Crystallite Distribution Research Articles

Representation of Hashin–Shtrikman Bounds in Terms of Texture Coefficients for Arbitrarily Anisotropic Polycrystalline Materials

The present work generalizes the results of Bohlke and Lobos (Acta Mater. 67:324–334, 2014) by giving an explicit representation of the Hashin–Shtrikman (HS) bounds of linear elastic properties in terms of tensorial Fourier texture coefficients not only for cubic materials but for arbitrarily anisotropic linear elastic polycrystalline materials. Based on the HS bounds as given by Walpole (J. Mech. Phys. Solids 14(3):151–162, 1966) and tensor functions for the representation of the crystallite orientation distribution function, it is shown that the HS bounds are represented in terms of the exact same second- and fourth-order texture coefficients which appear in the consideration of the Voigt and Reuss bounds. The derived representations in terms of tensorial texture coefficients are valid for all symmetry classes in elasticity and present expressions highly attractive for the description of physical quantities in terms of tensorial variables. In order to make these results also accessible for the community of quantitative texture analysis, transformation relations between experimentally obtained Bunge’s or Roe’s coefficients and the tensorial texture coefficients are given. The representation of the present work offers a finite and low dimensional parametrization of the fully anisotropic Hashin–Shtrikman bounds, which can be used in inverse materials design problems in order to explore the set of possible materials properties or for the determination of optimal microstructural influence with respect to prescribed material properties. Examples for orthotropic polycrystals of cubic materials and transversely isotropic polycrystals of hexagonal materials (showing the connection and applicability of the results also to fiber orientation distributions) are discussed. Finally, an implementation in Mathematica® 11 of the HS bounds for arbitrarily anisotropic materials is offered, such that readers can reproduce all the results of this work and use them for their own purposes.

Pre-coalescence scaling of graphene island sizes

Graphene grown using cold-wall chemical vapor deposition on Cu surfaces follows a classical nucleation and growth mechanism. Following nucleation at the earliest growth stages, isolated crystallites grow, impinge, and coalesce to form a continuous layer. During the pre-coalescence growth regime, the size distributions of graphene crystallites exhibit scaling of the form N(s) = θ/⟨s⟩2 g(s/⟨s⟩), where s is the island area, θ is the graphene coverage, ⟨s⟩ is the average island area, N is the areal density, and g(x) is a scaling function. For graphene grown on Cu surfaces that have been annealed in a reducing Ar + H2 ambient, excellent data collapse onto a universal Avrami scaling function is observed irrespective of graphene coverage, surface roughness, or Cu grain size. This result is interpreted to indicate attachment-limited growth and desorption of diffusing C-containing species. Graphene grown on Cu surfaces that were annealed in a non-reducing environment exhibits a qualitatively different scaling function, indicating diffusion-limited growth with a lower attachment barrier combined with C detachment from the graphene edges.

Education program for controversial defect of recent X-ray instrument termed as a simultaneous small angle X-ray scattering and wide angle X-ray diffraction measuring instrument

Abstract Simultaneous rotations of sample and X-ray detected counter are needed to evaluate orientation distribution of crystallites and amorphous chains oriented predominantly parallel to the film surface in addition to exact diffraction peak profiles obtained without the complicated intensity corrections. The rotation mode is known as “θ–2θ scanning” system (θ: film, 2θ: counter). The system has been mainly used in research and development institutes. However, such instruments are not produced at present. Recently, small angle X-ray scattering (SAXS) and wide angle X-ray diffraction (WAXD) intensities have been measured by using X-ray beam generated along one direction. The brand name of the instrument is “a simultaneous SAXS and WAXD measuring instrument”. The X-ray beam generated by the instrument has surely high luminance providing high degree resolution of peak profiles by diffraction and/or scattering. The sample stage and detector, however, are fixed, since the intensities for SAXS and WAXD are obtained by the digital display of the number of X-ray photons detected on the imaging plate. Such optical system contains controversial defect on evaluating orientation of crystal planes parallel to the surface of films prepared by T-die and inflation methods as well as the exact profile. The imaging plate cannot detect the diffraction intensity from the crystal planes existing in the angle range between incident beam and Bragg angle associated with the diffraction peak position of the individual crystal planes.

