Experimental Advances in the Real-Time Recording of Cross-Linking Alginate In Situ Gelation: A Review

Alginate-based hydrogels are promising smart materials widely employed in the food, bioengineering, and energy sectors. The development and optimization of their production require a thorough knowledge of gelation. In recent years, advanced experimental procedures have been developed for real-time cross-linking alginate reaction monitoring. Novel methods, such as customized rheometric setups, enable the recording of mechanical properties and morphological changes during hydrogel formation. These innovative techniques provide important insights into the gelation stages, the reaction rate, the diffusion of cross-linker to polymer chains, and the homogeneity of the gelling structures. Based on real-time experimental data, kinetic models are developed to enhance comprehension of the reaction mechanism and, eventually, to predict the gelation progress. The aim is to enable better control of the characterization of both the complex gelation and the propagated structures. This review aspires to present a comprehensive overview and evaluation of the breakthrough innovations of the real-time in situ recording of cross-linking alginate hydrogels and bead formation. A detailed analysis of the pioneering experimental developments provides a deep comprehension of the alginate gelation, including the parameters controlling the reaction.


Introduction
The favorable characteristics of alginate, such as biocompatibility, biodegradability, and antimicrobial properties [1][2][3], render it an ideal material for developing 'smart gelling' products with a wide range of applications in bioengineering, food, and energy domains [4][5][6]. The polysaccharide is primarily extracted from the cell walls of brown seaweeds [7,8]. To easily tailor the structure and characteristics of the macromolecule, biological production by two genera of soil bacteria, i.e., Azotobacter and Pseudomonas, is feasible as well [9,10]. Alginate is composed of (1-4)-linked β-D-mannuronic (M) and α-L-guluronic (G) acid monomers arranged in different sequences, i.e., MM-, GG-, and MG-, along the negatively charged long linear chain, as shown in Figure 1a [11,12]. Gelling networks are formed in the presence of alginate in aqueous solutions [13,14]. At a pH lower than the pKa of uronic acid, hydrogels are formed, while the reaction of the polysaccharide with divalent cations results in well-known ionic hydrogels [15,16]. The ionotropic gelation is based on the 'egg-box model, presented in Figure 1b,c, which arises from the initial chelation of cations with the alginate chains and the following interchain association, [17]. It critically depends on the physicochemical characteristics of the reactants, such as their concentration, the kind of cations, and the molecular weight, Mw, and M/G of alginate, as shown in Figure 2 [18]. The divalent metals bind mainly with G-and MGsequences of the polymer backbone through physical cross-linking [19][20][21]. The dimerization of alginate chains is induced, and multimer structures are formed following the follow ing the zipper m echanism [22,23]. T he selectivity of the polysaccharid e w ith ca increases in the ord er M g 2+ < C o 2+ , N i 2+ , Z n 2+ < C a 2+ < Sr 2+ < B a 2+ < C d 2+ < C u 2+ < Pb 2+ [24 T he reaction is controlled by the stoichiom etric m ole ratio f = [X 2+ ]/[C O O -], w here X 2 resents the d ivalent cations. A s f increases, the gelation is faster and stiffer structure form ed [26,27]. M oreover, the alginate gelation can originate from chem ica cross-lin For instance, som e bivalent cations and trivalent m etals, as w ell as other chem ical m ed iates, such as epichlorohydrin, cross-link w ith alginate chains through strong cov bond s [20,21]. A dd itionally, to fabricate m aterials w ith enhanced m echanical stre alginate can create double netw ork (D N ) hydrogels w ith other m acrom olecules, su gelatin or polyacrylam id e [28,29]. T he breakthrough applications of hyd rogels d ictate the extensive com prehensi the form ation m echanism [30,31]. A d vanced m ethod s, such as rheology and light gered techniques, allow record ing the gelation kinetics in situ [32,33]. M onitoring the cess d eterm ines the stages of the reaction, the factors that evolved in each stage, an tem poral m icro-structural rearrangem ents [34,35]. B ased on real-tim e experim ental kinetic m od els are d eveloped to enable a d eeper und erstand ing and accurate pred i of the gelation progress. T his leads to d eveloping strategies to optim ize the characte tion of com plex gelation [36,37]. T he present review dem onstrates and evaluates th vanced in situ m ethod s for m onitoring the alginate cross-linking gelation. N ovel con based on rheology and tim e-resolved techniques, such as sm all angle X -ray scatt (SA X S), m onitor the alginate hydrogel and bead com position. Inform ation is provid the tem poral evolution of the m echanical properties of the system to d etect the d om param eters affecting the reaction and further elaborate the d iffusion m echanism of c linker to alginate solution. A lso, the visualization of gelation captures the changes i gelling structures and investigates their hom ogeneity. W ith a thorough assessm ent o in situ m ethod s, the challenges and opportunities of real-tim e gelation record ing ar The breakthrough applications of hydrogels dictate an extensive comprehension of the formation mechanism [30,31]. Advanced methods, such as rheology and light-triggered techniques, allow for recording of the gelation kinetics in situ [32,33]. Monitoring the process determines the stages of the reaction, the factors which evolve in each stage, and the temporal micro-structural rearrangements [34,35]. Based on real-time experimental data, kinetic models are developed to enable a deeper understanding and accurate prediction of the gelation progress. This leads to the development of strategies to optimize the characterization of complex gelation [36,37]. The present review demonstrates and evaluates the advanced in situ methods for monitoring the alginate cross-linking gelation. Novel concepts based on rheology and time-resolved techniques, such as small angle X-ray scattering (SAXS), monitor the alginate hydrogel and bead composition. Information is provided on the temporal evolution of the mechanical properties of the system to detect the dominant parameters affecting the reaction and further elaborate the diffusion mechanism of the cross-linker to alginate solution. Additionally, the visualization of gelation captures the changes in the gelling structures and investigates their homogeneity. With a thorough assessment of the in situ methods, the challenges and opportunities of real-time gelation recording are discussed to shed light on potential future research that could further advance this domain.

