Introduction

Aqueous latex, in which polymer particles with sizes ranging from tens of nanometers to several micrometers are dispersed in water, is a material that plays an important role in our lives, even though these particles cannot be observed with the naked eye. As aqueous latex can be synthesized in water, it can be safely mass-produced and stored for a long time due to its high colloidal stability. The colloidal scale of aqueous latex provides it with a large specific surface area compared with that of bulk materials, thus improving the efficiency of its interactions with other molecules and enabling a rapid stimulus response. Owing to their beneficial properties, aqueous latexes are used in a wide range of applications, such as in molecular separation/transport materials [1,2,3,4,5], affinity materials [6, 7] and functional catalysts [8,9,10], as well as in two-dimensional and three-dimensional assemblies, including colloidal crystals [11,12,13], interface stabilizers [14,15,16,17], and stimuli-responsive sheets [18, 19].

Latex films obtained by drying particle dispersions are environmentally friendly materials that do not use volatile organic solvents (VOCs) and are applied as coatings and adhesives [20,21,22,23]. Much effort has been devoted to the discussion of the mechanism of latex film formation. Specifically, the particles are packed as the water evaporates, and when dried above the minimum film-formation temperature, the microparticles come into contact with each other and deform, and the polymer chains on the surface diffuse across the interfaces of the adjacent particles, i.e., the polymer chains interpenetrate to form a film [24, 25]. However, given that the particle interfaces remain inside the material, the mechanical performance of latex films is generally lower than that of corresponding bulk polymer materials, and they have not yet become a complete replacement for systems that use VOCs. Various approaches have been used to obtain mechanically stable latex films, such as thermal annealing and the addition of fusing assistants to promote interpenetration [26,27,28,29,30,31,32], the design of core‒shell particles [33,34,35,36], and interparticle cross-linking [37,38,39].

In this context, our group discovered that the introduction of a rotaxane cross-linker to the inside of particles toughens microparticle-based films [40]. Furthermore, we found that microparticle assemblies based on polymethyl acrylate (pMA), without any additives, are as tough as bulk pMA and can be disassembled into individual particles by immersion in a good solvent [41, 42]. To elucidate the nature of the toughening mechanism, a quantitative evaluation method for the interpenetration depth using small-angle X-ray scattering (SAXS) was developed and used to show the tendency for the fracture energy to increase with increasing interpenetration depth of the surface polymer chains [43, 44]. However, to fully understand the toughening mechanism, it is necessary to understand how the structure of microparticle-based films changes and fractures. Structural evaluation of microparticle-based films after deformation by scattering techniques [44,45,46] and microscopy [47, 48] has been reported. For example, SAXS and small-angle neutron scattering (SANS) provide information on the changes in the ordered structure of particles [46] and changes in the morphologies of the particles in the film [44]. Atomic force microscopy (AFM) can be used not only to visualize the arrangement and morphology of microparticles in films but also to evaluate changes in surface mechanical properties at the nanoscale [48]. Nevertheless, simultaneously evaluating the stress‒strain behavior and microscale deformation of microparticle-based films up to the point of fracture in real time remains challenging.

In this study, we attempted to establish a method to measure microscale structural changes in real time by combining optical microscopy (OM) with a stretching stage (Fig. 1). As a model for the dynamic evaluation of microparticle-based films up to failure, chemically cross-linked pMA microparticles of a size that is observable by OM were synthesized, and films exhibiting two types of crosslinking were formed, i.e., chemical crosslinking within the microparticle interiors and physical crosslinking between particles due to interpenetration of surface polymer chains.

Fig. 1
figure 1

Schematic and photograph of the setup used for the combination of optical microscopy with a stretching stage

Full size image

Materials and methods

Materials

Methyl acrylate (MA, purity 98%), potassium peroxydisulfate (KPS, 95%), 1,6-hexanediol dimethacrylate (HDD, 98%), sodium chloride (NaCl, 99.5%), ethanol (99.5%), and toluene (95.5%) were purchased from FUJIFILM Wako Pure Chemical Corporation (Japan) and used as received. Water was distilled and ion-exchanged with an EYELA SA-2100E1 before use in the experiments.

Microparticle synthesis

A round-bottom flask (1000 mL) was charged with water (590 g), MA (49.07 g), HDD (7.63 g) and NaCl (0.28 g). The monomer solution was heated to 70 °C under mechanical stirring (300 rpm) and purged with nitrogen gas for 30 min to remove oxygen. Then, KPS (0.32 g in 10 g of water) was injected into the flask via a syringe to initiate the polymerization. Owing to the formation of agglomerates, the polymerization was quenched 4 h after initiator injection by cooling to room temperature. Before the experiments, the obtained dispersion was purified via dialysis.