Helium diffusion parameters of hematite from a single-diffusion-domain crystal

This contribution reports new parameters for helium diffusion in hematite useful for interpretation of cosmogenic 3He and radiogenic 4He chronometry. Fragments of a coarse, euhedral single crystal of hematite from Minas Gerais, Brazil were subjected to bulk step-heating helium diffusion experiments after proton irradiation to make a uniform distribution of 3He. Aliquots of three different grain sizes ranging from ∼300 to ∼700 μm in equivalent-sphere radius yielded helium diffusion activation energies Ea ∼ 170 kJ/mol, very similar to previous estimates for Ea in hematite. Uniquely in this specimen, diffusivity varies with the dimensions of the analyzed fragments in precisely the fashion expected if the diffusion domain corresponds to the physical grain. This contrasts with previous studies that concluded that the analyzed hematites consist of polycrystalline aggregates in which helium migration is governed by the size distribution of the constituent crystallites. These new data permit a direct estimate of the helium diffusivity at infinite temperature for hematite of ln(Do) = −0.66 ± 0.35 in cm2/s.The major implication of the new diffusion parameters is that hematite is very retentive of helium even at very small crystal sizes. For example, a 20 nm radius hematite crystal, at the smallest end of the size range so far described in dated polycrystalline hematite specimens, will retain more than 99% of its ingrown He over 1 Myr at 30 °C, and more than 90% over 100 Myr. Under most conditions, hematite is close to quantitatively helium-retentive on the Earth's surface, simplifying radiogenic and cosmogenic helium dating of this phase. In a system cooling at 10 °C/Myr, the 20 nm hematite crystal has a He closure temperature of ∼70 °C, similar to a typical ∼100 μm apatite crystal.Helium is likely held tightly in hematite owing to its dense hexagonal closest packing structure and absence of migration-enhancing channels. The isostructural minerals corundum and ilmenite are likely to be similarly helium retentive.

Photoinduced Reactions of Benzophenone in Biaxially Oriented Polypropylene

The photoinduced reactions of benzophenone (B) in biaxially oriented polypropylene (BOPP) were studied with nanosecond laser photolysis (N2 laser, λ337.1 nm). The first observed transient was a triplet state 3B*. Decay of 3B* led to formation of a radical pair (RP) of BH• and R•, where R• is a radical formed by hydrogen abstraction from BOPP (RH) by 3B*. We studied BOPP after the preheating for a short time in a temperature range 298-423 K, which is essentially lower than its melting point of 453 K. All measurements with not-heated and with preheated (annealed) BOPP were made at 298 K. A radical pair (RP) apparently decays as a contact pair 3[BH•, R•] in nonheated BOPP. A critical phenomenon takes place: dissociation of RP with a formation of free radicals in the polymer bulk is observed at preheating temperature Tcrit ≈ 403 K and at a higher T. The physical process of heating and cooling of BOPP apparently resulted in the restructuring of crystallites, their agglomeration, shrinking of the distribution of crystallites according to their sizes in BOPP. Overall BOPP becomes softer which manifests itself in the radical kinetics. The decay kinetics of 3B* and RP in the cage fits well the first-order law. Rate constants were obtained. Radicals BH•, which exit into the polymer bulk at temperatures of preheating T ≥ 403 K, decay by cross-termination according to the second-order law. A relatively high rate constant ∼108 M-1·s-1 for this reaction was obtained due to diffusion of BH• enclosed in the soft amorphous phase of BOPP. Properties of BOPP containing B were studied with ESR, DSC, IR, and WAXD.

Critical Role of Vertical Phase Separation in Small-Molecule Organic Solar Cells

An inverted device structure is a more stable configuration than a regular device structure for solution-processed organic solar cells (OSCs). However, most of the solution-processed small-molecule OSCs (SM-OSCs) reported in the literature used the regular device structure, and a regular device normally exhibits a higher efficiency than an inverted device. Herein, a representative small-molecule DR3TBDTT was selected to figure out the reason for photovoltaic performance differences between regular and inverted devices. The mechanisms for a reduced open-circuit voltage ( Voc) and fill factor (FF) in the inverted device were studied. The reduced Voc and FF is due to the vertical phase separation with excess [6,6]-phenyl-C71-butyric acid methyl ester near the air/blend surface, which leads to a reduction in build-in voltage and unbalanced charge transport in the inverted device. Another reason for the reduced FF is the unfavorable DR3TBDTT crystallite orientation distribution along the film thickness, which is preferential edge-on crystallites in the top layer of the blend film and the increased population of face-on crystallites in the bottom layer of the blend film. This study illustrates that the morphology plays a key role in photovoltaic performance difference between regular and inverted devices and provides useful guidelines for further optimization of the morphology of solution-processed SM-OSCs.

