WO2025029196A1

WO2025029196A1 - Hydrogel ink and methods for direct 3d printing of hydrogel structures

Info

Publication number: WO2025029196A1
Application number: PCT/SG2024/050488
Authority: WO
Inventors: Changjin HUANG; Quan Zhou
Original assignee: Nanyang Technological University
Current assignee: Nanyang Technological University
Priority date: 2023-08-01
Filing date: 2024-08-01
Publication date: 2025-02-06
Anticipated expiration: 2026-02-01

Abstract

Disclosed herein is a method of forming a three-dimensional hydrogel structure, the method comprising (a) providing an ink formulation comprising water, a polymeric material and an organic solvent to an additive manufacturing device, and (b) using the additive manufacturing device to print the ink formulation onto a substrate to generate a three- dimensional structure, wherein in step (b) either the substrate is held at a temperature that is below a sol-to-gel transition temperature of the ink formulation, or in step (b) the printing step is conducted in an environment that has a temperature below the sol-to-gel transition temperature of the ink formulation, so as to enable the printed ink to undergo a sol-to-gel transition upon printing. Also disclosed herein is a method of forming hollow 1 D, 2D and 3D structures by immersing the solid 1 D, 2D or 3D structure into a gelation bath, where the gelation bath comprises water and a suitable crosslinking agent.

Description

HYDROGEL INK AND METHODS FOR DIRECT 3D PRINTING OF HYDROGEL STRUCTURES

Field of Invention

The present invention generally relates to three-dimensional printing, and more particularly relates to three-dimensional printing of hydrogels.

Background

The listing or discussion of a prior-published document in this specification should not necessarily be taken as an acknowledgement that the document is part of the state of the art or is common general knowledge.

Hydrogels are three-dimensional cross-linked polymeric networks capable of absorbing a large amount of water. Since hydrogels were first reported in 1960, their applications have been greatly expanded into various fields, such as drug delivery in biomedicine, wound dressings in tissue engineering, flexible electronics and soft robotics in engineering. Continuous advancement in hydrogel-based applications relies not only on novel hydrogel materials with superior properties but also on effective strategies that can facilitate easy generation of complex hydrogel structures, as those applications are often enabled by their special structural designs, such as the implantable microelectromechanical system developed by Yin Chin et al. (S. Y. Chin et al., Sci. Robot. 2017, 2 (2)), the soft actuators reported by Mishra et al. (A. K. Mishra et al., Sci. Robot. 2020, 5 (38)), and the hydrogel robot developed by Zhang et al. (J. Zhang et al., Cell Rep. Phys. Sci. 2022, 3, 101081). However, due to their soft and wet nature, the well-established techniques developed for shaping traditional engineering materials (e.g., metal, plastic, and elastomers), such as machining and lithography, are no longer applicable. Mold casting has become the most common approach to making hydrogel structures. Nevertheless, making complex structures with mold casting remains challenging as it is difficult to get a soft hydrogel out of a complex mold.

Hollow structures, including water pipes and blood vessels, are common in industry and nature. Moreover, hollow hydrogels are also important both in research and clinics. For instance, it was reported that hollow hydrogel tubes could be used to mimic complicated cellular microenvironments in living organisms and display potential for vascular grafts. However, due to hydrogels' high water content and soft nature, their manufacturing and processing may be limited, especially for hollow-structured hydrogels. Currently, some methods could be used to obtain hollow hydrogel structures, including coaxial nozzle extrusion, templating, casting, assembly and partial curing. For coaxial nozzle extrusion, templating, casting and assembly approaches, some relatively simple hollow structures can be obtained, such as linear tubes and simple branched constructs. However, these methods can only be used for the production of those relatively simple structures due to the limitations of the fabrication processes. Some relatively complex structures could be obtained through partial curing. However, patent-based manufacturing devices are complex and costly. Therefore, developing a new method to produce hollow hydrogel structures that combines simple manufacturing processes and the complexity of obtained constructs is important.

Three-dimensional (3D) printing, especially extrusion-based, has recently emerged as a powerful technique to generate complex hydrogel structures. In order to print directly into desired 3D structures, hydrogel inks are required to have a proper viscosity, good stackability, and shear-thinning behavior to be rapidly solidified after deposition. However, most hydrogel precursor solutions, especially at low concentrations, do not meet these requirements and therefore, special treatments are often needed to promote their printability. Yield stress fluids, which are shear-thinning materials that can flow like liquids only beyond critical stress, have been commonly adopted as 3D printing inks. It is required that the yield stress of the inks should be sufficiently high to prevent the action of capillary forces, which can be achieved by mixing multi-component polymers together, increasing polymer concentration, or adding particle reinforcements. However, their high viscosity poses a challenge to the printing process as a high extrusion pressure is required and the nozzle tends to get clogged easily, especially for fine nozzles.

In particular, alginates are a kind of linear anionic polysaccharides derived from brown algae and bacterial biosynthesis with blocks of (1 ,4)-linked a-L-guluronic acid (G) and |3-D- mannuronic acid (M) residues, and they can form a physical hydrogel by lowering pH value or introducing di- or tri-valent cations (e.g., Ca²⁺ and Fe³⁺). Due to their low toxicity, high biocompatibility and relatively low cost, alginates have been widely investigated and applied in many biomedical applications. However, similar to many other hydrogel materials, direct 3D printing of alginate hydrogel is generally challenging due to the low viscosity of alginate solutions. Therefore, alginate is usually mixed with other ingredients, including polymers (e.g., gelatin, collagen, poly(vinyl alcohol), methylcellulose, and polyethylene oxide) and nanoparticles, to increase the viscosity of the printing ink. However, the indiscriminate elevation of ink viscosity may paradoxically result in printing difficulties, such as unsmooth extrusion and even nozzle clogging. Alternatively, 3D printing of alginate hydrogel can be achieved by printing alginate solution into a supporting bath that contains viscous polymer solutions to help maintain the printed hydrogel patterns, such as a gelatin solution with calcium ions. However, the presence of the supporting polymers and the instantaneous gelation of the outmost layer of printed alginate may reduce the bonding strength between two sequentially printed layers, resulting in a weak mechanical strength. A special supporting stage whose Z- axis motion can be precisely controlled has been adopted to enable direct printing of alginate precursor (A. G. Tabriz et al., Biofabrication 2015, 7, 45012; and K. Christensen et al., Biotechnol. Bioeng. 2015, 112, 1047-1055). The stage was initially at the interface of calcium solution and air. As a layer of alginate solution was printed, the Z-axis of the stage was lowered down to sub- merge the printed layer into the calcium solution bath to crosslink the printed layer. However, when using this method to print tall constructs, the printed constructs are prone to collapse, resulting in a low printing quality.

Therefore, to overcome at least one of the aforementioned problems, there exists a need for new methods of forming three-dimensional hydrogel structures.

Summary of Invention

Aspects and embodiments of the invention will now be discussed by reference to the following numbered clauses.

1. A method of forming a three-dimensional hydrogel structure, the method comprising:

(a) providing an ink formulation comprising water, a polymeric material and an organic solvent to an additive manufacturing device; and

(b) using the additive manufacturing device to print the ink formulation onto a substrate to generate a three-dimensional structure, wherein: the polymeric material is selected from one or more of the group selected from a polysaccharide, a protein and a synthetic polymeric material that has functional groups that can generate hydrogen bonds; the organic solvent is selected from one or more of the group consisting of an alcohol, acetone, dimethylformamide and dimethylsulfoxide; and in step (b) either the substrate is held at a temperature that is below a sol-to-gel transition temperature of the ink formulation, or in step (b) the printing step is conducted in an environment that has a temperature below the sol-to-gel transition temperature of the ink formulation, so as to enable the printed ink to undergo a sol-to-gel transition upon printing. 2. The method according to Clause 1, wherein the organic solvent is selected from one or more of the group consisting of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol.

3. The method according to Clause 1 or Clause 2, wherein the polymeric material is selected from one or more of the group consisting of sodium alginate, K-carrageenan, chitosan, chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin, pullulan, poly(vinyl alcohol) (PVA), hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP), polyurethane (PU), gelatin, albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein and zein, optionally wherein the polymeric material is sodium alginate.

4. The method according to any one of the preceding clauses, wherein when: the polymeric material is sodium alginate, then the organic solvent is selected from one or more of the group of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol; the polymeric material is K-carrageenan, chitosan, chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin or pullulan, then the organic solvent is selected from one or more of the group of ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol; the polymeric material is poly(vinyl alcohol) (PVA), hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP) or polyurethane (PU), then the organic solvent is acetone; and the polymeric material is gelatin, albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein or zein, then the organic solvent is selected from one or more of the group of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol.

5. The method according to any one of the preceding clauses, wherein the ink formulation is one that has: from 2 to 10% w/v of the polymeric material; from 25 to 40% v/v of the organic solvent; and from 60 to 75% v/v of water. 6. The method according to Clause 5, wherein the polymeric material is present in an amount of from 2 to 4% w/v of the ink formulation, such as about 3.3% w/v.

7. The method according to Clause 5 or Clause 6, wherein the organic solvent is present in an amount of from 30 to 35% v/v of the ink formulation, such as about 33.3% v/v.

8. The method according to any one of the preceding clauses, wherein the temperature of the substrate or environment in step (b) is less than or equal to 15 °C and greater than or equal to -25 °C, such as less than or equal to 5 °C and greater than or equal to -20 °C (e.g. greater than or equal to -10 °C).

9. The method according to any one of the preceding clauses, wherein: the ink formulation contains sodium alginate at a concentration of 3.3% w/v, ethanol at a concentration of 33.3% v/v and water in an amount of 66.7% v/v; and the temperature of the substrate or environment in step (b) is less than or equal to 5 °C.

10. The method according to any one of the preceding clauses, wherein the method further comprises:

(c) immersing the three-dimensional structure generated in step (b) in a gelation bath, where the gelation bath comprises water and a suitable crosslinking agent for the polymeric material, so as to provide a crosslinked structure, optionally wherein the gelation bath further comprises the organic solvent, further optionally wherein the concentration of the organic solvent in the gelation bath is about the same as the concentration of the organic solvent in the ink formulation.

11. The method according to Clause 10, wherein:

(i) the temperature of the gelation bath is elevated above the boiling point of the organic solvent, so as to provide a hollow three-dimensional structure; or

(ii) the crosslinked structure is placed into a further bath comprising water at a temperature above the boiling point of the organic solvent, so as to provide a hollow three- dimensional structure, optionally wherein the further bath further comprises the organic solvent, further optionally wherein the concentration of the organic solvent in the further bath is about the same as the concentration of the organic solvent in the ink formulation.

12. The method according to any one of the preceding clauses, wherein the method further comprises: (d) immersing the three-dimensional structure generated in step (b) or step (c) in water to remove the organic solvent.