Field-emission scanning electron microscopy (FE–SEM)

The particle size and size uniformity of the microparticles were evaluated via FE–SEM (JSM-IT800SHL, Hitachi Ltd., Japan). To prepare the samples, 0.1 wt.% of pMA microparticles dispersed in 50 vol.% ethanol were spread on a glass Petri dish (Φ = 5 cm) with water poured on the surface. After 1 min, a cover glass was immersed in the liquid and slowly removed to allow the particles to adhere to the substrate [49]. The substrate was then sputter-coated with Pt/Pd (15 mA, 6 Pa, 80 s). After observation, the particle size was measured using the ImageJ software (version 1.54 d) package. The surface structure of the microparticle-based films was observed via FE‒SEM under the same sputtering conditions.

Differential scanning calorimetry (DSC)

DSC (DSC250, TA Instruments, USA) was used to evaluate the glass transition temperature (Tg) of the pMA microparticles. The particle dispersions were freeze-dried using an FDU-1200 (Tokyo Rikakikai Co., Ltd., Japan). Then, ~20 mg of the sample was placed into an aluminum pan. The temperature was increased twice between -30 °C and 100 °C at a rate of 10 °C/min, and the second measurement was used for analysis. Tg was defined as the intersection of the baseline of the heat-flow curve and the tangent to the curve after the inflection point.

Electrophoretic light-scattering (ELS) measurements

The zeta potential of the pMA microparticles was calculated via ELS measurements (Zetasizer NanoZS, Malvern, Zetasizer Software v.6.12) [5]. A 0.01 wt.% pMA-particle dispersion was prepared, and its ionic strength was adjusted to 1.0 mM using a 0.1 M NaCl solution. The samples were then allowed to equilibrate at 25 °C for 10 min. The electrophoretic mobility (\(\mu\)) of the particles was obtained via laser Doppler velocimetry with a scattering angle of 173° (20 measurements), and the zeta potential (\(\zeta\)) was calculated using the Smoluchowski equation (Eq. 1) [50].

$$\mu =\frac{{\varepsilon }_{{{{\rm{r}}}}}{\varepsilon }_{0}}{\eta }\zeta$$
(1)

Here, \({\varepsilon }_{{{{\rm{r}}}}}\) is the relative dielectric constant, \({\varepsilon }_{0}\) is the dielectric constant of the vacuum, and \(\eta\) is the viscosity; the three obtained \(\zeta\) values were averaged.

Film formation

The dispersions with 294 mg of pMA microparticles (~5 mL) were placed into a 3.5 cm × 3.5 cm × 1 mm silicon mold and dried at 40 °C in an incubator (LTI-2100, EYELA). The film was subsequently cut into a dumbbell shape (ISO 37-4) for elongation testing.

Gel content measurements

The gel content (GC) of the pMA microparticles was measured via Soxhlet extraction [44]. For that purpose, the films were dissolved in toluene for 12 h before Soxhlet extraction was performed at 170 °C for 10 h to dissolve the sol components. The insoluble components were separated via a cellulose-fiber filter (Thimble Filter No. 84, Advantec, retention particle size: 8 μm). After removing the toluene using a Buchi R-300 evaporator (Switzerland) and drying in air, the GC was determined from the remaining insoluble components via the following equation (Eq. 2):

$${{{\rm{GC}}}}=\left(\frac{{w}_{1}}{{w}_{2}}\right)\times 100$$
(2)

Here, w1 and w2 represent the weight of the gel (insoluble) fraction and the weight of the original film, respectively.

Tensile tests of the pMA microparticle-based films using optical microscopy

A stretching stage (Japan High Tech Co., Ltd.) was combined with an optical microscope (BX51, Olympus) to visualize the microscale structural changes that occur during uniaxial extension tests (tensile test speed: 10 mm/min). Owing to limitations of the experimental setup, such as the working distance of the objective lens, only the central part of the dumbbell-shaped test specimen could be observed. Changes in particle length and ordered structure were quantified via the ImageJ software (version 1.54 d) package.

