) Low magnification view at 100×; (b,c) local areas on SAP-2 (400×) and SAP-1 (400×), respectively; (d) low magnification view after EGDE modification at 100×; (e) clusters at 400×; (f) cluster surface at 2500×. The accelerating voltage is 3.0 kV for all images.

Table 3. Water Absorption Properties of SAP-1 Blended with SAP-2 in Different Ratios.

	7:3		5:5		3:7
SAP-1/SAP-2 (w/w) ratios	RCW-1 (g/g)	AAP-0.3 (g/g)	RCW-1 (g/g)	AAP-0.3 (g/g)	RCW-1 (g/g)	AAP-0.3 (g/g)
no surface modification	192.17 ± 1.80	11.08 ± 0.12	181.75 ± 2.79	16.03 ± 0.43	148.36 ± 2.87	20.19 ± 0.23
3% modifier (based on SAPs) with 3% EGDE	216.09 ± 5.06	17.79 ± 0.38	197.97 ± 8.08	21.51 ± 1.85	149.81 ± 0.27	24.65 ± 0.71
5% modifier (based on SAPs) with 10% EGDE	185.87 ± 11.74	18.98 ± 0.01	180.46 ± 3.07	23.56 ± 0.40	139.73 ± 1.74	24.49 ± 0.17
15% modifier (based on SAPs) with 10% EGDE	183.83 ± 13.75	19.53 ± 0.70	176.34 ± 4.35	23.81 ± 0.50	131.96 ± 2.01	25.81 ± 1.21

Theoretical Model of Capillary Diffusion and Swelling of Particles

Owing to the water absorption between SAP particles, the theoretical model of capillary diffusion and swelling of particles is relatively complicated. In consideration of the dominating factors of water absorption in a short time, we sought out some references to assist us in explaining our results. The Lucas–Washburn (LW) equation (also known as Washburn’s equation) shows that capillary penetration and fluid transport through porous or gap structures exhibit diffusive behavior.³¹ It describes the changing speed of the meniscus over time in a capillary, porous, or gap structural material. Further, some research modified the LW equation to find an equation for the frontal position of the advancing (imbibing) fluid in homogeneous porous media.³¹ The capillary force F_c from the curved meniscus is opposed by the force from gravity F_h, the force from Poiseuille flow is F_P (neglecting air resistance), and inertia is F_i(32,33)

1

and since F = P × A, and the initial inertial regime disappears quickly for a water–air system.³¹ It gives

2

where η is the viscosity of the liquid, γ is the surface tension of water, θ is the contact angle between the penetrating liquid and the solid, R is the pore radius, and h is the capillary height. If a horizontal capillary is considered (renaming h to x) and disregarding inertial forces, which only occur in the very beginning, via integration we receive the following simplified equation³³

3

where x is the traveled distance and is proportional to the square root of time.

The research employed porous alginate to estimate the rate of diffusion via the LW model, leading to a very early stage transmission of water among the porous hydrogel. Larger diameters result in faster capillary action. While the 630 μm-diameter capillary was already filled with water after 0.75 s, it took on average 2.56 s for the capillary 180 μm in diameter to diffuse over the x² of 3000 mm², equaling approximately diffusion rate of 21.4 mm/s (for alginate).³³ The pore gap of 180 μm in the reference research was suitably analogous to the average diameters and gaps between particles in our study.

The swelling of an SAP particle when in contact with water is driven by the difference in chemical potential between the SAP particle and the surrounding water, assuming that the swelling rate is governed by diffusion of water molecules into the SAP particle. Based on some simplifications, the absorption of water at the particle scale can be approximated with the following equation³⁴

4

where r_i is the particle radius, D is a diffusion coefficient for water molecules in SAPs, ρ_s is the density of dry SAPs, ρ_w is the density of water, Qiabs is the absorption ratio for each individual particle i, and the maximum value of Q^abs_i is denoted by Q^max. From the experimental data fitting with the equation, they found a good fitting (R² > 0.86) with D as 4.8 × 10^–4 cm² min^–1 and Q^max = 216 g/g ± 15%. The result showed that the radius change in 1 min was about 1.3–1.6 mm, approximately equal to the swelling rate of 0.022–0.027 mm/s. Therefore, we reasonably supposed that the capillary diffusion dominated the absorption mechanism in the very initial stage, suggesting that the key factors for enhancing the RCW-1 can be further controlled: wetting properties on the particle surface (contribution to cos θ) and pore diameter/size (contribution to R).