Introduction to In Situ Gelation Methods
The innovative applications of polymers require detailed knowledge of their molecular characteristics and reaction mechanisms [39,40]. The conventional ex situ preparation of hydrogels limits their study to a steady-state of kinetics, impeding a comprehensive investigation of their dynamic structural and mechanical properties [41]. In contrast, the in situ methods provide an opportunity to record the gelation in its original position from the initiation of the reaction [42,43]. As Figure 3 shows, in situ gelation can facilitate the precise control of the volume and concentration of reactants to modulate the reaction accurately. Furthermore, the detection of gel point, i.e., the instant that the transition from solution to gel phase occurs [44], is achieved along with the recording of the dynamic mechanical response and structural development of hydrogels. In this way, the kinetic stages and the parameters defining each reaction step are determined. Finally, kinetic models can be developed based on real-time experimental data to describe and predict the reaction.  Table 1, which is at the end of the review, includes the methods developed during the last years to characterize the cross-linking alginate in situ gelation. In brief, the mechanical properties of alginates are controlled by advanced rheometric and light-triggered techniques, as well as by cantilever sensors. Moreover, innovative procedures based on time-resolved methods, such as small-angle Χ-ray scattering, Fourier-transform infrared spectroscopy (FTIR), and dark field microscopy (DFM), are designed to detect the

Introduction to In Situ Gelation Methods
The innovative applications of polymers require detailed knowledge of their molecular characteristics and reaction mechanisms [39,40]. The conventional ex situ preparation of hydrogels limits their study to a steady-state of kinetics, impeding a comprehensive investigation of their dynamic structural and mechanical properties [41]. In contrast, the in situ methods provide an opportunity to record the gelation in its original position from the initiation of the reaction [42,43]. As Figure 3 shows, in situ gelation can facilitate the precise control of the volume and concentration of reactants to modulate the reaction accurately. Furthermore, the detection of gel point, i.e., the instant that the transition from solution to gel phase occurs [44], is achieved along with the recording of the dynamic mechanical response and structural development of hydrogels. In this way, the kinetic stages and the parameters defining each reaction step are determined. Finally, kinetic models can be developed based on real-time experimental data to describe and predict the reaction.

Introduction to In Situ Gelation Methods
The innovative applications of polymers require detailed knowledge of their molecular characteristics and reaction mechanisms [39,40]. The conventional ex situ preparation of hydrogels limits their study to a steady-state of kinetics, impeding a comprehensive investigation of their dynamic structural and mechanical properties [41]. In contrast, the in situ methods provide an opportunity to record the gelation in its original position from the initiation of the reaction [42,43]. As Figure 3 shows, in situ gelation can facilitate the precise control of the volume and concentration of reactants to modulate the reaction accurately. Furthermore, the detection of gel point, i.e., the instant that the transition from solution to gel phase occurs [44], is achieved along with the recording of the dynamic mechanical response and structural development of hydrogels. In this way, the kinetic stages and the parameters defining each reaction step are determined. Finally, kinetic models can be developed based on real-time experimental data to describe and predict the reaction.  Table 1, which is at the end of the review, includes the methods developed during the last years to characterize the cross-linking alginate in situ gelation. In brief, the mechanical properties of alginates are controlled by advanced rheometric and light-triggered techniques, as well as by cantilever sensors. Moreover, innovative procedures based on time-resolved methods, such as small-angle Χ-ray scattering, Fourier-transform infrared spectroscopy (FTIR), and dark field microscopy (DFM), are designed to detect the  Table 1, which is at the end of the review, includes the methods developed during the last years to characterize the cross-linking alginate in situ gelation. In brief, the mechanical properties of alginates are controlled by advanced rheometric and light-triggered techniques, as well as by cantilever sensors. Moreover, innovative procedures based on time-resolved methods, such as small-angle X-ray scattering, Fourier-transform infrared spectroscopy (FTIR), and dark field microscopy (DFM), are designed to detect the microstructural and macrostructural rearrangements during the gelation. A detailed description of these techniques is presented in Sections 3 and 4.
In situ experimental procedures are used to form alginate hydrogels or alginate beads. The hydrogels are generated by the diffusion of the cross-linking agent into alginate solution [45], while the beads derive from the drop-wise methods in which alginate solution drops into pools of gelling solution [46]. The source of the cross-linker determines whether the gelation is internal or external [47]. In the first case, insoluble salt is added to the alginate solution, and the cations diffuse to the chains of the polymer when the appropriate pH conditions are achieved by a catalyst such as D-glucono-δ-lactone (GDL). The slow hydrolysis of GDL releases the protons, and they react with the biopolymer [48,49]. In the second case, the gelation is achieved by the direct exposure of alginate solution to an active form of cations, specifically, a soluble salt [50,51]. The external gelation implies a faster reaction than the internal mechanism, but the slower internal gelation leads to more homogeneous gelling structures [52,53].

Conventional Rheometric Setups
Real-time recording of gelation kinetics provides valuable insights into the crosslinking reactions of alginate during hydrogel formation. Rheology, the study investigating the flow of matter [54], is extensively used to evaluate the progress of gelation in dynamically developing systems [55,56]. Specifically, oscillatory time sweeps are performed to study the temporal evolution of the mechanical properties of materials undergoing timedependent structural rearrangements [57]. The mechanical properties are represented by the storage (G ) and loss (G ) moduli, showing the elastic and the viscous responses of the system, respectively. When G > G , the material behaves as a solution, while when G < G , the gel behavior predominates [54,55,57]. At the crossover point of the moduli, the first microgel structures are created and the reaction initiates [54]. For this reason, it can be called the gel point, which is accurately determined by in situ rheological experiments, leading to the detection of the precise gelation time, in contrast to the vial inversion test, which detects gelation time by inverting the vials until the material no longer flows [58,59]. Over the years, the most common method for conducting these dynamic experiments has been the utilization of conventional rheometric plates [34,35,60]. Typically, a uniform mixture of reactants is prepared ex situ, and it is then applied to the bottom plate of the rheometer directly after mixing, where it is subjected to dynamic oscillatory tests [61,62]. The recording of modulus kinetics is effective and successful [63,64]. This rheological procedure is mainly applied to study internal gelation, as the slow reaction provides enough time for sample placement and initiation of the oscillatory experiments within the solution state of the material [65].
The study of Funami et al. [66] is remarkable, utilizing internal gelation to investigate the kinetics under different physicochemical characteristics of reactants. Alginate hydrogels rich in G-residues are more elastic in the presence of higher CaCO 3 concentrations (15 mM), while the elasticity increases in those rich in M-residues in the lowest availability of cations (3.75 mM). They obtain rod-like and network-like conformations, respectively, as shown in the additional atomic force microscopy images. Based on long-term oscillatory time sweeps, the authors propose the following first-order kinetic equation to describe the temporal evolution of the storage modulus, G : Equation (1) has been applied to determine the gelation time (t 0 ), the rate constant (k t ), and the saturated storage modulus (G sat ) of alginate hydrogels, i.e., the equilibrium modulus as t reaches infinity [66,67]. In their following research work, a simpler equa- tion with one exponential factor was employed to describe the internal alginate gelation adequately [67].