13. The method according to any one of Clauses 10, 11 and 12, the latter as dependent upon Clause 10 or Clause 11, wherein the polymeric material is sodium alginate and the crosslinking agent is a calcium salt (e.g. CaCh), optionally wherein the concentration of the calcium salt is from 0.05 to 0.2 M, such as about 0.1 M in the gelation bath, further optionally wherein the organic solvent is ethanol.

14. A method of forming hollow 1D, 2D and 3D structures, the method comprising the steps of:

(ai) providing a solid 1 D, 2D or 3D structure in an uncured state, the solid 1D, 2D or 3D structure comprising a polymeric material, an organic solvent and water; and either:

(aii) immersing the solid 1 D, 2D or 3D structure into a gelation bath, where the gelation bath comprises water and a suitable crosslinking agent for the polymeric material, where the temperature of the gelation bath is elevated above the boiling point of the organic solvent, so as to provide a cured, hollow 1 D, 2D or 3D structure; or

(aiii) immersing the solid 1 D, 2D or 3D structure into a gelation bath, where the gelation bath comprises water and a suitable crosslinking agent for the polymeric material to provide a solid, cured 1 D, 2D or 3D structure, followed by placing the solid, cured 1 D, 2D or 3D structure into a further bath comprising water at a temperature above the boiling point of the organic solvent, so as to provide a cured hollow 1 D, 2D or 3D structure, optionally wherein in steps (aii) and (aiii), the gelation bath and the further bath comprise the organic solvent, further optionally wherein the concentration of the organic solvent in the gelation bath and the further bath is about the same as the concentration of the organic solvent in the solid 1 D, 2D or 3D structure in an uncured state.

15. The method according to Clause 14, wherein the polymeric material is sodium alginate and the crosslinking agent is a calcium salt (e.g. CaCh), optionally wherein the concentration of the calcium salt is from 0.05 to 0.2 M, such as about 0.1 M in the gelation bath, further optionally wherein the organic solvent is ethanol.

Drawings Fig. 1 depicts ethanol-induced gelation of sodium alginate (SA) hydrogel. (A) Representative digital photographs of SA-ethanol mixtures with different ethanol concentrations at 25 °C and 4 °C. The corresponding ethanol concentrations (v/v) are labeled. (B) The change in storage modulus G’ and loss modulus G" of SA-ethanol mixtures with different ethanol concentrations (0%, 25%, 33.3% and 40%) during a cooling process from 25 °C to -5 °C. (C) The storage modulus G’ of SA-ethanol mixtures at 25 °C and -5 °C as a function of ethanol concentration. (D) Gelation temperature of SA-ethanol mixtures as a function of ethanol concentration. The concentration of SA polymer was 3.3% for all the samples used in (B)-(D). (E) The change in G’ and G” of SA-ethanol mixtures with different SA concentrations (2%, 3.33% and 4.5%) during a cooling process from 25 °C to -5 °C. (F) The storage modulus G’ of SA-ethanol mixtures at 25 °C and -5 °C as a function of SA concentration. (G) Gelation temperature of SA-ethanol mixtures as a function of SA concentration. (H-l) Representative scanning electron microscopy (SEM) images of freeze-dried SA-ethanol mixtures with different ethanol concentrations.

Fig. 2 depicts the relationship between the light transmittance and ethanol concentration of SA-ethanol mixtures at a SA concentration of 3.33% w/v under light wavelength of 750 nm. The sample thickness was 10 mm.

Fig. 3 depicts rheological temperature ramp tests (cooling from 25 °C to -5 °C and then heating from -5 °C to 40 °C) of SA-ethanol mixtures with various ethanol concentrations (0%, 25%, 33.3%, and 40% v/v).

Fig. 4 depicts rheological temperature ramp tests (cooling from 25 °C to -5 °C) of SA-ethanol mixtures with various ethanol concentrations (10%, 15% and 20% v/v).

Fig. 5 depicts SEM images of freeze-dried SA-ethanol mixture with 33.3% v/v ethanol (A) and SA gelled by Ca²⁺ (B). The SA concentration was 3.3% in both samples.

Fig. 6 depicts structural conformation characterization using small-angle X-ray scattering (SAXS). (A)-(C) SAXS profiles of the SA-ethanol mixture with 33.3% v/v ethanol at 25 °C, 15 °C, 5 °C, -5 °C and -10 °C (A) and the corresponding Kratky plots (B) and normalized Kratky plots (C). (D)-(F) SAXS profiles of SA-ethanol mixtures with 0%, 15%, 25%, 33.3% and 40% v/v ethanol at -10 °C (D) and the corresponding Kratky plots (E) and normalized Kratky plots (F). (G) The P(r) function of SA-ethanol mixtures with 0%, 15%, 25%, 33.3% and 40% v/v ethanol at -10 °C. (H) Fitting curves for SA-ethanol mixtures with 15%, 25%, 33.3% and 40% (v/v) ethanol at -10 °C. (I) The predicted hierarchical microstructure of the SA-ethanol gel. Dmax, R_g, and ro are the maximum size of the polymer-rich area, the radius of gyration of the polymer-rich area, the correlation length which represents the size of the cluster, and the size of the minimum unit of the internal polymer clusters, respectively.

Fig. 7 depicts molecular dynamics (MD) simulations of alginate chain assembly in the presence of Na⁺, Ca²⁺, and ethanol. (A) Atomistic models of poly-G (G16) and poly-M (M16) chains in stick representation after 20 ns of simulation. (B) Number of molecules in the largest cluster during simulation of five GIB chains in water in the presence of either 200 mM Na⁺ ions, 100 mM Ca²⁺ ions, or 40 % v/v ethanol with 200 mM Na⁺ ions. (C) Representative simulation snapshot showing clustering of Gie chains in the presence of Ca²⁺ ions with a zoomed-in view of the region showing inter-chain interaction via coordination of Ca²⁺ ions (large individual spheres) by carboxylate groups. (D) Representative simulation snapshot showing clustering of Gie chains in the presence of ethanol and Na⁺ ions with a zoomed-in view showing interchain H-bond interactions directly (right oval) or via bridging water molecules (left oval). Water molecules within 2 A of GIB chains are shown as sticks and ethanol molecules within 6 A of GIB chains are labelled with (i). (E) Similar plot as (B) for five M-I₈ chains. (F), (G) Representative simulation snapshots showing clustering of Mw chains in the presence of Ca²⁺ ions (large individual spheres) (F), or the presence of ethanol and Na⁺ ions (G), similar to (C), (D) for GIB chains.

Fig. 8 depicts simulation snapshots of the simulation system with five GIB chains in the presence of Ca²⁺ ions showing parallel and perpendicular arrangements. The zoomed-in view at 180 ns snapshot shows parallel arrangement of three chains in both van der Waals and stick representations with Ca²⁺ ions (large individual spheres) in between the chains near carboxylate groups. The zoomed-in view at 200 ns snapshot showing perpendicular arrangement of two chains in both van der Waals and stick representations with a Ca²⁺ ion (large individual sphere) between carboxylate groups from the chains. The chains are also directly hydrogen-bonded when they are within distance for hydrogen-bonding (0.3 nm).

Fig. 9 depicts example of inter-chain distances between two “clustered” alginate chains in the presence of 40% v/v ethanol. (A) Simulation snapshot of the two GIB chains shown in Fig. 7D inset using ball-and-stick representation and sticks for water molecules bridging across the chains (non-bridging waters are omitted for clarity). The short dashed lines are hydrogen bonds shown by VMD using default settings. The long dashed lines are inter-chain distances between linkage oxygen atoms from the two chains. The average value over three measurements is 9.29 A, which happens to match the value for 2r₀ or 9.30 A observed at 40% v/v ethanol (Table 1). (B) Simulation snapshot of the two Mie chains shown in Fig. 7G inset using ball-and-stick representation. The average value over two measurements was 11.9 A and larger than 2r₀.

Fig. 10 depicts rheological property characterization for pure SA solution and SA-ethanol mixtures at various printing stages. (A) The rheological properties of pure SA solution and SA- ethanol mixtures. Left: storage moduli (G’) and loss moduli (G”) of SA-ethanol mixture and pure SA solution as a function of the oscillation strain at 4 °C and 25 °C; Right: the viscosity of pure SA solution and SA-ethanol mixture as a function of shear rate at 4 °C and 25 °C. Both pure SA solution and SA-ethanol mixture contain 3.33% SA, and the SA-ethanol mixture contains 33.3% v/v ethanol. (B) Storage moduli (G’) of pure SA solution and SA-ethanol mixture under alternating high (500%, gray areas) and low (0.5 %) strain at 4 °C and 25 °C. (C) Dynamic yield stresses of pure SA solution and SA-ethanol mixture at 4 °C and 25 °C.

Fig. 11 depicts direct 3D printing of pure SA solution and SA-ethanol mixtures. (A) Digital photos of 3D printing processes using pure SA solution and SA-ethanol mixture with 33.3% v/v ethanol. (B) Digital photos of 3D printed SA-ethanol scaffold (right after printing, after gelation by Ca²⁺ ions and after soaking in deionized (DI) water for 24 h). Scale bar: 1 cm. (C) Fourier-transform infrared spectroscopy (FTI )-attenuated total reflectance (ATR) spectra of SA solution (SA), SA gelled by Ca²⁺ ions (SA-Ca), 33.3% v/v ethanol solution (Ethanol 33.3%), SA-ethanol solution (SA-ethanol), SA-ethanol gelled by Ca²⁺ ions (SA-Ca-ethanol) and SA- Ca-ethanol gel after removing ethanol (SA-Ca-deEthanol). (D) The volume reduction ratio of printed scaffolds after gelation by Ca²⁺ ions and soaking in DI water for various hours compared to the scaffolds after printing.

Fig. 12 depicts the mechanism and geometry of hollow gel tubes. (A) The regulation of the hydrogel structure by the competition between hydrogel gelation (external) and solvent gasification processes (internal). (B) Representative optical images of alginate gel fibers or tubs generated at various heating temperatures. The scale bars are all 200 pm.

Fig. 13 depicts the collapse and recovery of hollow structures.

Fig. 14 depicts the relationships between wall thickness and external (concentration of Ca ions) or internal (concentration of ethanol) factors. (A) Representative optical microscopy images of SA-EtOH tubes with the same ethanol concentration, gelled under various calcium ion concentrations (0.05, 0.1 and 0.2 M, respectively). (B-C) Outer and inner diameters (B) and wall thicknesses (C) of SA-EtOH tubes gelled under various calcium concentrations. (D) Representative optical microscopy images of SA-EtOH tubes with various ethanol concentrations (25%, 33.3% and 40% v/v, respectively) gelled under the same condition. The scale bar is 200 pm. (E-F) Outer and inner diameters (E) and wall thicknesses (F) of SA-EtOH tubes with various ethanol concentrations gelled under the same condition.

Fig. 15 depicts the relationship between gelation thickness and time of SA-EtOH samples gelled in CaCh aqueous or CaCk ethanol solution.