Structure-property relationships of basalt fibers for high performance applications

Abstract Continuous basalt fibers present a unique combination of good mechanical and functional properties at an attractive cost level. These characteristics can be useful in high performance polymer matrix composites where cost orientation and function integration are becoming priority objectives. The present work examines a currently unresolved problem of basalt fibers regarding high performance applications: the variation of fiber structural and mechanical properties due to inhomogeneity of the natural raw material and the spinning process. Commercial basalt fibers from seven different manufacturers were investigated regarding their fiber chemistry, diameter distribution and occurrence of defects and crystallites. In a second step, structure-property relations on different scales were established with single fiber, impregnated roving and composite tensile test results.

Structural characterization and functional correlation of Fe3O4 nanocrystals obtained using 2-ethyl-1,3-hexanediol as innovative reactive solvent in non-hydrolytic sol-gel synthesis

An original non-hydrolytic sol-gel approach, using 2-ethyl-1,3-hexanediol as reactive solvent, was proposed to synthetize nanostructured magnetite. Iron-oxide nanoparticles were prepared and studied as a function of the precursor-to-solvent ratio. The crystallization degree of nanoparticles was followed by the combined Rietveld and Reference Intensity Ratio method. This procedure has allowed the determination of both amorphous and crystalline content of nanomagnetite, using hematite as suitable internal standard. The results of Rietveld method show that the crystalline content decreases as the precursor-to-solvent is increased, ranging from 67 to 60 wt%. Information on the crystallite size-strain distribution and microstructural evolution of nanocrystals was supplied by line profile analysis of the powder diffraction patterns, employing the Whole Powder Pattern Modeling analysis: the obtained log-normal distribution curves become increasingly narrow and symmetrical, while nanoparticle microstrain increases as the precursor concentration is increased. The dimensional analysis of the Transmission Electron Microscopy images has allowed to obtain the nanoparticle grain-size distribution. Nanoparticle dimensions decreases from 15 to 9 nm increasing the precursor concentration. The comparison between the results of X-ray diffraction and microscopic characterization techniques highlighted the effect of several factors, such as size, shape and microstructure of magnetite nanoparticles, on their functional magnetic response. Magnetic characterizations show that magnetite nanoparticles are not in the superparamagnetic phase even at room temperature, independent of the precursor concentration. On the other hand, the room-temperature saturation magnetization, ranging from 73 to 60 emu/g, is a function of the nanoparticle average size, decreasing as the precursor concentration increases.

Fabrication and electrochemistry characteristics of nickel-doped diamond-like carbon film toward applications in non-enzymatic glucose detection

This research work focused on the fabrication of nickel-doped diamond-like carbon (DLC) films and their characteristics including of surface morphology, microstructure, and electrochemical aiming at applications in non-enzymatic glucose detection. Novel nanodiamond target was employed in unbalanced magnetron radio-frequency co-sputtering process to prepared high quality Ni-doped DLC thin film at room temperature. TEM analysis reveals a highly uniform distribution of Ni crystallites in amorphous carbon matrix with fraction ranged from 3 to 11.5 at.% which is considered as active sites for the glucose detection. Our cyclic voltammetry measurements using 0.1 M H2SO4 solution demonstrated that the as-prepared Ni-doped DLC films possess large electrochemical potential window of 2.12 V, and this was also observed to be significantly reduced at high Ni doping level owing to lower sp3 fraction. The non-enzymatic glucose detection investigation indicates that the Ni-doped DLC thin film electrode prepared under 7 W of DC sputtering power on Ni target possesses good detecting performance, high stability, and high sensitivity to glucose concentration up to 10 mM, even with the existence of uric acid and ascorbic acid. The peak current was observed to be proportional to glucose concentration and scanning rate, demonstrating highly reversibility redox process of the film electrode and glucose.