Similarly, Fernández Farrés et al. [68] have investigated the in situ formation of alginate-Ca 2+ fluid hydrogels using a four bladed vane in cup on a Kinexus Rheometer to control the shear profile. During the reaction, the viscosity increased due to the propagation of 'eggbox structures, while the shear rate caused a break-up in the network, and micro-gelled particles were formed. Under constant concentrations of reactants, the lower shearing (100 and 200 s −1 ) results in higher viscosities, while the higher ones (400 and 600 s −1 ) may cause a delay in the association of the initially formed particles and decrease the reaction rate. Based on real-time data of G obtained by dynamic oscillatory experiments with cone-and-plate geometry directly after the formation of structures at different shear rates, kinetic models are proposed to describe the gelation. An exponential equation applied to the high-Mw alginate hydrogels shows that the kinetic constant is the same for all structures, meaning that, at the time of shear removal, the strength rate is independent of the external forces, since the processing period is higher than the gelation time. For the reaction of low-Mw alginates, a linear first-order model has been employed, revealing that the ordering rate increases when hydrogels are formed at low shear rates, which accelerates the development of particles [68].
Another interesting study reports a customized mixing of reactants by developing a syringe-induced technique, one which utilizes internal gelation [69].
In brief, Larsen et al. [69] prepare one syringe containing soluble alginate and one with insoluble alginate/gelling ion particles, as shown in Figure 4a,b. Through a three-way connector, as presented in Figure 4c, the soluble alginate is transferred to the other syringe, and, upon mixing of reactants, it is placed through a channel onto the standard plate of the rheometer to record the gelation which occurs according to the mechanism in Figure 4d,e. The effect of cations on gelation kinetics can be controlled, and the proposed exponential kinetic equation, describing the temporal evolution of G , shows that the gelling rate is determined by the release rate of cations from the particles [69].
Furthermore, the conventional rheometric setups serve to study the temperature control kinetics of cross-linking alginate gelation due to the attached Peltier system [70,71]. Goudoulas et al. [72] have successfully studied the phase-transition kinetics of binary alginate-gelatin mixtures by applying isothermal time sweeps in a temperature range (5-20 • C) using a parallel-plate configuration on an Anton Paar MCR 502 rheometer. Additionally, the standard rheometric configurations can record the reaction of alginate with a covalent cross-linking agent upon ex situ mixing of reactants, since the sol-gel transition occurs relatively slowly. This has been achieved by Li et al. [73] in order to study the reaction of N,O-carboxymethyl chitosan with oxidized alginate without additional cross-linkers, as well as by Yang et al. [74] to record the gelation of alginate with adipic acid dihydrazide through coupling agents. However, they cannot provide adequate monitoring of the external gelation, which occurs instantaneously, i.e., within the first seconds of mixing the reactants. ling ion particles, as shown in Figure 4a,b. Through a three-way connector, as presented in Figure 4c, the soluble alginate is transferred to the other syringe, and, upon mixing of reactants, it is placed through a channel onto the standard plate of the rheometer to record the gelation which occurs according to the mechanism in Figure 4d,e. The effect of cations on gelation kinetics can be controlled, and the proposed exponential kinetic equation, describing the temporal evolution of G′, shows that the gelling rate is determined by the release rate of cations from the particles [69]. prepare one syringe containing soluble alginate and one w ith insoluble alginate/gelling 173 ion particles, as show n in Figures 4a and 4b. T hrough a 3-w ay-connector, presented in 174 Figure 4c, the soluble alginate is transferred to the other syringe, and, upon m ixing of 175 reactants, it is placed through a channel on the stand ard plate of the rheom eter to record 176 the gelation w hich occurs accord ing to the m echanism in Figures 4d and 4e. T he effect of 177 cations on gelation kinetics can be controlled, and the proposed exponential kinetic equa-178 tion, d escribing the tem poral evolution of G´, show s that the gelling rate is determ ined by 179 the release rate of cations from the particles [69].

Advanced Rheometric Setups
Advanced rheometric methods have been designed to enhance the investigation of the mechanical response in diffusion-controlled alginate hydrogels [75]; for instance, when crafted polysaccharide is exposed to light, conformational changes occur, resulting in gelling networks [76]; Bonino et al. [77] have utilized the light emitting diode (LED) bottom plate of a TA stress-controlled rheometer to monitor the gelation of alginate chemically modified with methacrylate groups when a photoinitiator presents in the system. This method facilitates the detection of the gel point and the evaluation of the reaction kinetics. Higher UV intensities accelerate the sol-gel transition and increase the storage (G ) and loss (G ) moduli at a constant concentration of the system. Additionally, based on shortterm UV oscillatory experiments, an empirical exponential model is applied to obtain the moduli at steady-state, which is found to be critically dependent on the degree of methacrylation [77]. Subsequently, the same research group observed that in addition to cross-linking due to the light-induced method, there is evidence of micro-structural development in the absence of UV light, i.e., dark curing. This is attributed to the acidic pH of the mixture, which facilitates the CaCO 3 dissociation and the cross-linking with alginate chains. Subjected or not to dark curing, the final properties of gelling formations are unaffected, reaching the same modulus plateau, but the gelation rate slows down when UV is not continuously applied to the gelation [78].