Results and discussion

To visualize each microparticle of a film during elongation via OM, micron-sized pMA microparticles were prepared via soap-free emulsion polymerization in the presence of salt [51]. During polymerization, oligomers that form due to the reaction between the ionic initiator and a small amount of monomer dissolved in water at the early stages of polymerization serve as surfactants [52]. Therefore, the added salt may shield the electrostatic repulsion between the oligomers and promote the aggregation of oligomers, resulting in a reduction in the number of particle nuclei and an increase in the particle size. Here, 5 mol% of crosslinker was fed into the reaction mixture during polymerization to prepare a microparticle-based film with small interpenetration distances between the surface polymer chains. SEM revealed that the diameter of the obtained pMA microparticles (948 ± 108 nm) was sufficient to be visualized via OM (Fig. 2a). The Tg of the pMA microparticles derived from DSC measurements (16.7 °C) is similar to that reported previously (Fig. 2b) [41]. The GC (86%) confirmed the introduction of the crosslinker, given that pMA microparticles without a crosslinker have a GC < 1% [44]. The zeta potential of the pMA microparticles (-29 ± 0.4 mV) indicated that residues of the anionic thermal initiator were present on the surface.

Fig. 2
figure 2

a SEM image of pMA microparticles and associated size histogram and b heat flow as a function of temperature obtained from DSC measurements (Tg = 16.7 °C)

Full size image

Next, a pMA microparticle-based film was obtained by drying the dispersion. To promote the interpenetration of the polymer chains between microparticles, the film was dried at 40 °C, which is higher than the Tg of the microparticles [53]. The obtained pMA microparticle-based film was ~95 μm thick and transparent. The structure of the film surface was observed via FE–SEM, which revealed that many of the constituent microparticles were deformed and in contact with each other (Fig. 3). A previous evaluation via X-ray scattering revealed that the interpenetration depth of the surface polymer chains between the particles is ~2.9 nm in the case of a 5-mol%-crosslinked pMA microparticle-based film [44]. Moreover, free-standing films are considered to be stabilized by the interpenetration between particles.

Fig. 3
figure 3

FE‒SEM image of the surface structure of the pMA microparticle-based film

Full size image

Next, a uniaxial tensile test was performed via a stretching stage combined with OM to visualize the microscale structural changes in the film in real time (Fig. 1). During the initial state, the film showed some undulation, and although the image was partially out of focus, the film remained flat after stretching (Fig. 4a). Notably, the observation area of the film did not remain constant because the position moved due to uniaxial stretching. With increasing elongation ratio of the film (λmacro), the particle morphology changed to an elliptical shape parallel to the elongation direction. At the beginning of elongation (λmacro < 1.4) and at the end of deformation (λmacro > 2.2), the image drifted considerably, and the microparticles that formed the film could not be observed via OM. Finally, the film fractured at λmacro = 2.3 (Fig. 4b). The major axis length of the microparticles (Lparticle) was measured at several points within the film, and it increased proportionally to λmacro, regardless of the measurement position (Fig. 4c). The elongation ratio of the microparticles (λparticle) was calculated from the average Lparticle values and compared with the changes in the λmacro values. The slope of the resulting approximation line of the obtained plot (0.33) indicated a deviation from affine deformation (slope = 1) (Fig. 4d).

Fig. 4
figure 4

a OM images of pMA microparticle-based films at different elongation ratios. The bottom panels show magnified images of the regions indicated by blue circles. Lparticle refers to the major axis of the microparticles. b Stress vs. λmacro curve obtained from the stretching stage, c L particle measured at the positions corresponding to the images in (a) as a function of λmacro, and d a plot of λmacro vs. λparticle; affine deformation: slope = 1

Full size image

The affine deformability of latex films has previously been investigated through a static evaluation. For example, Rharbi et al. used SANS to investigate the mechanical properties and structural changes of latex films, which were locally reswollen with a solvent, under tensile deformation [45]. They showed that the films consisted of particles with a hydrophobic core and a hydrophilic shell obtained via copolymerization of styrene, butyl acrylate, and acrylic acid. They concluded that when the continuous layer is swollen and weak, the particles deform as if they are sliding between each other, whereas when the continuous layer is hard, the film exhibits affine deformability. Zhang et al. subjected a latex film consisting of a polystyrene/butyl acrylate copolymer to elongation deformation and evaluated the structural changes via SAXS [46]. They demonstrated that the microscale elongation ratio calculated from the periodic structure was smaller than the macroscale elongation ratio. They also showed that the deformation approached affine deformation upon thermal annealing and that the nonaffine deformability was compensated for by the sliding of the colloidal-crystal domains. The structural changes in pMA microparticle-based films were also evaluated via SAXS, which revealed that intraparticle cross-linking reduces the deformability of the particles [44]. In other words, it was argued that the sliding of the polymer chains near the surface occurs preferentially at low strain. The microparticles loaded with 5 mol% of a crosslinking agent used in our study consisted of particles with low deformability, and the interpenetration sites between the particles may act as soft continuous layers with low crosslinking density; thus, deformation should occur preferentially between particles.