Fig. 16 depicts the relationship between wall thickness and external (gelation) or internal (heating caused gasification) factors by a stepwise method.

Fig. 17 depicts 2D interconnected hollow hydrogel constructs and photos of casted hollow tree-like hydrogel constructs. (A) Schematic diagram of preparing 2D interconnected hollow gel objects. (B) Designed structures of hollow hydrogels and screenshots from videos of injecting dye solution into hollow hydrogel constructs.

Fig. 18 depicts sol-to-gel transition of SA-EtOH at low temperature. (A) Rheological temperature ramp test (cooling from 37 °C to -5 °C) of SA-EtOH mixture. (B) Optical images showing the change in appearance of pure SA and SA-EtOH mixture before and after cooling.

Fig. 19 depicts 3D printed hollow gel scaffolds. (A) Storage moduli (G') and loss moduli (G") of SA-EtOH as a function of the oscillation strain at 4 °C and 25 °C. (B) Shear-thinning behavior of SA-EtOH at 4 °C and 25 °C. (C) Schematic diagram of 3D printing setup. (D) Photos of printed SA-EtOH scaffold at different stages: after printing, after heating and after injecting dye solution. The scale bar is 1 cm.

Fig. 20 depicts the proposed mechanism of ethanol-induced gelation of SA.

Description

It has been surprisingly found that ink formulations formed from water, a suitable polymeric material and an organic solvent can solve some or all of the problems identified above. Thus, in a first aspect of the invention, there is provided a method of forming a three-dimensional hydrogel structure, the method comprising:

(a) providing an ink formulation comprising water, a polymeric material and an organic solvent to an additive manufacturing device; and (b) using the additive manufacturing device to print the ink formulation onto a substrate to generate a three-dimensional structure, wherein: the polymeric material is selected from one or more of the group selected from a polysaccharide, a protein and a synthetic polymeric material that has functional groups that can generate hydrogen bonds; the organic solvent is selected from one or more of the group consisting of an alcohol, acetone, dimethylformamide and dimethylsulfoxide; and in step (b) either the substrate is held at a temperature that is below a sol-to-gel transition temperature of the ink formulation, or in step (b) the printing step is conducted in an environment that has a temperature below the sol-to-gel transition temperature of the ink formulation, so as to enable the printed ink to undergo a sol-to-gel transition upon printing.

In embodiments herein, the word “comprising” may be interpreted as requiring the features mentioned, but not limiting the presence of other features. Alternatively, the word “comprising” may also relate to the situation where only the components/features listed are intended to be present (e.g. the word “comprising” may be replaced by the phrases “consists of” or “consists essentially of”). It is explicitly contemplated that both the broader and narrower interpretations can be applied to all aspects and embodiments of the present invention. In other words, the word “comprising” and synonyms thereof may be replaced by the phrase “consisting of’ or the phrase “consists essentially of’ or synonyms thereof and vice versa.

The phrase, “consists essentially of’ and its pseudonyms may be interpreted herein to refer to a material where minor impurities may be present. For example, the material may be greater than or equal to 90% pure, such as greater than 95% pure, such as greater than 97% pure, such as greater than 99% pure, such as greater than 99.9% pure, such as greater than 99.99% pure, such as greater than 99.999% pure, such as 100% pure.

When used herein, the terms “additive manufacture” and “3D-printing” may be used interchangeably. These terms are intended to refer to a method of producing an object in a layer-by-layer fashion based on a computational model of said object, with a computer (or similar device) controlling a machine that is capable of generating said layers. Any such suitable devices may be used herein, and any suitable object may be obtained.

As noted herein, the organic solvent is selected from one or more of the group consisting of an alcohol, acetone, dimethylformamide and dimethylsulfoxide. For example, the organic solvent may be selected from one or more of the group consisting of methanol, ethanol, 1- propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide. In particular embodiments that may be mentioned herein, the organic solvent may be ethanol.

As noted above, the polymeric material is selected from one or more of the group selected from a polysaccharide, a protein and a synthetic polymeric material that has functional groups that can generate hydrogen bonds. As such, the polymeric material may include functional groups that can interact with water and/or the organic solvent. Non-limiting examples of such functional groups include hydroxyl groups, ketones, aldehydes, carboxylic acids, esters, amino groups, amides, and the like. Particular examples of suitable polymeric materials that may be mentioned herein include, but are not limited to sodium alginate, K-carrageenan, chitosan, chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin, pullulan, poly(vinyl alcohol) (PVA), hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP), polyurethane (PU), gelatin, albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein and zein, and combinations thereof. In particular embodiments that may be mentioned herein, the polymeric material may be sodium alginate.

It will be appreciated that certain combinations of the polymeric material and the organic solvent may provide particularly good results. This can be determined by the skilled person based on their own common knowledge in light of the examples provided below. Examples of particularly good combinations of polymeric materials and organic solvents are provided below.

In particular embodiments that may be mentioned herein, when the polymeric material is sodium alginate, then the organic solvent may be selected from one or more of the group of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide (e.g. such as ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol).

In particular embodiments that may be mentioned herein, when the polymeric material is K- carrageenan, chitosan, chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin or pullulan, then the organic solvent may be selected from one or more of the group of ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol.

In particular embodiments that may be mentioned herein, when the polymeric material is poly(vinyl alcohol) (PVA), hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP) or polyurethane (PU), then the organic solvent may be acetone.

In particular embodiments that may be mentioned herein, when the polymeric material is gelatin, albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein or zein, then the organic solvent may be selected from one or more of the group of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol.

While any suitable ink formulation that is capable of being printed to provide a product under the conditions mentioned hereinbefore, may be used, it is noted that ink formulations that have the following component amounts may be particularly good. That is, an ink formulation that has: from 2 to 10% w/v of the polymeric material; from 25 to 40% v/v of the organic solvent; and from 60 to 75% v/v of water.

For example, the polymeric material may be present in an amount of from 2 to 4% w/v of the ink formulation, such as about 3.3% w/v. The organic solvent may be present in an amount of from 30 to 35% v/v of the ink formulation, such as about 33.3% v/v.

In a particular embodiment that may be mentioned herein, the ink formulation may be one that has: from 2 to 10% w/v of sodium alginate; from 25 to 40% v/v of ethanol; and from 60 to 75% v/v of water.

In such embodiments, the sodium alginate may be present in an amount of from 2 to 4% w/v of the ink formulation, such as about 3.3% w/v. The ethanol may be present in an amount of from 30 to 35% v/v of the ink formulation, such as about 33.3% v/v.

As noted above, the printing step is conducted in an environment that has a temperature below the sol-to-gel transition temperature of the ink formulation, so as to enable the printed ink to undergo a sol-to-gel transition upon printing. As such, the temperature of the substrate, or the environment of the printing chamber (or space), or both may be controlled so that the substrate and/or the environment has a temperature that is selected to match the properties of the particular ink formulation that is being used. For example, the temperature of the substrate and/or environment in step (b) of the method may be less than or equal to 15 °C and greater than or equal to -25 °C, such as less than or equal to 5 °C and greater than or equal to -20 °C (e.g. greater than or equal to -10 °C).

The exact temperatures chosen will depend on the ink formulation that is used. That is, the highest temperature that can be used will be determined so as to be less than the sol-to-gel transition temperature of the particular ink formulation in question and the lowest temperature may be determined as one that does not result in delamination between layers of the printed ink formulation. The delamination may be caused due to a too fast gelation of the printed layers, which may result in less cohesion between the layers.

In embodiments of the invention making use of an ink formulation formed from sodium alginate, ethanol and water (e.g. as described hereinbefore), the temperature of the substrate and/or environment in step (b) of the method may be less than or equal to 15 °C and greater than or equal to -25 °C, such as less than or equal to 5 °C and greater than or equal to -20 °C (e.g. greater than or equal to -10 °C).

In particular embodiments that may be disclosed herein, the ink formulation may contain sodium alginate at a concentration of 3.3% w/v, ethanol at a concentration of 33.3% v/v and water in an amount of 66.7% v/v and the temperature of the substrate and/or environment in step (b) may be less than or equal to 5 °C.

The method disclosed herein may contain further steps, depending on the final product that is required. For example, the method may further comprise:

(c) immersing the three-dimensional structure generated in step (b) in a gelation bath, where the gelation bath comprises water and a suitable crosslinking agent for the polymeric material, so as to provide a crosslinked structure.

In such embodiments, the gelation bath may further comprise the organic solvent. In still further embodiments, the concentration of the organic solvent in the gelation bath may be about the same as the concentration of the organic solvent in the ink formulation. It is noted that having the composition of the gelation bath approximate the water and organic solvent concentration of the ink formulation may help to prevent possible release of the organic solvent from the printed structure into the gelation bath due to a difference in the concentration of the organic solvent in the two environments. As will be appreciated, the suitable crosslinking agent will depend on the polymeric material in question. Examples of suitable crosslinking agents include, but are not limited to CaCk and glutaraldehyde.

In embodiments of step (c), it is possible to then generate a hollow three-dimensional structure. This may be achieved by one of the following:

(i) the temperature of the gelation bath may be elevated above the boiling point of the organic solvent, so as to provide a hollow three-dimensional structure; or

(ii) the crosslinked structure may be placed into a further bath comprising water at a temperature above the boiling point of the organic solvent, so as to provide a hollow three- dimensional structure. In (ii), the further bath further may comprise the organic solvent. For example, the concentration of the organic solvent in the further bath is about the same as the concentration of the organic solvent in the ink formulation.

As will be appreciated, if a hollow three-dimensional structure is desired, then the organic solvent will be selected to be one that has a boiling temperature that is lower than that of water (e.g. dimethylsulfoxide and dimethylformamide are not suitable for this hollowing-out downstream process step).

As will be appreciated, it may be desirable to remove the organic solvent from the finalised product. This may be achieved by simply allowing the organic solvent to diffuse out of the product over time. However, it is possible to ensure that the organic solvent is removed by immersion in water. Thus, in a further step, the method may include:

(d) immersing the three-dimensional structure generated in step (b) or step (c) in water to remove the organic solvent.

As noted hereinbefore, step (d) may be conducted after step (b) or step (c). That is, step (c) is an optional step and is not required. For the avoidance of doubt, this is also the case for step (d).

As noted above, in certain embodiments that may be mentioned herein, the polymeric material may be sodium alginate and the crosslinking agent may be a calcium salt (e.g. CaCk). In such embodiments, the concentration of the calcium salt may be from 0.05 to 0.2 M, such as about 0.1 M in the gelation bath. In such embodiments, the organic solvent may be ethanol.

In a second aspect of the invention, there is provided a method of forming hollow 1 D, 2D and 3D structures, the method comprising the steps of: (ai) providing a solid 1 D, 2D or 3D structure in an uncured state, the solid 1D, 2D or 3D structure comprising a polymeric material, an organic solvent and water; and either:

For the avoidance of doubt, the 1 D, 2D and 3D structures referred to herein are all three- dimensional in nature, but vary depending on their complexity.

