Hydrogel Properties and Characterization Techniques
The unique structure of hydrogels as materials, including their soft mechanical properties, their typically high water contents, their capacity to respond to changes in their solvent environment, and their tunable and often multi-scale porosity, offers both significant challenges and specific opportunities in terms of their characterization. Herein, we describe the key properties associated with hydrogels (from both qualitative and quantitative perspectives) and review the major analytical techniques used to probe those properties, highlighting the strengths and weaknesses of various available strategies. Chemical, physical, and biological properties are all reviewed, with an emphasis on the techniques developed specific to hydrogels to measure swelling, mechanics, gelation time, and porosity.
Atomic absorption spectroscopy
Dynamic light scattering
Fourier transform infrared spectroscopy
Nuclear magnetic resonance
Nanoparticle tracking analysis
Pulsed gradient spin-echo nuclear magnetic resonance
Small-angle neutron scattering
Scanning electron microscopy
Transmission electron microscopy
Tunable resistive pulse sensing
Visible light (as in UV/vis spectroscopy)
X-ray photoelectron spectroscopy
Hydrogels have several key properties that must be understood both experimentally and theoretically in order to effectively design functional hydrogel-based materials for applications. However, the unique nature of hydrogels as materials (e.g., their high water contents, soft mechanics, internal porosities, networked chemical structures, etc.) makes many of the techniques typically used to assess the properties of other polymeric materials difficult if not impossible to apply to hydrogels. For example, contact angle measurements that are widely used to assess the hydrophilicity of a material interface are ineffective with hydrogels due to the typically rapid absorption of the droplet into the gel phase. Methods used to measure the porosity of hard porous materials (e.g., gas adsorption, mercury porosimetry) are ineffective at assessing the porosity of hydrogels given that hydrogels compress upon pressurization and the water inside hydrogels cannot effectively be displaced without significantly deforming the network structure. Many common techniques used to assess morphology (e.g., electron microscopy) or surface chemistry (e.g., x-ray photoelectron spectroscopy) require operation in a vacuum, posing challenges in the context of analyzing a highly water-swollen material. Furthermore, standard chemical characterization techniques used for polymers (e.g., nuclear magnetic resonance) are challenging or impossible to apply quantitatively to hydrogels based on the crosslinked networked structure which significantly alters relaxation times or other fundamental mechanisms used for chemical analysis. As such, hydrogel characterization is frequently a significant challenge, particularly from a chemical perspective. However, at the same time, the unique properties of hydrogels have led to the development of several techniques beyond those used for conventional “hard” materials that have enabled a significantly better understanding of both the fundamental physics as well as the application performance of hydrogel-based materials.
In this chapter, we outline the key properties of hydrogels from the chemical, physical, and (briefly) biological standpoints, describe the conventional methods/approaches for measuring such properties, and outline the key theoretical relationships used to understand the fundamental influencers to those properties (the latter of which are also essential to understand in order to rigorously apply the techniques described).
2 Chemical Properties
Chemically, there is interest in analyzing both the chemistry of the building blocks incorporated into a hydrogel (particularly when two or more different building blocks are used as the basis of hydrogel formation) as well as the extent of reaction among available crosslinking groups. However, the network structure of hydrogels makes the chemical analysis of hydrogels challenging, as many of the traditional solution-based chemical characterization methods often do not give accurate or consistent results with hydrogel networks. As such, while most of the methods described below are highly imperfect, they can collectively give insight into gel composition and crosslink density.
2.1 Nuclear Magnetic Resonance (NMR)
Proton NMR is the most common method used for the chemical characterization of polymers [1, 2]. However, in the context of hydrogels, NMR is challenging due to the significantly extended proton relaxation times observed within hydrated network structures, resulting in peak broadening that effectively reduces signal-noise [1, 2, 3]. In addition, functional groups at the periphery of the hydrogel and in the bulk of the hydrogel are often exposed to functionally different chemical environments (particularly with regard to the balance between free and bound water) and local crosslink densities, resulting in asymmetric signal generation throughout the hydrogel and thus selective over-weighting of functional groups in local areas with faster relaxation times. Reducing the dimension of the hydrogel from the bulk scale to the nanoscale somewhat improves the analysis , but quantitative analysis in the manner of small molecules or linear polymers remains challenging. Solid-state NMR using magic-angle spinning has been applied to address this problem, with some success in particular for tracking changes in crosslinking during the process of gelation . However, signal-to-noise can still be a challenge with solid-state NMR, particularly in probing hydrogels with lower crosslink densities if the main purpose of the analysis is to calculate crosslink density, and experiments are slow and expensive.