Polylactide single-polymer composites with a wide melt-processing window based on core-sheath PLA fibers

Polylactide (PLA) single polymer composites (SPCs) were prepared using PLA nonwovens made of core-sheath PLA fibers as raw materials by hot-pressing. The core and sheath materials were poly(l-lactide) (PLLA) and PLA with d-lactide of about 10mol% (PLA90), respectively. The melt processing window of PLA SPCs reached more than 40°C. The effects of hot-pressing temperature on the crystallinity, the size distribution of crystallites, the lamellar thickness and mechanical property of PLA SPCs were significant while those of hot-pressing pressure were small. The strong interfacial adhesion between matrix and reinforcement led to high mechanical properties of PLA SPCs. For pure PLA materials prepared at 180°C, the tensile strength (σb), elongation at break (εb), the work of rupture (W) and impact strength (αcU) were only 19MPa, 1.8%, 0.15J, and 2.8kJ/m2, respectively while those of the SPCs prepared at 130–160°C were 47–65MPa, 15–32%, 9–22J, and 14.9–67.2kJ/m2, respectively. With the increase of hot-pressing temperature, the σb, W and αcU decreased.

On Elastic Symmetry Identification for Polycrystalline Materials

The products made by the forming of polycrystalline metals and alloys, which are in high demand in modern industries, have pronounced inhomogeneous distribution of grain orientations. The presence of specific orientation modes in such materials, i.e., crystallographic texture, is responsible for anisotropy of their physical and mechanical properties, e.g., elasticity. A type of anisotropy is usually unknown a priori, and possible ways of its determination is of considerable interest both from theoretical and practical viewpoints. In this work, emphasis is placed on the identification of elasticity classes of polycrystalline materials. By the newly introduced concept of “elasticity class” the union of congruent tensor subspaces of a special form is understood. In particular, it makes it possible to consider the so-called symmetry classification, which is widely spread in solid mechanics. The problem of identification of linear elasticity class for anisotropic material with elastic moduli given in an arbitrary orthonormal basis is formulated. To solve this problem, a general procedure based on constructing the hierarchy of approximations of elasticity tensor in different classes is formulated. This approach is then applied to analyze changes in the elastic symmetry of a representative volume element of polycrystalline copper during numerical experiments on severe plastic deformation. The microstructure evolution is described using a two-level crystal elasto-visco-plasticity model. The well-defined structures, which are indicative of the existence of essentially inhomogeneous distribution of crystallite orientations, were obtained in each experiment. However, the texture obtained in the quasi-axial upsetting experiment demonstrates the absence of significant macroscopic elastic anisotropy. Using the identification framework, it has been shown that the elasticity tensor corresponding to the resultant microstructure proves to be almost isotropic.

Crystal growth patterns in DC and pulsed plated galvanic copper films on (1 1 1), (1 0 0) and (1 1 0) copper surfaces

Copper films for printed circuit board applications have to be fine-grained to achieve even filling of vias. Electroplated Cu films on roll annealed Cu substrates may have unacceptably large epitaxial crystals. Here galvanic films were plated on oriented single-crystal Cu substrates from an additive-free electrolyte, as well as DC plating and pulse reverse (PR) plating with additives. The distribution of crystallite orientations was mapped with XRD and compared with the microstructure determined by SEM. For the additive-free bath on [111] and [100] oriented surfaces a gradual transition from epitaxial to polycrystalline is seen, while films on [110] substrates are persistently epitaxial. Without bath additives, twinning is the main mechanism for the transition to polycrystalline texture. For DC plating, additives (carriers, accelerators and levelers) promote fine-grained films with isotropic grain orientations, with films on [110] substrates being partially isotropic. Plating with carriers and accelerators (no leveler) yields films with many distinct crystallite orientations. These orientations result from up to five steps of recursive twinning. PR plating produces isotropic films with no or very few twins (〈111〉 and 〈100〉 substrates, respectively), while on 〈110〉 oriented surfaces the deposits are about 20% epitaxial.

Tailor-Made Dispersion and Distribution of Stereocomplex Crystallites in Poly(l-lactide)/Elastomer Blends toward Largely Enhanced Crystallization Rate and Impact Toughness.