Another issue, gel blocking, also dominates the diffusion of water.¹² Gel blocking is an effect that happens when SAP particles swell too quickly when in contact with water, blocking the flow of liquid to other particles located in different areas.³⁵ We designed two convenient evaluation methods for in situ observation of water absorption. At first, in order to determine the water diffusion speeds in different SAPs, we packed the SAP-1, SAP-2, and SAP-1 + 2 powders in three separate capillary tubes to compare the capillary heights of water diffusion via addition of a fluorescent dye (rhodamine B) in water and irradiation under UV light (mainly at 365 nm wavelength). From the recorded video, the water absorption was the fastest in the tube packed with SAP-2; however, the larger particles with large gaps contributed to the instant capillary force for diffusion, but their loose binding also resulted in particles flowing out of the tube. In practical applications, SAPs will get crushed or broken into pieces. As for the SAP-1, owing to the smaller particle sizes, the SAP-1 tube appeared to be packed closely, leading to apparent gel-blocking in the head of the tube, and no meniscus could be found. The results can be seen in Figure 6 and the video in the Supporting Information.

Figure 6.

Observation of rapid water absorption of SAP-1, SAP-2, and SAP-1 + 2. The capillary heights of samples packed into the capillary tubes after dyed-water absorption (a) for 1 min and (b) for 5 min. (c) Fluorescence microscopic analysis for the real-time water absorption process in dry SAP powders fixed on glass slides. (Dye: rhodamine B). The scale bars in the fluorescent images represent 10 μm.

Water Absorption Process

In order to observe the structural changes of SAP microparticles and clusters during the initial water absorption, we pioneered a method to record the process. We fixed dry SAP powder on a glass slide and added one drop of diluted dyed water. We could then observe the fluorescence response accompanying water absorption. The bright field images taken via an optical microscope show the absorption of the dyed water by the dry SAPs, and the dark field images show their corresponding fluorescence images, proving the structural changes of the SAPs when absorbing water. In the beginning of the water absorption process, the dry SAPs with many gaps diffused water in a very short time (within 1 s), which could be attributed to the capillary force effect, as evidenced by the microscopic analysis of fluorescent orange color distribution. The absorbed water then caused the SAPs to swell. However, the SAP-1 particles appeared to swell quickly and congest each other, breaking down the interfaces between particles and resulting in gel blocking. SAP-2 with larger clusters absorbed water molecules but with larger gaps between clusters, resulting in slower capillary diffusion of water and retention of water on cluster shells, consistent with the low RCW-1 but high AAP-0.3. SAP-2 also showed a drawback when individually applied; that is, the larger particles or clusters were not well-cross-linked or fixed firmly. When they met water, the clusters were crushed or broken into small pieces, making SAP-2 unsuitable for application in diapers (shown in Figure 6 and the real-time changes of SAPs can be seen in the videos in the Movies S2, S3, S4, and S5 Supporting Information). SAP-1 + 2, prepared via blending SAP-1 with SAP-2 and assembling them into clusters with different sizes, had many gaps between clusters, which could be adjusted to improve the performance of RCW-1 and AAP-0.3. It showed rapid water absorption through capillary diffusion among the increased gap structures, and it retained water on the shell layers of the clusters without breaking apart, allowing for continuous water absorption to the maximum amount via the swelling effect. All the effects can be seen in Figure 6, and the videos of the absorption process in real time can be watched in Movies S2, S3, S4, and S5 in the Supporting Information. We also carried out water absorption of SAPs for 3 h until equilibrium, as shown in Figure S4, proving that the rapid water absorption within the first minute dominates the water absorption capacity of the SAP microparticles. FTIR was utilized to characterize the changes of SAP molecules before and after water absorption. The characteristic peaks for dry SAPs can be seen and particularly shown at the two peaks: 1570.5 cm^–1 is the antisymmetric stretching vibration peak of COO in −COO^–; 1408.9 cm^–1 is the symmetric stretching vibration peak of COO in −COO^–.⁴ Water absorbed into swollen SAPs can be seen in the spectra, showing the water occupying the gaps and insides of the SAPs (as shown in Figure S5). Furthermore, the water absorption amount of SAP-2 (large and aggregated SAPs) at equilibrium was found to be slightly lower than that of SAP-1, based on the observation of remaining trivial peaks at around 1570 cm^–1, revealing that the water absorption amount of SAP-2 is lower than SAP-1. Another water absorption phenomenon was interesting due to our design for observing the individual SAP-1 and SAP-2 particles swelling in water. The microscopic observation in Figure S6 shows that the swelling ratios for both SAP-1 and SAP-2 are about 5 times in diameter. The SAPs in a certain space (analogous to measurements in a tea bag) are congested to each other and limited in their ability to freely swell and develop in water. After rapid absorption via capillary diffusion, the SAPs fill the gaps between microparticles and form gel-blocking surfaces to resist their swelling. Also, the effect can enhance the water retention under pressure.