2.2 Infrared Spectroscopy (IR)
IR spectroscopy is a useful tool to identify bond types in hydrogels and other polymeric systems. While the conventional potassium bromide (KBr) pellet technique for transmission-based IR can be applied to hydrogels, the difficulty inherent in grinding hydrogels (which remain somewhat wet due to bound water even when “dry”) often makes it difficult to create uniform pellets with substantial transmission . Alternately, attenuated total internal reflectance techniques (AT-FTIR) may be used to probe the surface (0.5–5 μm) of the hydrogel , with the assumption that the surface chemistry is not substantially different from the bulk chemistry. This assumption should however be made with care, as AT-FTIR results are often carelessly interpreted in the literature. Sensitivity and quantification are also challenges in using IR for characterization of hydrogels (or indeed polymers in general). Interpretation of small crosslinker signals in particular should conservatively be limited to the qualitative appearance/disappearance of peaks related to the crosslinking compound before and after reaction.
2.3 Fluorescent Probes
2.4 X-Ray Photoelectron Spectroscopy (XPS)
XPS can identify the elemental composition of hydrogels and, when used in high-resolution mode, can give insight into the types of bonds present in hydrogels based on the subtle differences in binding energy depending on the types of substituents adjacent to the bond being studied . This technique is particularly useful in cases where it is sought to prove a particular type of crosslink has formed, but IR techniques do not give the required resolution. However, as with any XPS measurement, only the surface (~10–50 nm) of the sample is probed ; furthermore, due to the need to operate under a vacuum, the hydrogel must be dried prior to analysis, potentially altering the crosslinking chemistry of particularly dynamically crosslinked hydrogel systems relative to their solution state. Atomic absorption spectroscopy (AAS) may be a better choice in some cases, as it can probe the elemental composition of the entire bulk gel and thus give a clearer picture of overall crosslinking density (particularly if the crosslinker has a distinct element versus the building blocks) and/or monomer incorporation if the comonomers have different C/O/N/H ratios . However, AAS has no capacity to give any information about the bonding between these atoms and, as a result, cannot distinguish between single (graft) or multiple (crosslink) reacted crosslinking molecules.
3 Physical Properties
For the design of hydrogels for targeted applications, three key physical measurements are typically performed: (1) hydrogel swelling, associated with thermodynamics and/or changes in the crosslink density within the gel over time; (2) hydrogel mechanics, associated with the viscoelastic properties and/or the load-bearing potential of the networks; and (3) hydrogel morphology as it pertains to the porosity of the gel network (affecting the rate of molecular diffusion in or out of the hydrogel) and/or the capacity for domain formation inside hydrogels (affecting both diffusion and the capacity of the hydrogel to adsorb chemicals in its environment).
Given the diversity of ratios or percentage changes that can be used to represent swelling, it is critical to clearly express what type of swelling ratio is being reported in any case. All of these expressions have clear uses and relevance in different applications, and all are routinely reported in the literature. Note that any of these expressions may be applied equally to equilibrium swelling responses or non-equilibrium swelling responses.
For typical swelling measurements, the zero strain state is assumed to be equivalent to the polymer volume fraction during gel preparation, given that crosslinks should form in a manner to minimize the free energy (and thus strain) in the system. However, in other cases in which swelling changes between two different equilibrium swelling states are sought, the zero strain state may be taken as the initial equilibrium polymer volume fraction. The a term can be simply defined in spherical micro/nanogel systems as the particle radius but can also be estimated for any shape of hydrogel in terms of the characteristic length and the ratio between the total volume and surface area of a gel of any shape/dimension. The mixing term Πm incorporates both enthalpic (i.e., polar interactions between the polymer and water) and entropic (i.e., entropy gains by allowing lattice sites to be occupied by polymer or solvent) terms and may contribute to either swelling or deswelling depending on the value of the χ parameter, which is directly related to the polymer solubility parameters of the gel building blocks. Similarly, the elastic term Πe may promote swelling if the other terms promote gel contraction (reducing entropy versus the zero strain state via chain compression/coiling) or deswelling if the other terms promote gel expansion (reducing entropy versus the zero strain state in terms of chain stretching). The Donnan term Πd can only drive swelling due to the increased water content required to maintain high entropy of counterions to fixed charges on the gel network (if any).