The external cationic alginate gelation exhibits rapid kinetics and forms heterogeneous structures, facts making it challenging to in situ monitor the mechanical properties using the conventional rheometric plates. To address this issue, innovative customized rheometric setups have been developed. In 2016, Mahdi et al. [79] demonstrated a modified lower rheometric plate, as presented in Figure 5a, facilitating the initiation of reaction at the instrument. It consists of a petri dish with filter paper impregnated with CaCl 2 solution and a hydrated dialysis membrane on the top. The alginate solution is placed there, preventing its absorption by the filter paper. The fast initiation and the subsequent slower evolvement of the reaction can effectively be recorded, as shown in Figure 5b. Additionally, the systematic increase of moduli in the presence of more cations in the system can be recorded. Interestingly, the design enables the monitoring of hydrogel degradation by replacing the CaCl 2 filter with one containing a calcium chelator [79], while it has been utilized to study the effect of the physiological fluids on the mechanical response of cross-linking hydrogels [80]. Based on these advantages, a rheo-dissolution cell has been developed controlling both the in situ gelation and the drug release profile of gelling systems [81]. In 2020, Besiri et al. [82] introduced an advanced setup for mixing the reactants on the rheometer during the external gelation, one resembling a batch reactor operation. It consists of two parts with an inner void when coupled together. A cross-linking agent fills the internal space with a volumetric syringe connected to a side-feed hole. At the initiation of the rheological measurement, CaCl2 is injected through micro-holes into the alginate solution placed on the top of the setup, and the reaction begins. As shown in Figure 6a,c, the design can be easily adaptable to various rheometers with a flexible arrangement of micro-holes to study the process thoroughly. Figure 6b presents that the setup can successfully record the initial and quasi-steady-state kinetic stages and assists in the determination of the fast sol-gel transition, i.e., the crossover of moduli, where the first gelling structures are created. Additionally, the stoichiometry of reactants can be monitored. Higher concentrations and injected volumes of CaCl2 increase the growth rate of G′, signifying that the diffusion of cations controls the gelation. By controlling the angular frequency (ω) it is shown in Figure 6d that higher ω results in a sharper increase of G′, i.e., the formation of stiffer hydrogels, which is attributed to the acceleration of the propagation of multimer structures. The configuration can form typical soft hydrogels after 1 h of reaction, as it is observed in Figure 6e by frequency sweep experiments where the moduli are developed at a slow rate independently of the applied ω [82,83]. Further, in situ investigation demonstrates that alginates with higher Mw and M/G increase the elasticity of the network due to the higher availability of free polymer chains for cross-linking, as shown in Figure 7 [84]. Currently, the novel setup is utilized to provide insights into external alginate gelation concerning the spatial distribution of the ions', while a two-kernel equation, consisting of one exponential and one logarithmic factor, seems to describe and predict the reaction adequately [85]. In 2020, Besiri et al. [82] introduced an advanced setup for mixing the reactants on the rheometer during the external gelation, one resembling a batch reactor operation. It consists of two parts with an inner void when coupled together. A cross-linking agent fills the internal space with a volumetric syringe connected to a side-feed hole. At the initiation of the rheological measurement, CaCl 2 is injected through micro-holes into the alginate solution placed on the top of the setup, and the reaction begins. As shown in Figure 6a,c, the design can be easily adaptable to various rheometers with a flexible arrangement of micro-holes to study the process thoroughly. Figure 6b presents that the setup can successfully record the initial and quasi-steady-state kinetic stages and assists in the determination of the fast sol-gel transition, i.e., the crossover of moduli, where the first gelling structures are created. Additionally, the stoichiometry of reactants can be monitored. Higher concentrations and injected volumes of CaCl 2 increase the growth rate of G , signifying that the diffusion of cations controls the gelation. By controlling the angular frequency (ω) it is shown in Figure 6d that higher ω results in a sharper increase of G , i.e., the formation of stiffer hydrogels, which is attributed to the acceleration of the propagation of multimer structures. The configuration can form typical soft hydrogels after 1 h of reaction, as it is observed in Figure 6e by frequency sweep experiments where the moduli are developed at a slow rate independently of the applied ω [82,83]. Further, in situ investigation demonstrates that alginates with higher Mw and M/G increase the elasticity of the network due to the higher availability of free polymer chains for cross-linking, as shown in Figure 7 [84]. Currently, the novel setup is utilized to provide insights into external alginate gelation concerning the spatial distribution of the ions', while a two-kernel equation, consisting of one exponential and one logarithmic factor, seems to describe and predict the reaction adequately [85]. In addition to the innovative rheometric designs, interfacial shear rheology has been utilized to control the in situ generation of alginate foam hydrogels [86]. Ben Djemaa et al. [86] propose a gas-induced cross-linking reaction instead of the conventional internal mechanism for hydrogel formation. Gas steam, supplied with CO2 through a flow meter, is humidified and connected to a double-wall ring geometry containing an alginate solution with CaCO3 particles. The flow of CO2 in the interface of the pre-gelling system triggers the gelation. G′ and G″ increase rapidly within 90 min, since the diffusion of CO2 causes the acidification of the interface, providing the suitable pH conditions for the cations' release to alginate chains. Later, the moduli tend to reach a plateau, possibly meaning that the acidification process and the gel front have been developed in the entire system, i.e., from the interface to the bottom of the trough. The researchers state that the physicochemical characteristics of reactants and the CO2/air ratio determine the gelation progress and need further investigation [86].

Alternative Techniques beyond Rheology
Apart from employing customized rheometric techniques to understand alginate gelation, researchers propose to use cantilever sensors, as well. A cantilever sensor consists of a beam anchored at one end and having a free extension into space, and it is made from a thin film with mechanical flexibility. The movement of the beam, along with the combination of piezoelectric elements, enables the detection of the deflection of the cantilever, which results in the measurement of the mechanical response of materials. This experimental approach minimizes the time required for monitoring the dynamic properties, while the gelation can occur under high frequencies [87]. Recently, Ηaring et al. [88] have evaluated alginate-Ca 2+ kinetics with a piezoelectric-excited millimeter cantilever (PEMC) sensor [89]. In the implementation, it is immersed in a petri dish containing alginate, in which CaCl2 is injected 5 mm from the anchor of the cantilever upon stabilization of the signal. In this way, the resonant frequency is recorded and used to calculate the G′ and G″ of the developed gelling structures through fluid-structure interaction models. The gelation kinetics can be evaluated similarly to oscillatory rheological experiments, but the crossover of moduli is not detected due to the high frequency, i.e., G′ > G″ from the reaction's initiation [88].
The overall comprehension of the cross-linking gelation requires not only the recording of mechanical properties but also the investigation of the structural conformation. [90,91]. Yamamoto et al. [92] have designed a customized sampling system combined with In addition to the innovative rheometric designs, interfacial shear rheology w as uti-275 lized to control the in situ generation of alginate foam hydrogels [86]. Ben Djemaa et al. 276 [86] propose a gas-induced cross-linking reaction instead of the conventional internal 277 mechanism for the hydrogel formation. Gas steam, supplied w ith CO2 through a flow me-278 ter, is humidified and connected to a double-w all ring geometry containing an alginate 279 solution w ith CaCO3 particles. The flow of CO2 in the interface of the pre-gelling system 280 triggers the gelation. G´ and G´´ increase rapidly w ithin 90 min since the diffusion of CO2 281 causes the acidification of the interface, providing the suitable pH conditions for the cati-282 ons release to alginate chains. Later, the moduli tend to reach a plateau, possibly meaning 283 that the acidification process and the gel front have been developed in the entire system, 284 i.e., from the interface to the bottom of the trough. The researchers state that the physico-285 chemical characteristics of reactants and the CO2/air ratio determine the gelation progress 286 and need further investigation [86]. 287 288 A part from employing customized rheometric techniques to understand the alginate 289 gelation, researchers propose to use cantilever sensors, too. A cantilever sensor consists of 290 a beam anchored at one end and having a free extension into space, and it is made from a 291 thin film w ith mechanical flexibility. The movement of beam along w ith the combination 292 of piezoelectric elements enables the detection of the deflection of cantilever, w hich results 293 in the measurement of the mechanical response of materials. This experimental approach 294 minimizes the required time for monitoring the dynamic properties, w hile the gelation 295 can occur under high frequencies [87]. Recently, Ηaring et al. [88] have evaluated the algi-296 nate-Ca 2+ kinetics w ith a piezoelectric-excited millimeter cantilever (PEM C) sensor [89]. In 297 the implementation, it is immersed in a petri dish containing alginate, w here CaCl2 is in-298 jected 5 mm from the anchor of the cantilever upon stabilization of the signal. In this w ay, 299 the resonant frequency is recorded and used to calculate the G´ and G´´ of the developed 300 gelling structures through fluid-structure interaction models. The gelation kinetics can be 301 evaluated similarly to oscillatory rheological experiments, but the crossover of moduli is 302 not detected due to the high frequency, i.e., G´ > G´´ from the reaction initiation [88]. 303 The overall comprehension of the cross-linking gelation requires not only the record-304 ing of mechanical properties but also the investigation of the structural conformation. [90, 305 91]. Yamamoto et al. [92] have designed a customized sampling system combined w ith 306 SA XS to in situ monitor the polymer chains association in alginate-Ca 2+ structures devel-307 oped through competitive ligand exchange (CLEX) cross-linking [93,94]. After mixing 308 In addition to the innovative rheometric designs, interfacial shear rheology has been utilized to control the in situ generation of alginate foam hydrogels [86]. Ben Djemaa et al. [86] propose a gas-induced cross-linking reaction instead of the conventional internal mechanism for hydrogel formation. Gas steam, supplied with CO 2 through a flow meter, is humidified and connected to a double-wall ring geometry containing an alginate solution with CaCO 3 particles. The flow of CO 2 in the interface of the pre-gelling system triggers the gelation. G and G increase rapidly within 90 min, since the diffusion of CO 2 causes the acidification of the interface, providing the suitable pH conditions for the cations' release to alginate chains. Later, the moduli tend to reach a plateau, possibly meaning that the acidification process and the gel front have been developed in the entire system, i.e., from the interface to the bottom of the trough. The researchers state that the physicochemical characteristics of reactants and the CO 2 /air ratio determine the gelation progress and need further investigation [86].