In certain embodiments, the term “1D” here refers to the formation of single hollow tubes. These may be formed through the use of something as simple as a syringe to provide a “worm” of material that can then undergo processing to provide the desired hollow tube. In other words, an object in the tube can only move in one direction - forwards or backwards.

In certain embodiments, the term “2D” here refers to the formation of a series of interconnected hollow tubes that extend along a single plane (e.g. along the x-y plane). Such structures may be accomplished through moulding the ink formulation.

In certain embodiments, the term “3D” here refers to the formation of a series of interconnected hollow tubes that extend along all three planes (e g. along the x-y plane, the x-z plane and the y-z plane). Such structures may be achieved through the use of additive manufacturing.

In certain embodiments that may be mentioned herein (for the second aspect), the polymeric material may be sodium alginate and the crosslinking agent is a calcium salt (e g. CaCl2), optionally wherein the concentration of the calcium salt is from 0.05 to 0.2 M, such as about 0.1 M in the gelation bath, further optionally wherein the organic solvent is ethanol.

Further aspects and embodiments of the invention will now be discussed by reference to the following non-limiting examples.

Examples

Materials

Sodium alginate (SA) powder was purchased from Sigma. The rest of the materials used in the Examples are common chemicals purchased from commercial sources.

Example 1. Preparation of SA-ethanol precursors

The SA powder was first dissolved into deionized (DI) water to make a SA stock solution at a concentration of 10% w/v. The SA-ethanol gel precursor was then prepared by slowly mixing the SA stock solution with pure ethanol solution and DI water to reach a desired SA concentration (e.g. 2% w/v, 3.33% w/v and 4.5% w/v) and desired ethanol concentration (e.g. 0% v/v, 10% v/v, 15% v/v, 25% v/v, 30% v/v, 33.3% v/v, 35% v/v, 40% v/v and 45% v/v).

Example 2. Ethanol-induced temperature-responsive gelation

SEM characterization of samples

SEM was performed using a scanning electron microscope (JEOL JSM-5600LV, Japan). All samples were placed at 4 °C for 30 min, followed by a freeze-drying process using a freeze- drying machine (Christ Alpha 1-4 LSCbasic, Germany) before SEM characterization.

Light transmittance characterization

SA-ethanol precursors were prepared and placed in polystyrene cuvettes. The transmittance of the precursors was measured using a UV-Vis spectrophotometer (BioDrop Duo, USA) at 750 nm.

Rheological characterization

Discovery Hybrid Rheometer HR-2 was used to test oscillatory frequency sweeps, oscillatory amplitude sweeps, flow sweeps and temperature ramps. For the oscillatory frequency sweep tests, the applied strain was 1 %, while the shear frequency varied from 0.1 to 100 rad/s. The oscillatory amplitude sweep tests were preceded at a frequency of 1 rad/s and within a deformation range of 0.1-1000%. For the flow sweep tests, the shear rate varied from 0.1 to 100 s'¹. The temperature ramp tests were carried out at an angular frequency of 10 rad/s, and the temperature rate was 6 °C /min. The elastic modulus data was directly given by the rheometer.

Results and discussion

First, we mixed SA solution at a concentration of 3.3% w/v with ethanol at various concentrations by following the protocol in Example 1 and kept the mixtures at either room temperature (25 °C) or 4 °C for 30 mins. As shown in Figs. 1A and 2, when ethanol concentration exceeds 15% v/v, the SA solution becomes increasingly less transparent with the increase in the ethanol concentration. Interestingly, adding small amounts of ethanol (< 15 %) increases the light transmittance of the SA solution (Fig. 2), which may be associated with the hydrogen bond network enhancement in water facilitated by trace amounts of ethanol (M. Mijakovic et al., J. Mol. Liq. 2011 , 164, 66-73). At room temperature, the SA-ethanol mixture remains in the liquid phase when the ethanol concentration is lower than 40%, as indicated by the free-flowing of the mixture after the glass vials are turned upside down. The mixture becomes unstable and clear phase separation occurs when the ethanol concentration exceeds 40%. In contrast, evident gelation occurs at intermediate ethanol concentrations (30% - 40%) at 4 °C. The mixture remains in the liquid state when the ethanol concentration is below 30% and becomes unstable when the ethanol concentration is higher than 40%.

We then characterized the rheological properties of SA-ethanol mixtures to interrogate the ethanol-induced gelation process quantitatively. Rheological testing was conducted immediately after SA-ethanol mixtures were prepared. Temperature ramps were carried out, during which samples were cooled down from room temperature at 25 °C to -5 °C, held for 3 mins at -5 °C and then warmed up to 40 °C. For pure SA solution without ethanol, both storage modulus (G’) and loss modulus (G") increase linearly with the decreasing temperature and the modulus curves obtained during the cooling and heating processes perfectly coincide with each other (Figs. 1 B and 3). The increase (decrease) of G’ and G” in the cooling (heating) process could be attributed to the reduction (increase) in molecular chain mobility (S. Liu et al., J. Rheol. 2016, 60, 203-214). Adding ethanol into the SA solution increases both G’ and G" (Figs. 1 B and 3). The mixtures with ethanol concentrations below 25 % remain as a viscous liquid within the experimental temperature interval as indicated by their relatively smaller G’ value than G” (Fig. 4). When the ethanol concentration reaches or exceeds 25%, the mixtures undergo a sol-to-gel transition as the temperature is reduced from room temperature to -5 °C (Fig. 1 B). A substantial increase in elastic modulus by three orders of magnitude was observed when the ethanol concentration reached beyond 25% (Fig. 1C). The sudden decrease in the elastic modulus at an ethanol concentration of 40 % can be attributed to the reduced overall interaction between polymer chains as a result of the separation and precipitation of large polymer clusters from the solvent, which is consistent with the transition of the SA-ethanol mixture from a gel-like state to a turbid liquid observed in glass vials (Fig. 1 A). It is shown that the sol-to-gel transition temperature increases with the increase in the ethanol concentration (Fig. 1D). Since the transition temperature is always higher than the corresponding freezing point of the ethanol-water mixture (K. Takaizumi & T. Wakabayashi, J. Solution Chem. 1997, 26, 927-939), it rules out the possibility that the sharp increase in G’ is caused by the freezing of the mixture and confirms that the SA-ethanol mixture undergoes a sol-to-gel transition, which is consistent with the observations made in the glass vials (Fig. 1A). In addition, the concentration of SA also affects the rheological properties as well as the sol-to-gel transition behavior (Fig. 1 E). Specifically, the storage modulus of the SA-ethanol mixture increases with the increasing SA concentration at -5 °C (Fig. 1F), and a higher SA concentration results in a higher sol-to-gel transition temperature (Fig. 1G). The results above suggest that the sol-to- gel transition temperature can be controlled by varying the ethanol concentration and/or the SA concentration.

From a thermodynamic point of view, the sol-to-gel transition is governed by the Gibbs equation: AG = AH - TAS, where AH, AS and T are the enthalpy change associated with ethanol-induced gelation (AH < 0), the entropy change associated with the transition from a disordered liquid phase to an ordered gel phase (AS < 0) and the temperature, respectively. Hence, spontaneous sol-to-gel transition (AG < 0) is possible only when the temperature is below a critical value, i.e., T < T_c = AH/AS. Since the enthalpy change is expected to correlate positively with both the ethanol concentration and the SA concentration in the SA-ethanol mixture, the expected sol-to-gel transition temperature should increase with both the increasing ethanol concentration and SA concentration, which is consistent with our experimental observation (Figs. 1D and 1G).

Example 3. Effect of ethanol on SA hydrogel microstructure

Small-angle X-ray scattering (SAXS) analysis

SAXS patterns were obtained from a Xenocs Nanoinxider with Cu-Ka microsource (40 mm, A = 0.154 nm) at 30 W. Samples were prepared by injecting gel precursor solutions into capillary glass tubes with a diameter of 1.5 mm. Samples were first cooled from 25 °C to -10 °C and then heated from -10 °C to 40 °C at a rate of 6 °C/min for both cooling and heating processes. Samples were exposed for 60 s for every 5 °C change. All the data were background- subtracted. Data were analyzed using SasView, BioXTAS RAW (J. B. Hopkins et al., J. Appt. Crystallogr. 2017, 50, 1545-1553), and GNOM of ATSAS (K. Manalastas-Cantos et al., J. Appl. Crystallogr. 2021 , 54, 343-355). The P(r) function plots were obtained using BioXTAS RAW and GNOM of ATSAS.

The fractal model reported by Teixeira was used to fit our SAXS results (J. Teixeira, J. Appl.

Crystallogr. 1988, 21, 781-785). In this model, the scattered intensity is given by

where q is the scattering vector magnitude, < > is the volume fraction of building blocks with a radius of r₀, Vbiock is the volume of a single building block, p_biock and p_soivent are the scattering length densities of the building blocks and solvent, respectively, and B is the parameter about the background. P(q) and S q) are the scattering factors to describe the minimum units (building blocks) and the interference in a fractal structure from such building blocks, respectively. Herein, the minimum units (building blocks) are considered as uniform spheres. The sphere’s center is located at the midpoint of the shortest distance between two chains, and the shortest distance is the diameter of the sphere. The distance mentioned here includes the “thickness” of the polymer chain itself, i.e., this ball contains part of the chains. Thus

and

and D_f- are the correlation length and fractal dimension, respectively.

can be used to describe the size of the cluster in practice. F(x) is the gamma function of argument x. The radius of gyration R_g is given by

This fractal model could well fit the experimental results for the SA-ethanol mixtures with 25%, 33.3% and 40% v/v ethanol at -10 °C using SasView software and the fitting results were summarized in Table 1.

Table 1. Summary of fitting parameters of the fractal model.

Results and discussion

Ethanol-induced gelation was confirmed by the microstructures of SA-ethanol gel. Freeze- dried SA-ethanol mixtures with different ethanol concentrations were examined using SEM. As shown in Fig. 1 H, the dried pure SA sample features a lamellar structure, which is a typical microstructure of freeze-cast polymers due to the unidirectional solidification of polymer solutions. Incorporating ethanol into SA solution changes the microstructure from lamellar to network structure. The pore size of the network structure decreases with the increasing ethanol concentration. These porous microstructures are generally similar to that of SA hydrogel gelled with Ca²⁺ ions (Fig. 5). It is noted that the sample with 33.3 % v/v ethanol displays a hierarchical network structure with finer porous networks formed inside some of the large pores (Figs. 1 H and 5). Samples containing ethanol at concentrations higher than 40 % could not maintain their structures and became powdery after drying, which is consistent with the phase separation phenomenon observed in the glass vial (Fig. 1A).