More complex models can incorporate other effects on gel swelling, including chain stiffness (suppressing swelling/deswelling) , co-ion effects and background electrolyte (reducing Donnan equilibrium-based swelling) , direct repulsion between polymer-bound charges (increasing charge-driven swelling) , and counterion condensation (decreasing charge-driven swelling, particularly at higher overall ionic strengths) . These more complex models have been applied to both bulk gels  and particulate gels with well-defined charge/crosslinker distributions . However, unless hydrogels are made with liquid crystal-like building blocks (chain stiffness effects), very high charge densities (direct repulsion), or measured at high salt concentrations (counterion condensation/background electrolyte effects), these additional contributions are typically not necessary to consider to generate reasonable swelling estimates.
3.1.2 Experimental Measurements
Swelling in bulk hydrogels can be measured in multiple ways. The most common, and simplest, method is to track the change in the dimension(s) of the hydrogel upon exposure to a new solvent, allowing for the calculation of the volume-based swelling ratio. Measurements are made simpler when the hydrogel is fabricated in a container (e.g., a glass capillary)  that restricts gel swelling in only one dimension, increasing the accuracy of the volume swelling measurement at the cost of slowing down the response time of the gel due to the lower interfacial exposure of the gel to the new solvent conditions. Note that this approach inherently assumes the hydrogel is isotropic (i.e., swelling occurs equally in all dimensions if not otherwise constricted). Alternately, gravimetry is also widely used to estimate mass-based swelling ratios by weighing the hydrogel before and after exposure to the new solvent conditions [17, 18]. The hydrogel is most commonly contained in a wire mesh , a dialysis bag , or a cell culture insert  to enable facile handling of the gel between weighing steps without the need for tweezers that may break or damage the hydrogel; the pre-weighed container mass is subtracted from the overall mass at each time point. Gentle wicking of surface (nonabsorbed) water is typically performed using Kimwipes prior to measurement to ensure all water mass measured is adsorbed or absorbed within the hydrogel. Less commonly, the swelling pressure can be measured by loading a hydrogel on a mechanical test frame and tracking the change in measured normal force as a function of swelling . Quantitative measurements of swelling using this method require the confinement of the gel to direct all swelling in the normal direction, as gel swelling in the lateral direction is not directly assessed using z-direction force measurements. As such, swelling pressure measurements are not routine and are typically used only in cases where the forces exerted by a swelling hydrogel are relevant to the target application.
Microscale gel swelling can be measured using laser diffraction instruments that can probe particle sizes on the microscale . Alternately, optical microscopy can be used to track size changes visually, with particle size changes assessed using image analysis algorithms or software . In this latter case, care must be taken to ensure that an adequate number of representative particles (typically n > 100) is assessed to achieve statistically relevant particle size results.
The estimates provided by simple rubber elasticity theory are only estimates, as any nonhomogeneity in the network can significantly alter the results. More complex models have been derived for some of these inhomogeneous network cases, although still assuming non-Gaussian statistical distributions of crosslinks and accounting for entanglement effects .
3.2.2 Experimental Measurements
While all the above strategies effectively probe the average moduli of a bulk hydrogel, an interesting alternative approach called cavitation rheometry has also been developed to measure local moduli within bulk hydrogels . In this approach, a capillary or needle is inserted locally within a hydrogel sample and air is injected at a controlled rate. The corresponding pressure as a function of the air volume is tracked until the critical pressure of mechanical instability is reached, a pressure that can subsequently be correlated to the modulus of the material. This technique uniquely allows for multipoint modulus measurements within the hydrogel and can be performed using inexpensive instrumentation.