Stereocomplex (SC) crystallites, formed between poly(l-lactide) (PLLA) and poly(d-lactide), exhibit great potential to substantially enhance crystallization rate of PLLA-based materials as an eco-friendly nucleating agent. However, the nucleation efficiency of the SC crystallites is still far below an expected level, mostly on account of their strong aggregation tendency in PLLA/PDLA melts. Herein, taking PLLA/poly(ethylene-methyl acrylate-glycidyl methacrylate) (E-MA-GMA) blends as an example, we report a unique and facile strategy to control the dispersion and distribution of SC crystallites within the PLLA matrix by using elastomeric E-MA-GMA as carrier for the incorporation of PDLA. To do this, PDLA was first blended with E-MA-GMA or chemically grafted onto the E-MA-GMA. During subsequent melt-blending of PLLA and the E-MA-GMA/PDLA master batch, the PDLA chain clusters predispersed in the E-MA-GMA phase can gradually migrate into PLLA matrix and then collaborate with the matrix chains to form large amounts of tiny and well-dispersed SC crystallites. Compared with the SC-crystallite agglomerates formed by the direct melt-blending of PLLA and PDLA components, such tiny SC crystallites are much more effective in accelerating PLLA matrix crystallization. More interestingly, when PDLA chains are grafted onto the EMA-GMA, the formed SC crystallites tend to preferentially distribute at the blend interface and thus induce not only optimal nucleation efficiency but also superior impact toughness because these interface-localized SC crystallites can also serve as bridges to enhance interface adhesion. This work could open a new avenue in designing heat-resistant and supertough PLLA blends via controllable construction of SC crystallites.

Size, shape, orientation and crystallinity of cellulose Iβ by X-ray powder diffraction using a free spreadsheet program

X-ray powder diffraction is one of the most commonly used methods in cellulose science. This technique is used to identify the cellulose allomorphs, their crystallinity, and the size of their crystallites. In this paper, a novel model is introduced that implicitly takes into account the shape and size of cellulose Iβ crystallites in the interpretation of powder diffractograms. Because of the limited amount of data in cellulose powder patterns, this model focuses on a small number of adjustable parameters. The method hypothesizes that cellulose Iβ crystallites are straight crystalline rods with superelliptical cross-sections. This superellipse is a parametric curve that can, for example, describe various crystallite shapes as rectangles or ellipses. Additionally, preferred orientation along the (0 0 1) crystallographic planes can be modelled using the March–Dollase approach. The simulated background has a semi-empirical form. An initial model comprised cellulose Iβ crystallites and the amorphous background. A second model comprised a biphasic distribution of crystallites and the same amorphous background. In this second model, large cellulose Iβ crystallites coexisted with more slender crystallites, usually less than 20 A in lateral size. Cellulose IVI nanocrystals were selected as a modeling construct to represent these small and distorted forms of native cellulose. Both models produced simulations in excellent agreement with the experimental measurements.

Size and shape distributions of primary crystallites in titania aggregates

The primary crystallite size of titania powder relates to its properties in a number of applications. Transmission electron microscopy was used in this interlaboratory comparison (ILC) to measure primary crystallite size and shape distributions for a commercial aggregated titania powder. Data of four size descriptors and two shape descriptors were evaluated across nine laboratories. Data repeatability and reproducibility was evaluated by analysis of variance. One-third of the laboratory pairs had similar size descriptor data, but 83% of the pairs had similar aspect ratio data. Scale descriptor distributions were generally unimodal and were well-described by lognormal reference models. Shape descriptor distributions were multi-modal but data visualization plots demonstrated that the Weibull distribution was preferred to the normal distribution. For the equivalent circular diameter size descriptor, measurement uncertainties of the lognormal distribution scale and width parameters were 9.5% and 22%, respectively. For the aspect ratio shape descriptor, the measurement uncertainties of the Weibull distribution scale and width parameters were 7.0% and 26%, respectively. Both measurement uncertainty estimates and data visualizations should be used to analyze size and shape distributions of particles on the nanoscale.