Alternative Techniques beyond Rheology
Apart from employing customized rheometric techniques to understand alginate gelation, researchers propose to use cantilever sensors, as well. A cantilever sensor consists of a beam anchored at one end and having a free extension into space, and it is made from a thin film with mechanical flexibility. The movement of the beam, along with the combination of piezoelectric elements, enables the detection of the deflection of the cantilever, which results in the measurement of the mechanical response of materials. This experimental approach minimizes the time required for monitoring the dynamic properties, while the gelation can occur under high frequencies [87]. Recently, Haring et al. [88] have evaluated alginate-Ca 2+ kinetics with a piezoelectric-excited millimeter cantilever (PEMC) sensor [89]. In the implementation, it is immersed in a petri dish containing alginate, in which CaCl 2 is injected 5 mm from the anchor of the cantilever upon stabilization of the signal. In this way, the resonant frequency is recorded and used to calculate the G and G of the developed gelling structures through fluid-structure interaction models. The gelation kinetics can be evaluated similarly to oscillatory rheological experiments, but the crossover of moduli is not detected due to the high frequency, i.e., G > G from the reaction's initiation [88].
The overall comprehension of the cross-linking gelation requires not only the recording of mechanical properties but also the investigation of the structural conformation [90,91]. Yamamoto et al. [92] have designed a customized sampling system combined with SAXS to in situ monitor the polymer chains' association in alginate-Ca 2+ structures developed through competitive ligand exchange (CLEX) cross-linking [93,94]. After mixing two alginate solutions, the one containing a cross-linking agent (CaEDTA) and the other an exchange ion (ZnEDDA), Zn 2+ ions are exchanged with Ca 2+ ions, which are subsequently available to cross-link with polymer chains. As shown in Figure 8a, syringe pumps injected the solutions in a sample cell so that an interface between them was created, and the reaction was controlled only by the diffusion of reactants. The X-ray irradiated the interface in short time intervals to continuously monitor the process. The Kratky plots presented in Figure 8b show that q 2 I increases linearly with q (I the intensity, and q the magnitude of scattering), meaning that the polymer chain is in a single coil condition at pH = 7, while the maximum value at 0.5 nm −1 indicates the chain association in the first 2 min of the experiment. The researchers have observed that, at higher pH, the signal of SAXS is weaker due to higher aggregation and less cross-links of alginate chains. Based on this experimental data, a two-component broken rod model is proposed to evaluate the cross-sectional form of junction zones [92]. Recently, Pragya et al. [95] probed the dynamic structures of an alginate-acrylamide double network in the presence of both cationic and covalent cross-linking agents by applying a series of time-induced methods, such as FTIR, UV-vis and Raman spectroscopy. In situ hydrogel characterization after the direct mixing of reactants indicates that the gelation is developed in two phases. Initially, physical entanglements of polymers and covalent cross-linking occur. Later, the cross-linking interactions control the reaction [95]. SAXS to in situ monitor the polymer chains' association in alginate-Ca 2+ structures developed through competitive ligand exchange (CLEX) cross-linking [93,94]. After mixing two alginate solutions, the one containing a cross-linking agent (CaEDTA) and the other an exchange ion (ZnEDDA), Zn 2+ ions are exchanged with Ca 2+ ions, which are subsequently available to cross-link with polymer chains. As shown in Figure 8a, syringe pumps injected the solutions in a sample cell so that an interface between them was created, and the reaction was controlled only by the diffusion of reactants. The X-ray irradiated the interface in short time intervals to continuously monitor the process. The Kratky plots presented in Figure 8b show that q 2 I increases linearly with q (I the intensity, and q the magnitude of scattering), meaning that the polymer chain is in a single coil condition at pH = 7, while the maximum value at 0.5 nm −1 indicates the chain association in the first 2 min of the experiment. The researchers have observed that, at higher pH, the signal of SAXS is weaker due to higher aggregation and less cross-links of alginate chains. Based on this experimental data, a two-component broken rod model is proposed to evaluate the cross-sectional form of junction zones [92]. Recently, Pragya et al. [95] probed the dynamic structures of an alginate-acrylamide double network in the presence of both cationic and covalent cross-linking agents by applying a series of time-induced methods, such as FTIR, UV-vis and Raman spectroscopy. In situ hydrogel characterization after the direct mixing of reactants indicates that the gelation is developed in two phases. Initially, physical entanglements of polymers and covalent cross-linking occur. Later, the cross-linking interactions control the reaction [95]. In situ visualization of the process provides important insights into the morphological rearrangements during the reaction [96,97]. Chueh et al. [98] have observed the formation of hydrogels in microchannels upon the release of cations to alginate chains from a caged calcium complex through UV light exposure. The non-gelled alginate is removed from the system through washing, and the gelling structures are dissolved in the presence of EDTA, resulting in the recovery of the flow in channels. The method controls the flow of the system in microfluidic devices applicable to in vitro 3D cell culture. Similarly, Braschler et al. [99,100] have monitored the gel front in an alginate-Ca 2+ system with a sequential gel layer synthesis in a microfluidic chip, using fluorescent polymer combined with the theoretical study. They reported that the diffusive depletion zone affects the concentration profile on the gel front, while higher flow rates decrease the gel's growth rate. This may be attributed to an interplay between the more alginate chains diffusing in the gel front, which require more Ca 2+ ions, and the washing out of partially cross-linked tw o alginate solutions, the one containing a cross-linking agent (C aED TA ) and the other 309 an exchange ion (Z nED D A ), Z n 2+ ions are exchanged w ith C a 2+ ions, available subse-310 quently to cross-link w ith polym er chains. A s show n in Figure 8a, syringe pum ps injected 311 the solutions in a sam ple cell so that an interface betw een them is created and the reaction 312 is controlled only by the d iffusion of reactants. T he X -ray irrad iates the interface in short 313 tim e intervals to continuously m onitor the process. T he K ratky plots presented in Figure 314 8b show that q 2 I increases linearly w ith q (I the intensity and q the m agnitud e of scatter-315 ing), m eaning that the polym er chain is in a single coil cond ition at pH =7, w hile the m ax-316 im um value at 0.5 nm -1 ind icates the chain association in the first 2 m in of the experim ent. 317 T he researchers have observed that at higher pH , the signal of SA X S is w eaker d ue to 318 higher aggregation and less cross-links of alginate chains. B ased on this experim ental 319 d