For microscale hydrogels, cantilever-based systems coupled with a microscope are available for performing compressive testing of single gel microspheres, with a capacity to perform measurements on microparticle-based hydrogels as small as 20–30 μm in diameter . Micropipette aspiration measurements have also been developed in which the gel microparticle is aspirated into a capillary tube and the modulus is calculated based on either the percentage of the material sucked into the tube at a particular applied negative pressure and/or the angle of deformation at the tip-sphere interface . Multisphere indentations coupled with Johnson-Kendall-Roberts (JKR) soft adhesion theory  and ultrasonic pulse-echo methods  have been reported as alternative strategies for gel microparticle modulus measurements.
For nanoscale hydrogels, atomic force microscopy (AFM) is the only viable technique for measuring a modulus via an indentation/compression mechanism . Since this technique practically requires the adsorption or at least tethering of the nanogels to a substrate, care must be taken to ensure minimization of the known “pancaking” of nanogels upon interaction with surfaces that would complicate the conversion of the measured forces to moduli values.
3.3 Gelation Time
Here, pgel is the fractional monomer conversion at the gel point π and Ni is the number of monomers with a functionality of fi (note that at least one monomer must have an fi greater than two in order to induce gelation). An analogous approach could be used to predict gelation conversions of pre-functionalized polymers that crosslink via a step-growth approach, with the average number of functional groups per precursor polymer chain used as the functionality of the “monomer” units in Eq. 12. In this context, the kinetics of the functional group reaction driving crosslink formation determine the rate of gelation. Note that an analogous gelation point calculation for chain-growth hydrogels is not physically meaningful since crosslinks are formed simultaneous to chain formation depending on the statistical incorporation of the crosslinking or backbone monomer(s) during the polymerization process, as determined by the copolymerization kinetics of the monomer(s) relative to the crosslinker(s).
3.3.2 Experimental Measurement
The gelation time can be experimentally determined using a vial inversion test, in which a vial containing the pre-gel components is rotated at pre-determined time intervals until no flow is observed during the defined measurement time . Alternately, shear rheology experiments can be conducted to track G′ and G″ as a function of time, with the time at which G’ = G″ (or G′/G″ = 1, conventionally defined as the gel point) corresponding to the gelation time of the hydrogel [40, 41].
3.4 Porosity/Network Structure
However, such an approach ignores the substantial heterogeneities that are present in most hydrogel samples and are essential to understand, demanding the development of experimental techniques to gain an improved understanding over internal network structure.
3.4.2 Experimental Measurements
188.8.131.52 Electron Microscopy
Scanning electron microscopy (SEM) is the most common method used in the literature to assess hydrogel morphology, although transmission electron microscopy (TEM) is also frequently used to probe nanogel structures [43, 44, 45]. The obvious limitation of the technique in the context of hydrogel analysis is that it must be conducted under vacuum and thus requires removal of the water phase from the hydrogel prior to imaging. This is typically done either via a lyophilization process (i.e., quick-freezing the sample in liquid nitrogen or dry ice and then sublimating the water under reduced pressure)  or via a solvent exchange process followed by critical point drying (i.e., stepwise replacement of water with an organic solvent prior to drying) [47, 48]. Both these methods are inherently problematic. During lyophilization, free water inside hydrogels can form organized ice crystal structures upon quick-freezing that deform the gel and, following removal of the ice via lyophilization, leave behind macropores that are then imaged. Indeed, it is these micro/macropores resulting from ice crystal growth that are most commonly observed in SEM hydrogel images in the literature and are often highly misleading relative to the actual swollen state of an unperturbed gel, although it should be acknowledged that a correlation does exist between the size of the ice crystallites and the modulus of the hydrogel (i.e., stiffer gels result in smaller ice crystallites forming) . The size of such ice crystals can be suppressed by improving control over the freezing process [50, 51]. Solvent exchange/critical point drying avoids these problems but introduces other challenges in terms of correlating the swelling state of the hydrogel in the exchanged organic solvent relative to that observed in water. Performing smaller gradient solvent steps helps to preserve the aqueous structure of the hydrogel  but is not a complete solution. Cryo-EM approaches can offer somewhat improved images by avoiding the need to remove solvent, as imaging is performed on a substrate maintained at low temperature typically using liquid ethane . However, crystallite formation upon freezing can still deform the native microstructure to some degree, particularly in weaker hydrogels, and the resolution (absent the use of reconstruction algorithms) can still be limited. As such, while SEM/TEM has some degree of utility in terms of analyzing hydrogel morphology, results only from SEM/TEM must be interpreted judiciously in light of the sample preparation techniques required.