To gain information about the in situ structural conformations of the hydrogel formed in the presence of ethanol, we performed SAXS experiments on samples with various ethanol concentrations as they were cooled from 25 °C to -10 °C. As an X-ray beam is brought to a sample, the scattered X-rays form a scattering pattern representing average information about the sample’s nanoscale structures, such as pore sizes, shapes and orientations (S. Skou et al., Nat. Protoc. 2014, 9, 1727-1739). The 2D scattering intensity depends on the scattering vector magnitude, known as q - (4n sinS) , where 2d is the scattering angle, and A is the wavelength of the incoming X-ray beam.

Fig. 6A shows the 2D SAXS scattering curves of the SA-ethanol mixture with 33.3% v/v ethanol at various temperatures. The increase in scattering intensity with the decrease in temperature within a wide range of wave numbers (0.005 to 0.1 A ¹) suggests an upper critical solution temperature (UCST) behavior, which is consistent with the abovementioned results. Presenting the scattering profiles of the samples as Kratky plots in Fig. 6B exhibits peaks at the low-q region. The peaks in the Kratky plot indicate the existence of inhomogeneity, i.e. , the clustering of polymer chains. It is shown that the peak becomes sharper and shifts to a lower q value with the decrease in temperature, indicating the gradual growth of polymer clusters as the temperature is reduced (Y. Yuguchi et a!., Carbohydr. Polym. 2016, 152, 532- 540). The cooling-induced formation and growth of polymer clusters are also confirmed by comparing the slopes of scattering profiles within the high-q region at different temperatures in the normalized Kratky plot in Fig. 6C. In general, a larger magnitude of the slope at the high- q region corresponds to a higher degree of polymer chain flexibility. Specifically, a higher slope corresponds to worm-like polymer chains, while a lower slope suggests that chains behave more like a rod (W. Burchard, Light Scattering from Polysaccharides as Soft Materials, in: R. Borsali, R. Pecora (Eds.), Soft-Matter Characterization, 1st ed., Springer, New York, 2008, pp. 463-603). Our results demonstrate that the flexibility of alginate chains generally decreases with the decrease in temperature due to the formation and continuous growth of clusters. Fig. 6D shows the scattering profiles of SA-ethanol samples with various ethanol concentrations at -10 °C. As the ethanol concentration increases from 0% to 40%, the scattering intensity within the low-q range (<0.1 A'¹) increases, indicating a promoted polymer cluster formation caused by ethanol. Similarly, the corresponding Kratky and normalized Kratky plots in Figs. 7E-F further confirm that introducing ethanol could enhance the formation of clusters and cause a decrease in polymer chain flexibility.

Fitting our scattering intensity profiles using a Bayesian indirect Fourier transformation (I FT) method allows us to obtain the Pair-distance distribution function (i.e., P(r) function), which provides valuable information about the shape and size of the samples’ internal microstructures. The P(r) function falls gradually to zero at the maximum dimension (D_max) of the internal features (Fig. 6G). For the pure alginate solution, our fitting reveals that D_max = 18 A, and the symmetry of the curve indicates that the shape of internal polymer clusters in pure alginate is almost spherical. However, when ethanol is introduced into the SA solution, the Dmax value increases to about 600 A. Interestingly, the D_max value displays no progressive increase with ethanol concentration. The notable rise in D_max suggests a transition from freely diffusing polymer chains to polymer clusters as a result of ethanol-induced interpolymer interactions. Moreover, the peak of P(r) curve shifts to a higher r region with the rise in ethanol concentration, suggesting that the shape of clusters becomes more disc-like from rod-like (D. I. Svergun & M. H. J. Koch, Rep. Prog. Phys. 2003, 66, 1735-1782).

For a more quantitative understanding of the microstructure of the SA-ethanol gel, the SA- ethanol gel system was considered from a fractal aspect, since the rise in the scattering intensity, when ethanol is incorporated, can be attributed entirely to the increase in the local inhomogeneity caused by ethanol. It was assumed that the minimum unit of this inhomogeneity can be treated as two alginate chains (or different parts of one folded chain) that are close to each other after being dehydrated by ethanol, as illustrated in Fig. 6I. The fractal behavior can be described using a fractal model developed by Teixeira (J. Teixeira, J. Appl. Crystallogr. 1988, 21, 781-785). The fitting curves are shown in Fig. 6H, and the fitting results are summarized in Table 1. The fitting results reveal that the average radius of gyration (R_g) of the alginate clusters in the SA-ethanol gel was around 250 A and only slightly increased as ethanol concentration increased from 25% to 40%. This result is consistent with the Dm_ax value predicted from the P (r) function. Within each polymer cluster, some fractal polymer-rich areas exist with an average characteristic size of

100 A. The fractal dimension Df increased as ethanol concentration increased, indicating an increase in inhomogeneity. Although fractal polymer-rich regions’ size did not notably change with ethanol concentration, the presence of ethanol could effectively bring alginate chains closer and contribute to enhanced inhomogeneity. The average radius of minimum units (r₀) decreased from 8.35 ± 0.31 A to 4.65 ± 0.11 A with the increase in ethanol concentration from 15% to 40%.

Example 4. Molecular dynamics (MD) simulations of alginate chain assembly

Molecular dynamics simulation of alginate chains

The poly-G and poly-M chains consisting of 16 residues each were generated using ChemDraw3D. The solution builder module of CHARMM-GUI web-server was then used to generate simulation models and obtain the required parameter files for MD simulations (S. Jo et al., J. Comput. Chem. 2008, 29, 1859-1865; and J. Lee et al., J. Chem. Theory Comput. 2016, 12, 405-413). MD simulations were performed using GROMACS version 2018.2 with the all-atom CHARMM27 force-field to equilibrate the single poly-G and poly-M chains (M. J. Abraham et al., SoftwareX 2015, 1-2, 19-25; S. Pall et al., Tackling Exascale Software Challenges in Molecular Dynamics Simulations with GROMACS, in: S. Markidis, E. Laure (Eds.), Solving Software Challenges for Exascale, Springer, Cham, 2015, pp. 3-27; and S. Pronk et al., Bioinformatics 2013, 29, 845-854). Each chain was placed in a periodic cubic simulation box which was then filled with TIP3P water molecules. Na⁺ and Cl' ions were included to make the system electrically neutral. Steepest descent energy minimization was first carried out for 5000 steps to remove any inter-atomic steric clashes. Next, dynamics simulation was carried out with positional restraints on the alginate atoms using canonical (NVT) ensemble for 125 ps at 303 K. Electrostatic interactions were computed using the Particle Mesh Ewald method with a cut-off distance of 1.2 nm. Van der Waals interactions were computed using a cut-off method with a cut-off distance of 1 .2 nm. System temperature was maintained with the Nose-Hoover method with a time constant of 1 ps. Lastly, restraints on the alginate atoms were removed and production simulation was carried out for 20 ns using canonical (NPT) ensemble with temperature maintained at 303 K using the Nose-Hoover method as above, and the system pressure maintained at 1 bar using the Parrinello-Rahman method with a time constant of 5 ps and compressibility of 4.5 x 10⁵ bar¹.

To investigate the gelation of alginate chains under the influence of Ca²⁺ ions or ethanol molecules, five copies of the equilibrated poly-G or poly-M chains were placed at random locations inside a cubic periodic box with a side length of 8.7 nm. For simulations to study the effect of Na⁺ or Ca²⁺ ions on self-assembly of alginate chains, the simulation box was then filled with TIP3P water molecules and either Na⁺ or Ca²⁺ ions were added to neutralize the system electrically. For simulations to investigate the role of ethanol in inducing alginate gelation, after placing the alginate chains, the simulation box was then filled with the appropriate number of ethanol molecules to reach a final concentration of 40% v/v. This was then followed by the placement of TIP3P water molecules and then Na⁺ ions to neutralize the system electrically. MD simulations for all these systems were performed as outlined above for single chain equilibration but with production simulation carried out for 200-300 ns. Visual Molecular Dynamics (VMD) software v1.9.4 was used to visualize simulation results (W. Humphrey, A. Dalke & K. Schulten, J. Mol. Graphics 1996, 14, 33-38). Molecular clustering was analyzed using GROMACS tool clustsize with a cut-off distance of 6 A. This cut-off distance corresponds to the peak location of the radial distribution function computed for GG or MM association in the presence of Ca²⁺ ions (H. Hecht & S. Srebnik, Biomacromolecules 2016, 17, 2160-2167).

Results and discussion

To understand the molecular mechanism underlying ethanol-induced gelation, atomistic MD simulations of alginate self-assembly in the presence of charge-neutralizing Na⁺ or Ca²⁺ ions in solution or the presence of different volume fractions of ethanol were conducted. We first simulated a single poly-G or poly-M chain with sixteen monomers per chain for 20 ns (Fig. 7A) and then placed five copies of each type of alginate chain into a simulation box at random locations. The simulation system had an alginate concentration of 3.3% w/v, which is the same as used in the experiments described herein. As expected, Na⁺ ions do not induce any notable clustering to form clusters with more than two chains for both the poly-G and poly-M systems, while notable cluster formation occurs in the presence of Ca²⁺ ions for the poly-G system (Fig. 7B) (O. Wichterle & D. Lim, Nature 1960, 185, 117-118; and A. C. Daly et al., Nat. Rev. Mater. 2020, 5, 20-43). The Ca²⁺ ions bring the poly-G chains close enough for side-chain hydrogen bonding to occur, forming both parallel and perpendicular arrangements (Figs. 7C and 8), as observed in previous simulation studies (W. Plazinski, J. Comput. Chem. 2011 , 32, 2988- 2995; and M. B. Stewart et ai., Carbohydr. Polym. 2014, 102, 246-253). In contrast to poly-G chains, poly-M chains showed a lower degree of clustering with Ca²⁺ ions (Fig. 7E), although clusters of five chains could still occur transiently as shown in Fig. 7F with the bridging Ca²⁺ ions. The networks formed by aggregates also seem more open than poly-G ones but also in predominantly perpendicular arrangements. This agrees with previous experimental and simulation observations of Ca²⁺-induced alginate chain self-assembly (H. Hecht & S. Srebnik, Biomacromolecules 2016, 17, 2160-2167; M. B. Stewart et al. , Carbohydr. Polym. 2014, 112, 486-493; and H. Hecht & S. Srebnik, Carbohydr. Polym. 2017, 157, 1144-1152).