ata, a tw o-com ponent broken rod m od el is proposed to evaluate the cross-sectional form 320 of junction zones [92]. R ecently, Pragya et al. [95] probed In situ visualization of the process provid es im portant insights into the m orphologi-333 cal rearrangem ents d uring the reaction [96,97]. C hueh et al. [98] have observed the for-334 m ation of hydrogels in m icrochannels upon the release of cations to alginate chains from 335 a caged calcium com plex through U V light exposure. The non-gelled alginate is rem oved 336 from the system through w ashing, and the gelling structures are d issolved in the presence 337 of ED TA , resulting in the recovery of the flow in channels. T he m ethod controls the flow 338 of the system in m icrofluid ic d evices, applicable to in vitro 3D cell culture. Sim ilarly, 339 B raschler et al. [99,100] have m onitored the gel front in an alginate-C a 2+ system w ith a 340 sequential gel layer synthesis in a m icrofluid ic chip using fluorescent polym er com bined 341 w ith the theoretical stud y. T hey have reported that the d iffusive d epletion zone affects 342 the concentration profile on the gel front, w hile higher flow rates d ecrease the gel grow th 343 rate. T his m ay be attributed to an interplay betw een the m ore alginate chains d iffusing in 344 the gel front, w hich require m ore C a 2+ ions, and the w ashing out of partially cross-linked 345 chains d uring the d eposition of layers. O verall, these d ecrease the concentration of cations 346 in the reaction zone [99,100]. N unam aker et al. [101] have visualized d iffusion-based ge-347 lation by dropping C aC l2 into an alginate solution consisting of a com plex ind icator. T he 348 In situ visualization of the process provides important insights into the morphological rearrangements during the reaction [96,97]. Chueh et al. [98] have observed the formation of hydrogels in microchannels upon the release of cations to alginate chains from a caged calcium complex through UV light exposure. The non-gelled alginate is removed from the system through washing, and the gelling structures are dissolved in the presence of EDTA, resulting in the recovery of the flow in channels. The method controls the flow of the system in microfluidic devices applicable to in vitro 3D cell culture. Similarly, Braschler et al. [99,100] have monitored the gel front in an alginate-Ca 2+ system with a sequential gel layer synthesis in a microfluidic chip, using fluorescent polymer combined with the theoretical study. They reported that the diffusive depletion zone affects the concentration profile on the gel front, while higher flow rates decrease the gel's growth rate. This may be attributed to an interplay between the more alginate chains diffusing in the gel front, which require more Ca 2+ ions, and the washing out of partially cross-linked chains during the deposition of layers. Overall, these decrease the concentration of cations in the reaction zone [99,100]. Nunamaker et al. [101] have visualized diffusion-based gelation by dropping CaCl 2 into an alginate solution consisting of a complex indicator. The gel rate is detected by the color change of the indicator from dark red to yellow in 50% of hydrogel, and it is influenced by the physicochemical properties of reactants, i.e., the higher the cationic concentration, the faster the gel's development.
Interestingly, Secchi et al. [102] have investigated the microscopic dynamics of crosslinking gelation by applying photo correlation imaging (PCI), a light scattering method, on a gelling system in which CaCl 2 slowly permeates through a membrane to an alginate solution. Additionally, the speckle motion in PCI controls the local flow field within the sample, revealing that the gel front is regulated by the diffusion of cations and molecules and critically depends on the convection of reactants through the local flow [102]. Another captivating research effort utilizes dark field microscopy (DFM) to record the progress of the gel front during an alginate cross-linking reaction [103,104]. As shown in Figure 9a, two microscope slides with an alginate droplet in the center, in the gap between them, are gently pressed together to create a flow cell. CaCl 2 is injected in this space at the periphery of the cell, and the diffusion-controlled reaction initiates. The flux of cations occurs in the radial direction, as Figure 9b shows. In this way, the gel front increases from the periphery to the center of alginate droplet during the cross-linking reaction, as can be observed in the DFM images in Figure 9d. The difference between gelled, i.e., narrow bright band, and ungelled alginate is distinct. This visualization is utilized to calculate the velocity of the gel front. Figure 9c shows that the gel front velocity is higher when more cations are available. In this way, the distance of non-gelled alginate solution from the periphery of the cell sharply increases due to the high number of cross-links formed in the reaction zone. Finally, a reaction-diffusion model is applied to describe and predict the dynamic gel front for a variety of geometries in which the gelation is developed [104]. chains during the deposition of layers. Overall, these decrease the concentration of cations in the reaction zone [99,100]. Nunamaker et al. [101] have visualized diffusion-based gelation by dropping CaCl2 into an alginate solution consisting of a complex indicator. The gel rate is detected by the color change of the indicator from dark red to yellow in 50% of hydrogel, and it is influenced by the physicochemical properties of reactants, i.e., the higher the cationic concentration, the faster the gel's development. Interestingly, Secchi et al. [102] have investigated the microscopic dynamics of crosslinking gelation by applying photo correlation imaging (PCI), a light scattering method, on a gelling system in which CaCl2 slowly permeates through a membrane to an alginate solution. Additionally, the speckle motion in PCI controls the local flow field within the sample, revealing that the gel front is regulated by the diffusion of cations and molecules and critically depends on the convection of reactants through the local flow [102]. Another captivating research effort utilizes dark field microscopy (DFM) to record the progress of the gel front during an alginate cross-linking reaction [103,104]. As shown in Figure 9a, two microscope slides with an alginate droplet in the center, in the gap between them, are gently pressed together to create a flow cell. CaCl2 is injected in this space at the periphery of the cell, and the diffusion-controlled reaction initiates. The flux of cations occurs in the radial direction, as Figure 9b shows. In this way, the gel front increases from the periphery to the center of alginate droplet during the cross-linking reaction, as can be observed in the DFM images in Figure 9d. The difference between gelled, i.e., narrow bright band, and ungelled alginate is distinct. This visualization is utilized to calculate the velocity of the gel front. Figure 9c shows that the gel front velocity is higher when more cations are available. In this way, the distance of non-gelled alginate solution from the periphery of the cell sharply increases due to the high number of cross-links formed in the reaction zone. Finally, a reaction-diffusion model is applied to describe and predict the dynamic gel front for a variety of geometries in which the gelation is developed [104].