184.108.40.206 Macromolecular Probes
The uptake (or exclusion) of molecular or nanoparticle-based probes with well-defined hydrodynamic diameters into hydrogels can be used as an indirect strategy to characterize the pore size distribution. Analogous to the molecular weight cutoff of a membrane, macromolecules or nanoparticles with diameters larger than the maximum pore size of the hydrogel will be fully excluded from the gel phase, while probes of sizes smaller than the maximum pore size will be able to penetrate into the hydrogel to a degree proportional to the percentage of hydrogel pores equal to or larger than the size of the probe used. The most common probes used include fluorescein-labeled dextrans, which are commercially available in a range of molecular weights from ~4 kDa up to 2 MDa  (albeit with somewhat different polydispersities which must be considered in interpreting the data). Proteins are also highly useful as probes given their very well-defined geometries and molecular weights, eliminating any complications associated with the polydispersity of polymer probes. For example, using a set of fluorescently labeled proteins including bovine catalase (~240 kDa), glyceraldehyde 3-phosphate dehydrogenase (~146 kDa), avidin (68 kDa), trypsinogen (24 kDa), and myoglobin (17 kDa) allows for probing of pore sizes from the ~2 to 10 nm size range most typical of hydrogels, as successfully demonstrated on both bulk hydrogels and gel micro/nanoparticles [55, 56]. DNA can also be used as a probe given the precise control possible over the number of base units, often coupled with gel electrophoresis to accelerate the measurement . Well-defined nanoparticles or microparticles are also useful as probes for analyzing larger pore sizes in hydrogels, as previously demonstrated for natural mucin . It should be noted that this technique inherently assumes that the probe(s) can freely move through the gel micro/nanostructure without significant entrapment or adsorption. Given the noted capacity of proteins in particular to adsorb to interfaces, care should be taken to first understand any non-specific interactions that may occur to a substantial enough degree to interfere with the pore size measurements before quantitative pore size distributions are estimated from the data.
220.127.116.11 NMR Relaxation
18.104.22.168 Small-Angle Neutron Scattering (SANS)
In this model, the two key hydrogel parameters extracted are the correlation length ε between the static inhomogeneities and the characteristic size of the inhomogeneities Ξ, parameters that give insight into both the size (Ξ) and spacing (ε) of inhomogeneous features in hydrogels (note that both numerators are intensity-related weighting constants to facilitate fitting of the function to real scattering curves). Other functions including Porod, Lorentzian, and Guinier (some of which can be used to extract a correlation length that approximates the mesh size of the hydrogel as defined in Eq. 13) can also be used , with multiple static inhomogeneity terms added in some cases to account for the multiple types of inhomogeneities present in some gels .
Contrast matching experiments are also possible in which one of the hydrogel components is deuterated. By adjusting the ratio between D2O and H2O in the solvent used to swell the hydrogels, the scattering between the solvent and protonated component(s) and the deuterated component(s) can be manipulated to effectively suppress signal from one or the other component by matching the scattering length densities of the solvent with that of the polymer phase desired to be hidden . This is analogous to observations with light, as zero net scattering is observed when a polymer phase and the solvent have equal refractive indices. The scattering length density is a direct function of the elemental composition of the hydrogel components, such that if the composition of the hydrogel is known, the H2O:D2O solvent ratio facilitating index matching can be reasonably calculated. Such analysis allows for investigation of the internal distribution of one gel component without interference from other components of the gel, another unique feature of SANS analysis of hydrogels. However, it should be noted that the information gained from SANS is highly model-dependent and requires both choosing the right physical model for the system and performing accurate fits to that model. In addition, highly specialized instrumentation is required to perform the experiments, requiring a neutron source with a sufficient flux and appropriate wavelength to probe the desired length scales of hydrogel features.
The transparency of hydrogels is critical for the use of hydrogels for optical applications. Transparency can be simply measured using UV/vis absorbance/transmittance measurements, with the threshold for transparency depending on the thickness of the hydrogel used and the needs of the specific application . Refractive index measurements can also be performed, although the high water content of most hydrogels results in refractive index readings rarely far from those of water ; this is true even in hydrogels that are visually opaque due to the formation of internal scattering domains.