Consistent with our experimental observation of ethanol-induced gelation, we found that, with ethanol at a concentration of 40% v/v, both the poly-G and poly-M systems exhibit a clustering level that falls between Na⁺- and Ca²⁺-induced clustering (Figs. 7B and 7E). Compared to poly- G chains, ethanol enhances the clustering of poly-M chains to a lower degree, with cluster sizes of mostly two to three chains. The largest cluster size in both systems could reach five, as in the case of Ca²⁺ ions, though the largest clusters are more transient. Similarly, the perpendicular arrangement of the chains seems to be preferred, and this type of arrangement has been previously proposed to facilitate the formation of open 3D networks (M. B. Stewart et al., Carbohydr. Polym. 2014, 102, 246-253). Examples of a four-chain poly-G aggregate and a three-chain poly-M aggregate are shown in Figs. 7D and 7G, respectively, where the chains are bridged by hydrogen-bonded water clusters and/or directly hydrogen-bond to each other via the hydroxylate and carboxylate groups. In contrast, water molecules do not appear to take part in bridging poly-G chains in the presence of Ca²⁺ ions, as their chains could associate much more closely. Ethanol molecules are not observed to participate directly in bridging alginate chains. Nevertheless, they could disrupt the percolating H-bond network of water at a concentration higher than 30% (S. Y. Noskov et al., J. Phys. Chem. B 2005, 109, 6705-6713). The first hydration shell of ethanol is depleted of water with the structuring of water in the second hydration shell (S. Y. Noskov et al., J. Phys. Chem. B 2005, 109, 6705- 6713). The disruption of the water H-bond network by ethanol as the alcohol concentration increases (again more substantial at higher than 30%) has also been shown in another study (S. Choi et al., Phys. Chem. Chem. Phys. 2020, 22, 17181-17195). Thus, disruption of the percolating water H-bond network by ethanol beyond a critical concentration facilitates the association of alginate chains as it becomes easier for the chains to come closer to each other.

During our clustering analysis, we used a cut-off distance of 6 A for chains to be considered in the same cluster. This distance matches the peak location of the radial distribution function of the center of mass of different poly-G or poly-M chains in the presence of Ca²⁺ ions (H. Hecht & S. Srebnik, Carbohydr. Polym. 2017, 157, 1144-1152). This is also comparable with the value of ro obtained from SAXS fitting (Table 1). In fact, when we measured the inter-chain spacing observed from the clustered configurations in our simulation, we found that the sampled average space between two G16 in the presence of ethanol shown in Fig. 7D happened to be very close to 2ro or the diameter of the minimum units shown in Figs. 6I and 9.

Example 5. Direct 3D printing of SA-ethanol mixtures

3D printing

All samples (pure SA solution and SA-ethanol mixtures prepared in Example 1) were loaded into a printing syringe with a printing needle (20G 0.023" or 22G 0.016") and printed onto a cold plate surface (about -10 °C) using a 3D printer (Cellink Bio X, Sweden). The gelled scaffolds were obtained by immersing the printed structures into a 0.1 M CaCl2 solution with 33.3 % v/v ethanol for overnight. Then, the ethanol inside the scaffolds was removed by soaking them in DI water for 1 h.

FTIR-ATR spectra acquisition

Infrared spectra were recorded on a Fourier-transform infrared spectroscopy (FTIR) spectrophotometer (Thermo Fisher Scientific Nicolet iS50, United States) with an attenuated total reflectance (ATR) accessory. The FTIR-ATR spectra were obtained against the background of the air spectrum. Background and samples were measured at room temperature from 4000 to 400 cm'¹. The SA-Ca sample was prepared by immersing the SA sample into 0.1 M CaCh until it fully gelled. The SA-Ca-ethanol sample was obtained by soaking the SA-ethanol sample into a 0.1 M CaCh solution with 33.3% v/v ethanol overnight, while the SA-Ca-deEthanol sample was prepared by immersing the SA-Ca-ethanol sample into DI water for 1 h.

Results and discussion

Using a rheometer, the rheological properties of moduli of SA-ethanol mixtures were first measured by following the protocol in Example 2. The SA-ethanol mixtures exhibited higher values than its loss moduli at 4 °C under oscillation strains below 30% (Fig. 10A, left), which enables the printed SA-ethanol objects to be stackable at this temperature. The emergence of a peak in the G” plot of the SA-ethanol mixture at 4 °C at a relatively high strain suggests the presence of an interconnected network in the sample (H. Ma et al., ACS Nano 2019, 13, 4302—4311) Furthermore, the presence of ethanol significantly augmented the viscosity of the SA solution by several orders of magnitude (Fig. 10A, right). More importantly, while the viscosity of pure SA solution is weakly dependent on shear rate, the viscosity of SA-ethanol mixtures notably decreased with increasing shear rate, showing an enhanced shear-thinning behavior. Cyclic strain time sweeps were applied to samples to closely simulate the time scales and mechanical forces experienced by the ink during printing. Fig. 10B shows that all the samples display sharp decreases in G’ at high strains (500 %), which are recovered immediately at low strains (0.5 %). The rapid transition from solid-like to liquid-like behavior when strain is applied (i.e., thixotropy) makes the SA-EtOH well-suited for extrusion-based printing. In addition to a high G’, shear-thinning behavior and rapid recovery of elasticity, maintaining a sufficient yield stress following extrusion is crucial for preventing possible distortion to printed structures. As shown in Fig. 10C, the dynamic yield stress increases from less than 50 Pa to about 170 Pa after cooling due to ethanol-induced gelation. The observed yield stress level at 4 °C is higher than other 3D printing inks reported in the literature (D. Kokkinis et al., Nat. Commun. 2015, 6, 8643; and M. Schaffner et al., Sci. Adv. 2017, 3, eaao6804). Thus, the introduction of ethanol makes the printing of the formerly non-printable SA solution feasible.

We then demonstrated that the SA-ethanol mixture could be used to print SA hydrogel structures directly using an extrusion-based 3D printer. To facilitate quick gelation and maintain printed patterns, we directly printed the ink solution onto a cold plate whose surface temperature was maintained at about -10 °C, below the sol-to-gel transition temperature of the SA-ethanol mixture. As shown in Fig. 11 A, the pure SA solution at a concentration of 3.3% w/v cannot be directly printed into the desired pattern. The extruded SA solution gradually fuses on the cold plate into large droplets rather than the desired linear structures before being frozen. In contrast, the SA-ethanol ink with the same alginate concentration can be reliably printed. More importantly, the sol-to-gel transition could quickly occur on the cold plate so the printed SA-ethanol solution could maintain its shape without collapse. We further printed a scaffold structure (20 mm x 20 mm x 3 mm) to show good accuracy of printing with SA-ethanol ink, as demonstrated in Fig. 11 B.

To fix the 3D structure of the printed scaffold, we immersed the scaffold in a CaCk solution mixed with ethanol at the same concentration as used in the SA-ethanol ink to allow Ca²⁺- induced gelation. The presence of ethanol in the CaCk solution was to prevent the release of ethanol from the printed structure to avoid any possible disruption to the scaffold’s shape during the gelation process.

Ethanol could then be removed after gelation by soaking the gelled SA-ethanol gel in DI water. The FTIR-ATR test results confirm that ethanol could be removed entirely in less than 1 h (Fig. 11C). The spectrum of the gel after soaking for 1 h becomes similar to that of a regular SA gel crosslinked with Ca²⁺ ions, without peaks associated with ethanol (highlighted by the shaded bands in Fig. 11C, including C-0 stretches at around 1000 to 1300 cm^-1 and C-H stretch at around 2900 cm'¹). Interestingly, the scaffold volume was reduced during the gelation and soaking processes (Fig. 11 B). As shown in Fig. 11D, due to the crosslinking action of Ca²⁺ ions, the volume of the sample has decreased by about 17 %. The removal of ethanol could further reduce the sample’s volume by ~17 %, resulting in a total reduction of ~34 % compared to its original size. Such an isotropic volume reduction can be considered as an effective way to enhance printing resolution. Considering the 1 D length change, the printing resolution increases by approximately 11 % in this regard.

In the present disclosure, we report the gelation of SA induced by ethanol, based on which direct 3D printing of alginate hydrogel structures was achieved. We first systematically investigated the UCST behavior of SA-ethanol mixtures with different ethanol concentrations by characterizing their rheological properties. We then characterized the influence of ethanol on the microscopic structures of the alginate hydrogel using SEM and SAXS techniques. MD simulations are performed to provide molecular insights into the mechanism underlying ethanol-induced gelation. The presence of ethanol notably elevates the viscosity of SA solution and enhances its shear-thinning feature, which makes the SA-ethanol mixture an ideal 3D printing ink. The UCST behavior of the SA-ethanol mixture allows us to easily maintain the printed patterns by depositing the printed patterns onto a cold plate whose temperature is kept below the sol-to-gel transition temperature of the SA-ethanol mixture. With this method, the ink is extruded out of the nozzle at room temperature (with low viscosity and low yield stress) and rapidly gels after cooling, significantly increasing viscosity and yield stress. This approach not only avoids extrusion difficulties caused by high viscosity but also enables direct printing of 3D hydrogel structures using low-concentration SA solutions. In addition, further gelling the printed structures with Ca²⁺ ions and removing ethanol lead to an isotropic structural shrinkage by ~34 %, thereby effectively improving printing resolution to generate hydrogel structures with feature dimensions even smaller than the size of the nozzle. The present disclosure provides a breakthrough in achieving direct printing of hydrogel materials without introducing additives that are difficult to remove and may benefit hydrogel-based applications in many other fields.

Example 6. Preparation of one- to three-dimensional hollow hydrogels

Preparation of SA-EtOH precursor

A mixture of 0.3 g sodium alginate (Sigma), 3 mL deionized (DI) water and 6 mL (50% v/v) ethanol was stirred thoroughly to obtain the SA-EtOH hydrogel precursor. Meanwhile, the same concentration of pure alginate solution (3.3% w/v) was prepared by mixing 0.3 g sodium alginate and 9 mL DI water. Producing one- to three-dimensional hollow hydrogels

To obtain the 1 D hollow hydrogel tubes, we loaded the precursor in a printing syringe with a printing nozzle (16G, 20G, or 22G) to directly press into a reaction container, which was a container filled with 0.1M CaC ethanol solution (33.3%) on a hotplate. The typical heating temperature was 86 °C for all the constructs if not specifically stated. For the 2D hollow hydrogel constructs, the SA-EtOH precursor was injected into a polydimethylsiloxane (PDMS) mould with specific shapes. After injection, the precursor and the mould were immersed in liquid nitrogen to solidify the precursor for demolding. The shaped hydrogel precursor was then placed into the reaction container to form hollow cavities inside. The 3D hollow hydrogel constructs were obtained by 3D printing at a low temperature (about -10 °C) based on CAD models to obtain hydrogel scaffolds. According to our previous research, the precursor can transfer to a gel state at a low temperature. This gelation process makes layer-by-layer printing possible. After heating, the hollow objects were immersed in DI water to remove ethanol and recover the shape.