In Situ Alginate Hydrogels in Membranes
The in situ control of cross-linking alginate hydrogels in membranes provides important information for their filtration, which is advantageous for industrial and environmental applications [105]. Hassan et al. [106] have obtained the pseudo-first-order kinetics of multimembrane alginate hydrogels in capillary columns at different temperatures. The gelation is developed rapidly in the beginning and at a slow rate across longer times due to the high density of the network of the gelled membrane [106]. Additionally, Sioutopoulos et al. [107] have investigated in situ the mechanical properties of fouled alginate segments on reverse osmosis membranes using a stress-controlled rheometer. The researchers reported that the elasticity increases when low concentration of cations flow to the membrane, possibly due to additional strong cross-links of Na + with alginate macromolecule [107]. Epstein et al. [108] have recently proposed an advanced configuration combining microrheology with microcopy to control both the mechanical and conformational changes of a foulant layer at a customized alginate-Ca 2+ membrane cell connected to a pressurized feed vessel. The mechanical properties are calculated by the particle-tracking method. At higher fluxes, compaction of the foulant layer occurs, and stiffening of the material is observed [108].

In Situ Real-Time Recording of Alginate Bead Formation
The broad implementation of alginate beads in environmental and biological applications highlights their significance [109,110]. Electrodispersion reactors and electrospray methods are only a few of the techniques applied to form cross-linking beads [111,112]. The enhanced comprehension of their synthesis makes the in situ control of their reaction kinetics imperative in order to evaluate their dynamic properties [113][114][115]. Lee et al. [116] injected an alginate droplet into baths of calcium sources with mixing times up to 4000 s. An osmotic gradient is created between the beads and the solution bath, which is responsible for cation diffusion to the gelling matrix. By rheological investigation of beads of 8 mm diameter, it was observed that, after 100 s of mixing, the mechanical properties of alginate-CaCl 2 beads were constant, meaning that the preparation time is short compared to this of calcium lactate and calcium gluconate sources [116]. Additionally, a microfluidic flow-focusing device was employed to produce alginate microparticles, which were then injected into a CaCl 2 tank. When continuous stirring was applied, the diffusion of cations to alginate particles was accelerated, and their faster solidification was achieved, as opposed to the particles gelled in the non-stirred tank [117].
Moreover, visualized extrusion-dripping methods are utilized to capture the shape and size of alginate-Ca 2+ beads, which are affected by the physicochemical characteristics of solutions, the size of the droplets, and the collecting distance [118,119]. Haldar et al. [120][121][122] have developed the configuration presented in Figure 10a. As alginate drops into a calcium solution bath, a high-speed imaging system records the formation of beads. As observed in the sequential snapshots of Figure 10b during gelation, the crater is propagated in the liquid surface of cross-linking agent, while the dynamic structural arrangements are distinct. The crater is restored to its shape upon the injection of the alginate droplet into the CaCl 2 solution [120]. Furthermore, Stößlein et al. [123] have constructed a customized sample holder on a texture analyzer, as depicted in Figure 11a. The configuration creates inhomogeneous alginate beads with an alginate-Ca 2+ shell and a core of pure cross-linking agent. Upon the beads' formation, the beads are pressurized, and their mechanical response is obtained as a function of time, as shown in Figure 11b. At approx. 300 s, a plateau was recorded indicating the two stages of kinetics. The authors also observed a decrease in the volume at 300 s, indicating that the fast gelation will not occur after 300 s [123]. Recently, Posbeyikian et al. [124] have controlled the gel front in alginate-Ca 2+ beads using an optical-video method, as presented in Figure 12a-d. The configuration comprises Furthermore, Stößlein et al. [123] have constructed a customized sample holder on a texture analyzer, as depicted in Figure 11a. The configuration creates inhomogeneous alginate beads with an alginate-Ca 2+ shell and a core of pure cross-linking agent. Upon the beads' formation, the beads are pressurized, and their mechanical response is obtained as a function of time, as shown in Figure 11b. At approx. 300 s, a plateau was recorded indicating the two stages of kinetics. The authors also observed a decrease in the volume at 300 s, indicating that the fast gelation will not occur after 300 s [123]. Furthermore, Stößlein et al. [123] have constructed a customized sample holder on a texture analyzer, as depicted in Figure 11a. The configuration creates inhomogeneous alginate beads with an alginate-Ca 2+ shell and a core of pure cross-linking agent. Upon the beads' formation, the beads are pressurized, and their mechanical response is obtained as a function of time, as shown in Figure 11b. At approx. 300 s, a plateau was recorded indicating the two stages of kinetics. The authors also observed a decrease in the volume at 300 s, indicating that the fast gelation will not occur after 300 s [123]. Recently, Posbeyikian et al. [124] have controlled the gel front in alginate-Ca 2+ beads using an optical-video method, as presented in Figure 12a-d. The configuration comprises Recently, Posbeyikian et al. [124] have controlled the gel front in alginate-Ca 2+ beads using an optical-video method, as presented in Figure 12a-d. The configuration comprises a vibration-stabilized stereoscopic microscope with a digital camera and a specialized chamber to position the bead for undisturbed image focus during cross-linking. Based on this technique, a fast automatic protocol is provided for the characterization of alginate beads, such as calculating their volume, which can be correlated to their mechanical properties. Furthermore, Lin et al. [125] have observed the in situ shrinkage of alginatebased emulsion gel beads. It seems that the increase of Young s modulus during gelation is combined with water loss and, consequently, shrinkage, leading to the formation of compact gelling structures. To conclude, the present section clarifies that alginate beads are produced mainly by extrusion-dripping methods, which can provide accurate visualizations of the formation process. Nevertheless, it is very challenging to characterize in situ the microstructural conformations. The Table 1 presents cumulates the representative studies of Sections 3 and 4 for the real-time recording of alginate gelation. a vibration-stabilized stereoscopic microscope with a digital camera and a specialized chamber to position the bead for undisturbed image focus during cross-linking. Based on this technique, a fast automatic protocol is provided for the characterization of alginate beads, such as calculating their volume, which can be correlated to their mechanical properties. Furthermore, Lin et al. [125] have observed the in situ shrinkage of alginate-based emulsion gel beads. It seems that the increase of Young′s modulus during gelation is combined with water loss and, consequently, shrinkage, leading to the formation of compact gelling structures. To conclude, the present section clarifies that alginate beads are produced mainly by extrusion-dripping methods, which can provide accurate visualizations of the formation process. Nevertheless, it is very challenging to characterize in situ the micro-structural conformations. The Table 1 presents cumulates the representative studies of Sections 3 and 4 for the real-time recording of alginate gelation.   [124] have controlled the gel front in alginate-C a 2+ bead using an optical-vid eo m ethod presented in Figures 12a-d . T he configuration com prises vibration-stabilized stereoscopic m icroscope w ith a d igital cam era and a specialize cham ber to position the bead for und isturbed im age focus d uring cross-linking. B ased o this technique, a fast autom atic protocol is provid ed for the characterization of alginat bead s, such as calculating their volum e, w hich can be correlated to the m echanical prop erties. Furtherm ore, L in et al. [125] have observed the in situ shrinkage of alginate-base em ulsion gel bead s. It seem s that the increase of Y oung' s m od ulus d uring gelation is com bined w ith w ater loss and , consequently, shrinkage lead ing to the form ation of com pac gelling structures. C onclud ing, the present section clarifies that alginate bead s are pro d uced m ainly by extrusion-dripping m ethod s, w hich can provid e accurate visualizatio of the form ation process. N evertheless, it is very challenging to characterize in situ th m icro-structural conform ations.