4 Biological Properties
While a comprehensive coverage of biological assays associated with the use of hydrogels as biomaterials is outside the scope of this chapter, the reader is referred to the excellent recent review on the use of hydrogels for cell culture applications from Burdick’s group for guidance on best practices for culturing cells on (2D) or inside (3D) hydrogels and assessing the capacity of those cells to remain viable, proliferate, grow, migrate, or differentiate . Protein adsorption to hydrogels is also of interest to measure, as the high hydrophilicity of hydrogels is often useful for suppressing non-specific protein adsorption and, subsequently, the inflammatory response following implantation in vivo [74, 75, 76]. Fluorescent  or radioactive probe  labeling of model proteins or of complex protein mixtures (e.g., whole blood or blood plasma) is typically used to enable detection of the fraction of added protein that binds to a hydrogel. Interpretation of the results of a protein adsorption experiment is however sometimes challenging with hydrogels given that proteins may both adsorb (as with conventional hard biomaterials) and absorb (due to the internal porosity) in hydrogels, depending on the size of the proteins relative to the hydrogel pore size; indeed, proteins are used as probes of gel porosity due to this capacity for non-specific absorption into the gel phase. As such, a hydrogel with smaller pores may be noted to exhibit less overall protein binding but actually interfacially adsorb more protein, likely to ultimately result in a more severe inflammatory reaction. Confocal imaging of fluorescently labeled protein distributions across a hydrogel cross-section can assist in differentiating these contributions and improve the prediction of the potential compatibility of a hydrogel in a biological application .
The characterization of hydrogels is inherently more challenging than with many other materials given the dynamic nature of hydrogels as their environment (solvent type, humidity, etc.) changes. Indeed, many of the core techniques used to characterize in particular the chemistry of “hard” polymer materials (e.g., XPS, NMR) are at best challenging to apply to quantitative hydrogel characterization. There are significant needs (and thus opportunities) to develop improved techniques for (a) performing traditional characterization techniques accurately on hydrogels (e.g., developing new NMR pulse sequences) and/or (b) developing strategies of drying hydrogels (e.g., critical point drying) that minimally perturb the swollen structure of the hydrogel in water. However, significant recent advancements have been made in the context of accurately measuring the mechanics of hydrogels under various stress regimes using a variety of creative techniques. The continued development of such techniques is expected to yield further insight into the fundamental properties and physics of hydrogels, enabling improved engineering of hydrogel structures to achieve targeted hydrogel performance.
- 12.P.J. Flory, Principles of Polymer Chemistry (Cornell University Press, Ithaca, 1953)Google Scholar
- 27.O.S. Lawal, J. Storz, H. Storz, D. Lohmann, D. Lechner, W.M. Kulicke, Hydrogels based on carboxymethyl cassava starch cross-linked with di- or polyfunctional carboxylic acids: Synthesis, water absorbent behavior and rheological characterizations. Eur. Polym. J. 45(12), 3399–3408 (2009)CrossRefGoogle Scholar
- 35.T. Boudou, J. Ohayon, Y. Arntz, G. Finet, C. Picart, P. Tracqui, An extended modeling of the micropipette aspiration experiment for the characterization of the Young’s modulus and Poisson’s ratio of adherent thin biological samples: numerical and experimental studies. J. Biomech. 39(9), 1677–1685 (2006)CrossRefGoogle Scholar
- 37.A.V. Salsac, L. Zhang, J.M. Gherbezza, Measurement of mechanical properties of alginate beads using ultrasound. 19eme Congres Francais de Mecanique, 2009Google Scholar
- 39.K.J. De France, K.J.W. Chan, E.D. Cranston, T. Hoare, Enhanced mechanical properties in cellulose nanocrystal-poly(oligoethylene glycol methacrylate) injectable nanocomposite hydrogels through control of physical and chemical cross-linking. Biomacromolecules 17(2), 649–660 (2016)CrossRefGoogle Scholar
- 79.D. Luensmann, M.A. Glasier, F. Zhang, V. Bantseev, T. Simpson, L. Jones, Confocal microscopy and albumin penetration into contact lenses. Invest. Ophthalmol. Vis. Sci. 84(9), 839–847 (2007)Google Scholar