Stable monolithic hole-conductor-free perovskite solar cells using TiO2 nanoparticle binding carbon films

Using TiO2 nanoparticles as binders, highly conductive carbon films were prepared and used to fabricate monolithic hole-conductor-free perovskite solar cell in air. Carbon films were condensed by dehydration reaction between surface hydroxyls from both TiO2 nanoparticles and carbon particles, which is beneficial for adhesion ability and conductivity of carbon film, and the formation and filling properties of perovskite crystallites in the monolithic structure of “FTO/CTiO2/M-TiO2/M-ZrO2/Carbon” (C-Compact, M-Mesoporous). A reduced sheet resistance from 10 to 2.74 Ω/sq was obtained for carbon films annealed under relatively low temperature annealing. Confined distribution of perovskite crystallites in the region containing M-TiO2/M-ZrO2 layers and bottom of carbon film was obtained from 350 °C annealing, which helped to achieve upgraded power conversion efficiency (reverse scan) from 5.40 (±1.32) % to 9.21(±1.04) % (averaged) and 11.07% (champion) at testing area of ∼0.18 cm2, as well as stability of 1000 h in relative high humidity environment (RH:70–90%). The air-compatible fabrication process, sound device ability as well as low-cost of carbon materials make such monolithic device structure realistic candidate for mass production of perovskite solar cells.

Homogenization and Materials Design of Anisotropic Multiphase Linear Elastic Materials Using Central Model Functions

It is shown in the present work that for linear elastic multiphase materials the orientation average of a stiffness tensor can be given in a closed form if central model functions are taken into consideration for the description of the material orientation distribution. This holds for arbitrary numbers of material constituents and for arbitrary degree of anisotropy of any of the constituents. The first-order bounds of Voigt and Reuss, the geometric average of Aleksandrov and Aisenberg and, specifically, the Hashin–Shtrikman bounds are given in closed forms depending on the single crystal stiffnesses of the material constituents, on their volume fractions and, for elasticity, on a central orientation and two scalar parameters per central model function. The central orientation and the two scalar parameters, referred to as texture eigenvalues in the present work, reflect the complete influence of the corresponding central model function on the elastic properties. The set of all admissible texture eigenvalues is derived. The closed form expressions can be used, e.g., for efficient and low dimensional homogenization procedures taking the crystallographic texture into account or, e.g., for the determination of estimates for unknown single crystal behavior. It is further shown, that central crystallite orientation distributions exist which conserve the anisotropy of the single crystal behavior but invert the direction of the anisotropy. The closed form expressions documented in the present work are used for an exemplary materials design problem in order to determine favorable volume fractions and texture eigenvalues for the orientation distributions of considered constituents for given prescribed anisotropic properties.

Toward a better understanding of conjugated polymer blends with non-spherical small molecules: coupling of molecular structure to polymer chain microstructure

Abstract

Optical, structural and electrical characterizations of stacked Hf-based and silicon nitride dielectrics

High-k stacked dielectric structures were fabricated by a combination of RF magnetron sputtering and plasma-enhanced chemical vapor deposition. Their structural properties were studied versus deposition and annealing conditions by means of attenuated total reflection and high-resolution transmission electron microscopy techniques. All samples demonstrated smoothed surface (with a roughness below 1nm) and abrupt interfaces between the different stacked layers. No crystallization of Hf-based layers was observed after annealing at 800°C for 30min, demonstrating their amorphous nature and phase stability upon annealing. Uniform capacitance–voltage characteristics were measured along the wafers for all stacks. Besides, after round-voltage sweep they demonstrate significant flat-band voltage hysteresis due to charging of the stack caused by carrier injection from the substrate. These phenomena were found to be more pronounced for the stacks with pure HfO2 layers. The stacked structures were implemented for the formation of Ge nanocrystals by means of ion implantation followed by the thermal treatment mentioned above. It was found that the spatial distribution of Ge crystallites in stacked dielectrics affects significantly their electrical properties including the trapping of charge.

Covalent/crystallite cross-linked co-network hydrogels: An efficient and simple strategy for mechanically strong and tough hydrogels

Covalent/crystallite cross-linked co-network hydrogels have been prepared using epoxy and PVA through a cyclic freezing-thawing process. The PVA/epoxy hydrogels show enhanced mechanical strength and toughness. PVA/epoxy hydrogels with 4wt% epoxy loading display maximum tensile strength and toughness of 1.1MPa and 2838kJ/m3 respectively. The fracture toughness of PVA/epoxy hydrogels ranges from 160 to 450J/m2. Radius of gyration and fractal information of the hydrogels were obtained by fitting the SAXS data to the Guinier and power law models. The enhanced mechanical properties are attributed to the increase in covalent bonding and decrease in crystallite distribution with an increase in epoxy content. However a larger hysteresis is shown for PVA/epoxy hydrogels due to irreversible destruction of covalent bonds between epoxy and PVA.