We observed that with increasing λmacro, not only do the particle shapes become elliptical, but the particles also move apart from each other, resulting in the formation of gaps between them (Fig. 4a). To quantify the deformation, including that between particles, a group of several particles was defined as “local”, and the local elongation ratio (λlocal) was calculated via Eqs. (3) and (4) (Eqs. (3) and (4)) [54].

$${\varepsilon }_{{xy}}=\frac{2}{N{d}^{2}}{\sum }_{i=1}^{N}\left(\begin{array}{cc}{x}_{i}{x}_{i} & {x}_{i}{y}_{i}\\ {y}_{i}{x}_{i} & {y}_{i}{y}_{i}\end{array}\right)$$
(3)
$${P}^{-1}{\varepsilon }_{{xy}}P=\left(\begin{array}{cc}{\lambda }_{{{{\rm{local}}}}}^{2} & 0\\ 0 & {C}_{{{{\rm{local}}}}}^{2}\end{array}\right)$$
(4)

Here, εxy is the local strain tensor, N is the number of neighboring particles, xiyi is the relative vector from the center to a neighboring particle, P is an invertible matrix, and Clocal is the local compression ratio. In essence, we found that λlocal increases with increasing λmacro. In contrast to Lparticle, a significant location dependency was observed, and the difference between the maximum and minimum λlocal values increases with increasing λmacro (Fig. 5). The minimum value showed a similar change to that of λparticle, whereas the maximum value showed almost affine deformation compared with that of λparticle (slope = 0.72). Considering that the change in λparticle does not considerably depend on the location, a large λlocal value represents the area where the interpenetrating polymer chains between particles are pulled out and the particles begin to separate. In addition, the λlocal values become more distributed closer to the fracture strain, suggesting that this may contribute to the fracture of the film. One of the reasons for the high dispersion of the λlocal values is that the microparticle-array structure is more heterogeneous than the uniformity of the microparticle size due to the influence of colloidal crystal angles and structural defects. Lui et al. reported that in composite materials with hard and soft components, the deformation of the soft part tends to be more dominant [55]. In the case of pMA microparticle-based films, the interpenetration of surface polymer chains, i.e., the particle interface with physical crosslinking, may be more easily deformed than the chemically crosslinked pMA microparticles, which may also have influenced the increase in the location dependence of λlocal.

Fig. 5
figure 5

a Representative λlocal values at different positions when λmacro = 2.14 and b the relationship between λmacro, λlocal, and stress at each tensile stage

Full size image

Since it was not possible to observe the same location continuously, deformations such as the sliding of the colloidal-crystal domains discussed in a previous study [46] may have occurred outside the observation area. Nevertheless, this study establishes a method for simultaneously analyzing microscale film structure in real time along with stress‒strain behavior and can provide important insights into the toughening and recycling mechanisms of microparticle-based films.

Conclusions

In this study, a stretching stage was combined with an optical microscope to construct a system that allows evaluation of the microscale structural changes in a pMA microparticle-based latex film during a uniaxial tensile test. To evaluate the microstructural changes in the film until fracture, large-sized poly(methyl acrylate) (pMA) microparticles were synthesized via soap-free emulsion polymerization in the presence of salt with a crosslinker loading of 5 mol%. Observation of the structural changes in the obtained pMA-microparticle-based film during uniaxial tensile tests revealed that the deformation of the particles deviated from affine deformation, suggesting that the crosslinker introduced inside the particles reduced the deformability of the microparticles that formed the film. Furthermore, the deformation of a group of several (local) particles was more location-dependent than that of a single particle, and the maximum local elongation ratio showed deformation closer to affine deformation. This finding suggests that the removal of the superficial polymer chains that ensure interpenetration between particles and the resulting separation of the neighboring particles may contribute to the fracture of the film. The observation method established in this study allows a comparison of the macroscopic stress‒strain behavior of the film with the microscopic structural changes and can be expected to contribute to the creation of new design guidelines for tough microparticle-based films.