We concluded that the water absorption process for SAP-1 + 2 hybrid clusters could be described as 5 stages (as shown in Figure 7): first, the assembled SAP clusters were dried and placed in the carrier. When the SAP cluster was in contact with water, the moving meniscus wetted the cluster surfaces. Then the meniscus interacted with the gaps between particles to initiate capillary diffusion and spread all over the gap structure within several seconds. Small particles (like SAP-1) were then fully swollen, and larger clusters (like SAP-2) kept swelling toward inner cores. Eventually, the SAP clusters were fully swollen and achieved equilibrium, similar to the sum of force vectors being zero.

Figure 7.

Illustration of water absorption stages for SAP-1 + 2 clusters. (1) Dry SAPs, (2) wetting surfaces, (3) capillary diffusion, (4) swelling toward inner cores, and (5) fully swollen and in equilibrium.

Environmental and Safety Concerns

As these SAPs may be used in the manufacture of personal hygiene products, it is important to consider the potential impact on the environment, as well as human health and safety. The production process involves the use of organic solvents and synthetic surfactants, but we mitigated the potential for harm in the following ways. First, during the production of SAP-1, the product was dried at 120 °C for 2 h to eliminate the solvent (n-heptane), which was then collected and recycled for future use. Second, we were careful to select surfactants that are commonly used in cosmetic products, hair conditioners, and even in the food and medicine industries. Further testing is needed to verify that surfactants and residual organic solvents are present at or below safe amounts in the final product.

Conclusions

In this study, we successfully utilized two-stage inverse suspension polymerization to prepare SAP microparticles. The key process parameters of the first stage of polymerization were controlled to find suitable surfactants, partial neutralization degrees, initiators, oil/water ratios, cross-linkers, and so on to prepare SAP-1. Owing to SAP-1’s relatively low AAP-0.3, we continued to perform the second-stage polymerization via control of oil/water ratios, surfactant with an HLB value of 8, addition timing for initiators and cross-linkers, initiator amounts, etc. to obtain larger SAP clusters (SAP-2) with a low RCW-1 and sufficiently high AAP-0.3. In order to obtain practical and applicable SAPs for use in hygiene, we blended SAP-1 and SAP-2 in different weight ratios to assemble SAP-1 + 2 clusters to fabricate gap structures among particles, providing both high RCW-1 and high AAP-0.3. The increased gaps between particles largely enhanced capillary diffusion, which was beneficial to rapid water absorption based on fluorescence observation in a capillary tube and on the glass slide. Real-time observation of the SAPs in contact with water provided evidence to support the SAP-1 + 2 design with adjustable SAP-1/SAP-2 ratios for various requirements in hygiene products. SAPs assembled in this way via inverse suspension polymerization can be applied to the usage domains requiring both high RCW-1 and AAP-0.3, such as infant diapers and napkins, and may be further developed into hemostatic materials. Compared to other SAPs developed for similar applications, ours achieve competitive RCW-1 and AAP-0.3 while being cheaper to produce due to the use of only AA instead of more expensive ingredients. Furthermore, we pioneered a way to observe the initial water absorption using fluorescent microscopy to record the capillary diffusion between gaps, which is beneficial to the development of rapidly absorbent polymers.

Glossary

Abbreviations

SAPs

superabsorbent polymers

RCW-1

retention capacity in water after 1 min

AAP-0.3

absorption against pressure at 0.3 psi

PAA

poly(acrylic acid)

AA

acrylic acid

EGDE

ethylene glycol diglycidyl ether

AIBA

2,2′-azobis(2-amidinopropane) dihydrochloride.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c09509.

• Detailed operation conditions in the inverse suspension polymerization and the related particle size analysis and SEM images (PDF)

• Video of water absorption for SAP1(AVI)

• (AVI)

• Video of water absorption for SAP2 (AVI)

• Video of water absorption for SAP-1 + 2 (AVI)

Author Contributions

J.C.C. and K.M.W. performed the polymerization, purification, characterization, and measurement; Z.Y.C. and Y.C.L. provided valuable discussion, sample verification, and measurement; Y.C.C. contributed to the concept formation, fulfilled the manuscript, and led the whole project.

This work was led by the YCC and funded by the Formosa Plastics Corporation with a R&D grant no. 1115065.

The authors declare no competing financial interest.