Results and discussion

The formation of hollow structures is the outcome of the competition between two mechanisms. These two mechanisms of the competitive relationship can be broadly categorized into external and internal factors, facilitating our discussion. The external being primarily the permeation and gelation induced by calcium ions into the alginate precursor, while the internal factor involves the vaporization of ethanol (although this vaporization is also caused by the introduction of external heat into the interior of the material). To demonstrate the feasibility of this approach, we created one-dimensional hollow tubes by directly injecting sodium alginate (SA) precursor solution mixed with ethanol (EtOH) into a calcium chloride (CaCh) solution mixed with ethanol at the same concentration. The presence of ethanol in the CaCh solution prevented the possible release of ethanol mixed in the SA-EtOH precursor through diffusion. As illustrated in Fig. 12A, SA chains started to cross-link immediately from the outer surface after the SA-EtOH mixture was continuously injected into the CaCh solution, forming a gel shell. The shell helped maintain the overall gel shape and, at the same time, hindered the outflow and leakage of the inside precursor. When the temperature reaches the ethanol boiling temperature, ethanol gradually gasifies and forms gas bubbles. Small, isolated gas bubbles continuously grew and merged, leading to a hollow tube. The uniformity of the tube wall is related to the position and movement of bubbles. Therefore, we added stirring and rotation to the heated object, ensuring that it is continuously in a rolling state. This is done to keep the gas relatively located in the central region, ensuring the uniformity of the tube wall. Example 7. Investigation of wall thickness vs gelation time and/or heating time

Investigation of wall thickness vs gelation time and/or heating time

The prepared SA-EtOH precursor in Example 6 was loaded into a 10 ml_ syringe with a 16 G needle. Then the precursor was injected as a fibre into a container filled with 0.1 M CaCh ethanol solution (33.3 %) to gel the precursor for 30, 60, 120, 240, 300, and 600 s, respectively. After gelation, the fibres were immediately transferred into another container filled with 33.3 % (v/v) ethanol solution at 86 °C for 2 min, 5min, 10 min and 20 min, respectively. These fibres were then observed by a microscope.

Results and discussion

It is noted that the hollow structure tends to collapse after heating when the temperature is reduced below the boiling temperature. We have demonstrated that the collapsed structure can be easily recovered by soaking it in deionized (DI) water (Fig. 13). After fully restoring its shape in DI water, the hollow structure becomes filled with water, providing evidence that the structure is intact and undamaged.

Example 8. Characterization of one- to three-dimensional hollow hydrogels

The one- to three-dimensional hollow hydrogels prepared in Example 6 were characterized.

Characterization of SA-EtOH fibers

The SEM observation was performed by a scanning electron microscope (JEOL JSM-5600LV, Japan). Before observation, samples were operated by freeze-drying.

Results and discussion

In Fig. 12B, optical and SEM images illustrate the SA-EtOH hollow tubes (bottom right). As control groups, SA-EtOH fibers gelled in CaCh at room temperature for the same duration, as well as SA fibers obtained at both room temperature and after heating, are also displayed. As anticipated, these controls exhibit no hollow structures. Fig. 12C demonstrates the influence of heating temperature on SA-EtOH tubes. Here, the CaCl2 ethanol solution's boiling point is around 85 °C (within a covered heating container). When the heating temperature is well below the boiling point (70 °C), virtually no bubble formation is observed, resulting in solid gel fibers. As the temperature approaches the boiling point (80 °C), bubbles gradually emerge but cannot merge to form interconnected hollow tubes; instead, the generated bubbles are trapped within the gel tubes. At temperatures above the boiling point (90 °C), the vaporization rate of ethanol surpasses the gelation rate of SA by calcium ions. This leads to the fusion of bubbles, ultimately forming connected hollow tubes. However, at significantly elevated temperatures, ethanol vaporization occurs too rapidly. The rapidly generated gas exerts pressure, squeezing the ungelled SA-EtOH precursor solution. This causes the precursor to be pushed upward along the tube radius in a perpendicular direction, ultimately extruding the precursor outside the gel shell (rupture). Consequently, only a small portion of SA is gelled by calcium ions, resulting in a hollow tube with an extremely thin wall. Due to the extrusion of internal material, upon cessation of heating, this very thin tube rapidly collapses into a double-layered film, as depicted in Fig. 12C (bottom right). Therefore, an appropriate heating temperature is imperative for obtaining hollow tubes. Unless otherwise specified, the heating temperature in this study was consistently maintained at around 86 °C.

Example 9. Impact of calcium ion concentration on wall thickness

To investigate the influence of external factors, we specifically studied the impact of calcium ion concentration on wall thickness by following the protocol in Example 6. Hollow tubes were obtained (Fig. 14A) by uniformly extruding SA-EtOH precursor (containing 33.3% ethanol, prepared in Example 6) into boiling CaCl2 ethanol solutions of varying calcium concentrations (0.05, 0.1 , 0.2 M). Subsequently, after the reaction concluded, the samples were immersed in DI water to eliminate excess ethanol. Optical microscopy was employed to observe the hollow tubes, and ImageJ software was utilized to measure both the outer and inner diameters, allowing for the calculation of wall thickness and the ratio of wall thickness to outer diameter.

Results and discussion

The outer and inner diameters are shown in Fig. 14B. With the increase in calcium ion concentration, there is a slight reduction in the outer diameter, from 450 pm (0.05 M) to 425 pm (0.2 M). This could be attributed to gel contraction induced by calcium ion gelation, with higher concentrations leading to more substantial contraction. From the perspective of the egg-box model, the presence of calcium ions brings the chains closer together, inevitably resulting in macroscopic volume contraction (Grant, G. T. et al., Febs Lett. 1973, 32, 195- 198). In contrast, the inner diameter sharply decreases with increasing calcium ion concentration, decreasing from 315 pm (0.05 M) to 250 pm (0.1 M) and further to 190 pm (0.2 M). This reduction is a result of the faster penetration and cross-linking of SA induced by calcium ions. Utilizing the data of outer and inner diameters, we can calculate the average wall thickness and the wall thickness ratio (Fig. 14C). Since the wall thickness is calculated using the equation: average wall thickness = (outer diameter - inner diameter)/2, the reduction in inner diameter corresponds to an increase in wall thickness, ranging from approximately 65 pm (0.05 M) to 110 pm (0.2 M). The results align with our hypothesis that increasing calcium ion concentration accelerates gelation. The reinforcement of this external factor gives it an advantage in the competition, leading the outcome towards conditions favorable to external factors. This is manifested in the increase in wall thickness. Certainly, the permeation rate of calcium ions (or SA gelation thickness) also influences wall thickness. We investigated the gelation thickness of SA-EtOH in CaCl2 at room temperature. It shows an exponential relationship between penetration thickness and time (Fig. 15). Simultaneously, SA-EtOH exhibited a faster gelation rate in a calcium solution without ethanol than its counterpart, consistent with previously reported conclusions (Li, J. et al., Carbohydr. Polym. 2015, 123, 208-216). However, since this test was conducted at room temperature and without the other party's presence in the competitive relationship, it can only serve as a reference.

Similarly, we studied the relationship between wall thickness and ethanol concentration for the internal factor. Hollow tubes made from SA-EtOH with various ethanol concentrations (25%, 33.3% and 40% v/v) could be obtained in the 0.1 M CaCh solution (Fig. 14D). We also measured and recorded the outer and inner diameters of the hollow tubes (Fig. 14E), calculating their average wall thickness and the wall thickness ratio (Fig. 14F). The results appear to be the opposite of the influence of external factors discussed above: the outer diameter shows a slight increase with the rise in the variable (here is ethanol concentration), while the inner diameter experiences a significant enhancement. Both of these increases can be attributed to the expansion pressure exerted on the tube wall after ethanol vaporization. The presence of more ethanol results in the generation of a greater volume of gas, leading to increased pressure within the tube. Additionally, more gas occupies a larger volume. Consequently, the increase in ethanol concentration causes a reduction in tube wall thickness from 90 pm (25% v/v) to 50 pm (40% v/v). Therefore, we can control wall thickness by altering either party in the competitive relationship or modifying both simultaneously. The current competitive relationship occurs within the same time dimension. Alternatively, we can separate these two competitive processes, breaking them down into an initial external gelation process followed by an internal vaporization process (controlled by adjusting the heating time). Fig. 16 presents the wall thickness data for the hollow tubes obtained through this separated competition (stepwise method). These data align with the conclusions drawn above.

Due to the use of a larger needle, the wall thickness data for the tubes in Fig. 16 (16 G needle was used in these experiments) are slightly larger than those in Fig. 14 (20 G needle was used in these experiments).

Example 10. Formation of 2D interconnected hollow objects The formation of 2D interconnected hollow objects was further investigated by following the protocols in Example 6 (Fig. 17). We previously reported a characteristic of SA-EtOH, wherein it undergoes a sol-to-gel transition at low temperatures (ethanol concentration range: 25% - 40% v/v), as shown in Fig. 18. In brief, the formation of hydrogen bonds between ethanol and water could be enhanced by reducing temperature. It causes a decrease in the hydrogen bonds between water and SA chains. That means the water molecules between SA chains decrease. As a result, SA chains are pushed together, and phase separation occurs. Leveraging this property, we used molds to obtain connected tree-like structures. The preparation process for the 2D interconnected tree-like structures is illustrated in Fig. 17. After obtaining the samples, they were immersed in a Ca&2 ethanol solution reaching its boiling point for the hollowing reaction, following the same principle as the formation of 1 D tubes. Ultimately, we could obtain connected hollow tree-like structures.

We used molds to prepare a series of 2D structures to validate the hollowness of complex tree-like constructs, as shown in Fig. 17A. After heating in a CaCh ethanol solution, those constructs were immersed in DI water to remove ethanol and recover. Because these constructs were free of defects, we cut the ends of these objects and injected dyed water from one end. Fig. 17B reveals that all structures demonstrated interconnected hollow channels. It indicates that the simple and effective method for preparing interconnected hollow hydrogel structures exhibits tremendous potential for various applications.

Example 11. 3D printing with SA-EtOH precursor

The 3D printing setups consist of an extrusion-based commercial 3D printer and a cold plate, as depicted in Fig. 19C. The SA-EtOH precursor (prepared in Example 3) was extruded through a nozzle onto the cold plate at room temperature, and within a few seconds of contact with the cold plate, SA-EtOH undergoes a sol-to-gel transition, forming a gel that allows the printing process to proceed smoothly. After printing, the printed scaffold was moved into a boiling CaCIz ethanol solution to make it hollow.

Preparation of 3 mL sodium-ethanol 3D printing ink Materials:

Sodium alginate (SA)

De-ionized (DI) water

Pure ethanol (EtOH) (> 99.5%) Procedures:

1. Prepare 50% v/v EtOH solution from pure EtOH solution

2. Prepare SA stock solution (10% w/v) from SA powder

3. Prepare SA-EtOH ink (SA: 3.33% w/v; EtOH: 33.3% v/v)

As will be appreciated, the printing ink prepared above may be used for making 1 D and 2D structures by using a syringe pump and a physical mould, respectively.