Conclusions and outlook
T he present review highlights the significance of real-tim e cross-linking alginate i situ gelation record ing. D em onstrating such ad vanced experim ental m ethod s provid es a assessm ent of the reaction progress and the m ain factors that evolved in each stage. C us tom ized rheom etric setups and m icroscopical and structural tim e-resolved technique have been d eveloped over the last years to control cross-linking gelation. T herefore, bot m echanical and m orphological properties can be record ed d uring the form ation of alg  Incubation of alginate in internal cross-linker on rheometer In situ evaluation of gelation behavior [66] LED bottom rheometric plate for UV cross-linking gelation Real-time evaluation of mechanical properties 2011 [77] Rheological characterization upon direct mixing of reactants Evaluation of gelation speed 2012 [116] In situ rheological characterization of fouled alginate membrane Assessment of the parameters manifesting alginate fouling layer 2013 [107] Fluid hydrogel formation on rheometer Controlling the shearing profile on gelation kinetics 2014 [68] UV irradiation on rheometer for photo-activated ionic gelation Monitoring of the two-step gelation mechanism [78] Rheological investigation upon ex situ mixing of reactants Recording of gelation kinetics 2015 [69] Petri dish with filter paper and a dialysis membrane on the rheometer Recording of alginate-Ca 2+ kinetics 2016 [79] Visible light-curable alginate Control of furfuryl alginate gel formation 2020 [76] Rheometric setup for injection of Ca 2+ into alginate through micro-holes Detailed recording of cross-linking gelation [82] In situ micro-rheology of foulant membrane surface Visualization of gelation on a petri dish Determination of sol-gel transition and gelling time 2011 [101] Photo correlation imaging during alginate gelation Investigation of gelling kinetics formation 2013 [102] Flow cell between two microscope slides Real-time studying of gel front 2016 [103,104] Microscopy on a drop-wise method Microscopical characterization of alginate-Ca 2+ beads 2017 [119] High-speed imaging system on drop impact gelation Effect of drop kinetic energy on gelation 2018 [120,121] Custom setup for optical video microscopy Measurement of alginate bead volume change over time 2021 [124] Light-triggered methods UV light in microchannels Progress of UV-triggered gelation 2010 [98] Capillary columns Capillary formation of ionotropic alginate membranes Propose equation for gelation kinetics 2013 [106] Cantilever sensors Dynamic-mode cantilever sensors on alginate gelation Real-time monitoring of gelation 2020 [87,88] Texture analyzer Customized bead formation on a texture analyzer In situ mechanical texture evaluation of alginate beads 2019 [123] SAXS Small-angle X-ray scattering oncompetitive ligand exchangegelation Determination of the local structure of hydrogels 2019 [92] Time-induced methods FTIR, Uv-vis, and Raman spectroscopy on alginate-acrylamide gelation In situ mapping of gel structure 2021 [95] Polymers 2023, 15, 2875 16 of 21

Conclusions and Outlook
The present review highlights the significance of real-time cross-linking alginate in situ gelation recording. Demonstrating such advanced experimental methods provides an assessment of the reaction progress and the main factors evolved in each stage. Customized rheometric setups and microscopical and structural time-resolved techniques have been developed over the last several years to control cross-linking gelation. Therefore, both mechanical and morphological properties can be recorded during the formation of alginate hydrogels and beads. This allows researchers to observe the process as it occurs and to investigate the effects of concentration and the physicochemical properties of reactants on the gelation evolved in an initial fast stage, followed by a slower evolution. Additionally, based on real-time experimental data, kinetic models are developed to describe and adequately predict the cross-linking alginate reaction. After extensive analysis of the research studies, it is observed that the fast external gelation, in which the gel point appears in time scales of seconds, can be quantified in terms of its mechanical properties and gel front position. Therefore, with the present review, an in-depth comprehension of alginate gelation is provided for the purpose of optimizing the conditions of the cross-linking reactions.
The prospects for in situ gelation are encouraging, since it provides many advantages compared to conventional gelation procedures. The transition of the liquid reactants to gelling structures on the characterization location, e.g., the rheometric bottom plate and microscopical cell, provides precise tuning of the mechanical and morphological properties over an extended time. The present review constitutes a valuable guide for identifying the appropriate technique to study the dynamic response, not only of alginate gelation, but also of other complex systems. The advanced setups can be utilized to study in situ sensitive biological hydrogels without the presence of an increased temperature which may cause a degradation of the material. Additionally, the in situ characterization of other cross-linking polymers, such as pectin and polyurethane, can be achieved. For the abovementioned reasons, developing methods to control the mechanical and structural characteristics simultaneously is deemed imperative. This will lead to the synthesis of smart hydrogels for state-of-the-art applications.

Conflicts of Interest:
The authors declare no conflict of interest.