Rheological characterization

Discovery Hybrid Rheometer HR-2 tested oscillatory frequency sweeps, oscillatory amplitude sweeps, and temperature sweeps. SA-EtOH samples were prepared to the same concentration, with 3.3% SA and 33.3% ethanol. For the oscillatory frequency sweep tests, the applied strain was 1%, while the shear frequency varied from 0.1 to 100 rad/s. In the oscillatory amplitude sweep tests, a frequency of 1 rad/s was used, and the deformation range was set from 0.1 to 100%. The temperature ramp tests were carried out at an angular frequency of 10 rad/s, and the temperature rate was set at 6 °C/min.

Tensile and compression tests

All the tests were performed using an Instron 3366 Tensile Tester (USA). The tensile and compression samples were both in the form of tubular specimens. The displacement rate was set at 2 mm/min. All samples were prepared using cylindrical molds with an inner diameter of ~ 4.5mm. After gelation at lowered temperature, they were heated in an 86°C calcium chloride ethanol solution for 20 minutes. Upon completion of the heating process, the samples were transferred to DI water at 40°C and stirred until their shapes were fully restored. The tensile test was exclusively conducted on the axial tensile testing of the hollow tube samples, while compression tests included axial compression, lateral compression, and compression of sealed hollow tubes (containing water). Except for the compression test of the sealed hollow tube, the ends of the other samples were cut before testing, removing the free water in the middle.

Results and discussion

Similarly, we can leverage the low-temperature gelation property of SA-EtOH (prepared in Example 6) for extrusion-based 3D printing. Rheological results are vital for 3D printing, as shown in Figs. 19A and 19B. The moduli (both storage and loss moduli) and viscosity of SA- EtOH significantly increase at low temperatures. The small hump at high strain (10% - 100%) indicated SA-EtOH at 4 °C formed a cross-linked network³, as shown in Fig. 19A. Moreover, at 4 °C and 25 °C, SA-EtOH exhibits shear-thinning properties. All the results above allow filaments printed on the cold plate to maintain their shape without collapsing. Conversely, SA- EtOH extruded at room temperature has a lower viscosity, facilitating smooth printing without requiring excessive extrusion pressure. This also helps reduce the possibility of nozzle clogging.

Fig. 19D shows photos of a printed scaffold at different stages. After heating, the volume of the scaffold reduces because of the Ca-induced gelation. Hence, the scaffold is smaller than before heating. And there are some bubbles inside of the scaffold after heating. Then, we used a syringe with a needle to inject red dye inside the scaffold to prove its hollowness. During imaging, the red dye did not fully mix with the pre-existing DI in the scaffold. As a result, the distinctive layering, combined with the presence of bubbles, could provide evidence of a connected hollow scaffold.

Example 12. Solvent-polymer combinations

Preparation of solvent-polymer samples

Different solvent-polymer combinations were prepared by first mixing the polymer with the DI water at a 3.3% w/v concentration, followed by mixing the polymer solution with the solvent to a final concentration of 33.3% v/v.

Results and discussion

Table 2. Solvent, polysaccharide, synthetic polymer, and protein-based polymer combinations.

“+” means gelation was observed within 30 mins after sample preparation, means gelation was not observed even after several hours.

Based on the responses of the three polysaccharide polymers tested, we expect ethanol, 1- propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide should also work for all other polysaccharide-based hydrogel materials, including chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin and pullulan.

Similar to polysaccharide polymers, based on the response of gelatin, we expect methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide should also work for all other protein-based hydrogel materials, including albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein and zein.

Based on the two synthetic polymers tested, we expect that acetone should work for polymers with many side chains or functional groups that can contribute to hydrogen bonding, such as hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP), and polyurethane (PU).

Conclusion

In summary, the ethanol-induced gelation of SA hydrogel was systematically investigated. The sol-to-gel transition was observed when the temperature was below a critical temperature and confirmed by both the change in rheological properties and the porous network structures using SEM. The transition temperature positively correlates with the ethanol and SA concentrations. Based on our SAXS experiments and all-atom MD simulations, the possible mechanism underlying the observed ethanol-induced gelation of SA is illustrated in Fig. 20. Without ethanol, the hydrophilic nature of the alginate chains makes alginates highly water- soluble, forming hydrogen bonds with water molecules. Ethanol molecules tend to destroy the hydrogen bonding network formed among alginate chains and water molecules and lead to local dehydration, effectively reducing inter-molecular separation distance among alginate chains. Consequently, alginate chains get close enough to each other to form a relatively more robust hydrogen bonding network, especially for the G blocks of alginate. Gelation becomes possible when the temperature is reduced to make the ethanol-induced enthalpic change outweigh the entropy change associated with the transition from a disordered liquid phase to a more ordered gel phase. However, too much ethanol (> 40% v/v) forces alginate chains to form large aggregates, causing macroscopic phase separation. Since ethanol-induced gelation and change in the viscosity of SA are achieved by altering the hydrogen bonding interactions among SA chains and water, we expect that our method may also be applicable to other gel systems. We successfully generated desired alginate hydrogel structures by directly printing the SA-ethanol mixture onto a cooling stage whose temperature was maintained below the sol-to-gel transition temperature of the SA-ethanol mixture. Since ethanol-induced gelation and change in the viscosity of SA are achieved by altering the hydrogen bonding interactions among SA chains and water, we expect that our method may be also applicable for other hydrogel systems. Therefore, the present discosure provides an effective approach for achieving direct printing of hydrogel and may largely benefit hydrogelbased applications in many fields.

Compared to the conventional approach of viscosity enhancement derived from yield stress fluids, our method avoids excessively raising the materials’ viscosity during printing, thus effectively mitigating potential extrusion issues, such as high-pressure requirements and nozzle clogging. The gelation upon cooling ensures the stability of the printed structure. Additionally, the significant reduction in the size of the printed structure as a result of removing ethanol after fixing the structure with Ca²⁺-induced gelation offers a means to enhance printing resolution. Therefore, our study provides an effective approach for achieving direct printing of hydrogel and may largely benefit hydrogel-based applications in many fields.

In addition, we have developed a simple method to obtain defect-free hollow hydrogels. The formation of a hollow hydrogel structure can be attributed to the result of two competing processes: the external factor (the calcium-induced gelation of SA) and the internal factor (the gasification of ethanol). When the SA-EtOH precursor solution is injected into the calcium solution, the SA chains start to cross-link immediately on the exposed outer surface, forming a cross-linked gel shell. The shell can help maintain the shape and, meanwhile, hinder the outflow of free alginate solution. Meanwhile, ethanol inside SA-EtOH gradually gasifies, forming gas bubbles with a temperature increase. Small and isolated gas bubbles continuously grow and then merge, leading to a continuous hollow hydrogel structure. We have proved that we could control the wall thickness of the hollow tubs by adjusting the parameters of the external or internal factors. This method enables the facile production of hollow hydrogel constructs without needing specific equipment or devices, such as coaxial nozzles. Due to the sol-to-gel transition of SA-EtOH at low temperatures, we could create more complicated constructs by casting and 3D printing.

Claims

(b) using the additive manufacturing device to print the ink formulation onto a substrate to generate a three-dimensional structure, wherein: the polymeric material is selected from one or more of the group selected from a polysaccharide, a protein and a synthetic polymeric material that has functional groups that can generate hydrogen bonds; the organic solvent is selected from one or more of the group consisting of an alcohol, acetone, dimethylformamide and dimethylsulfoxide; and in step (b) either the substrate is held at a temperature that is below a sol-to-gel transition temperature of the ink formulation, or in step (b) the printing step is conducted in an environment that has a temperature below the sol-to-gel transition temperature of the ink formulation, so as to enable the printed ink to undergo a sol-to-gel transition upon printing.

2. The method according to Claim 1 , wherein the organic solvent is selected from one or more of the group consisting of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol.

3. The method according to Claim 1 or Claim 2, wherein the polymeric material is selected from one or more of the group consisting of sodium alginate, K-carrageenan, chitosan, chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin, pullulan, poly(vinyl alcohol) (PVA), hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP), polyurethane (PU), gelatin, albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein and zein, optionally wherein the polymeric material is sodium alginate.

4. The method according to any one of the preceding claims, wherein when: the polymeric material is sodium alginate, then the organic solvent is selected from one or more of the group of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol; the polymeric material is K-carrageenan, chitosan, chondroitin sulfate, dextran, guar gum, gum arabic, hyaluronic acid, lignin or pullulan, then the organic solvent is selected from one or more of the group of ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol; the polymeric material is poly(vinyl alcohol) (PVA), hydroxypropyl methylcellulose (HPMC), poly(acrylic acid) (PAA), poly(glycolic acid) (PGA), poly(N-vinylpyrrolidone) (PVP) or polyurethane (PU), then the organic solvent is acetone; and the polymeric material is gelatin, albumin, casein, collagen, eastin-like polypeptide, fibrin, keratin, resilin, silk fibroin, soy protein, whey protein or zein, then the organic solvent is selected from one or more of the group of methanol, ethanol, 1-propanol, 2-propanol, acetone, dimethylformamide and dimethylsulfoxide, such as ethanol.

5. The method according to any one of the preceding claims, wherein the ink formulation is one that has: from 2 to 10% w/v of the polymeric material; from 25 to 40% v/v of the organic solvent; and from 60 to 75% v/v of water.

6. The method according to Claim 5, wherein the polymeric material is present in an amount of from 2 to 4% w/v of the ink formulation, such as about 3.3% w/v.

7. The method according to Claim 5 or Claim 6, wherein the organic solvent is present in an amount of from 30 to 35% v/v of the ink formulation, such as about 33.3% v/v.

8. The method according to any one of the preceding claims, wherein the temperature of the substrate or environment in step (b) is less than or equal to 15 °C and greater than or equal to -25 °C, such as less than or equal to 5 °C and greater than or equal to -20 °C (e.g. greater than or equal to -10 °C).

9. The method according to any one of the preceding claims, wherein: the ink formulation contains sodium alginate at a concentration of 3.3% w/v, ethanol at a concentration of 33.3% v/v and water in an amount of 66.7% v/v; and the temperature of the substrate or environment in step (b) is less than or equal to 5 °C.

10. The method according to any one of the preceding claims, wherein the method further comprises:

11. The method according to Claim 10, wherein:

12. The method according to any one of the preceding claims, wherein the method further comprises:

13. The method according to any one of Claims 10, 11 and 12, the latter as dependent upon Claim 10 or Claim 11, wherein the polymeric material is sodium alginate and the crosslinking agent is a calcium salt (e.g. CaCh), optionally wherein the concentration of the calcium salt is from 0.05 to 0.2 M, such as about 0.1 M in the gelation bath, further optionally wherein the organic solvent is ethanol.

15. The method according to Claim 14, wherein the polymeric material is sodium alginate and the crosslinking agent is a calcium salt (e.g. CaCh), optionally wherein the concentration of the calcium salt is from 0.05 to 0.2 M, such as about 0.1 M in the gelation bath, further optionally wherein the organic solvent is ethanol.