Zephyrnet Logo

Feedback-controlled hydrogels with homeostatic oscillations and dissipative signal transduction

Date:

Materials

N‐Isopropylacrylamide (NIPAm, 99.0%, recrystallized from n‐hexane), poly(ethylene glycol) diacrylate (PEGDA, Mn = 10,000), N,N′‐methylenebisacrylamide (BIS, ≥99.5%), acrylamide (AAm, ≥99.0%), agarose (ultralow gelling temperature, A5030), gold(iii) chloride trihydrate (HAuCl4·3H2O, >99.9%), sodium citrate tribasic dihydrate (BioUltra, 99.5%), poly(ethylene glycol) methyl ether thiol (PEG‐SH, Mn = 2,000), 3‐(trimethoxysilyl)propyl acrylate (≥92.0%), 2‐hydroxy‐4′‐(2‐hydroxyethoxy)‐2‐methylpropiophenone (photoinitiator, Irgacure 2959, 98.0%) and 2,2-dimethoxy-2-phenylacetophenone were purchased from Sigma‐Aldrich. 1,4-Bis-[4-(6-acryloyloxyhexyloxy)benzoyloxy]-2-methylbenzene was purchased from SYNTHON Chemicals. 6-Amino-1-hexanol and dodecylamine were purchased from TCI. Sodium hydroxide (97%) and hydrochloric acid (1 M) were purchased from Fisher Scientific. Ethanol (99.5%) was purchased from Altia Oyj. Deionized water (18.2 MΩ; Direct-Q 3 ultraviolet (UV), Millipore) was used in all the experiments.

Synthesis and modification of AuNPs

Citrate‐stabilized AuNPs were prepared by the classical citrate reduction of gold salt in water40. Briefly, trisodium citrate solution (2 ml, 1.00 wt%) was quickly injected in a boiling aqueous solution of HAuCl4·3H2O (100 ml, 0.01 wt%) under vigorous stirring. The solution was further refluxed for 10 min under stirring to complete the reaction. The AuNPs were analysed by transmission electron microscopy (Tecnai 12), which showed an average diameter of 20.0 ± 2.3 nm (Supplementary Fig. 2). The 20 nm AuNPs were stabilized using PEG‐SH by adding an ethanolic PEG‐SH solution (8 ml, 5 mg ml−1) to the AuNP solution (85 ml), which was incubated overnight on an orbital shaker. Finally, the modified AuNPs were purified three times by centrifugation (16,000×g for 25 min) and rediluted in pure water to yield a concentrated stock solution (0.88 ml) of PEGylated AuNPs. The concentration of AuNPs in this stock solution has been measured to be 4.3 mg ml−1 according to the dry weight, which is slightly lower than the theoretical value of 4.8 mg ml−1 due to loss during centrifugation. The concentration of AuNPs in the PAAm gel is, thus, 0.86 mg ml−1.

Preparation of gel oscillators

To prevent the undesired swelling and shrinking of hydrogels, the capillaries have been silanized to introduce covalent bonds with the hydrogels. Borosilicate glass tubes with a square cross section (2.0 mm × 2.0 mm (inner) and 2.8 mm × 2.8 mm (outer), VitroTubes) were cut into an appropriate length (~5 cm) with a glass cutter and cleaned by sonication in deionized water. The tubes were then activated by oxygen plasma for 5 min (Pico, Diener Electronic) and functionalized by storing the slides overnight in an evacuated desiccator containing 100 µl of 3‐(trimethoxysilyl)propyl methacrylate at 1 × 10−1 mbar. Subsequently, the liquid silane was removed, and the desiccator was further evacuated to 1 × 10−3 mbar for 2 h to remove any unbound silane on the glass surface. The silanized glass slides were stored in a sealed vial in the fridge and used within a week after preparation.

To prepare the gel, the desired amount of agarose was dissolved in deionized water by heating and vortexing until full dissolution to make a 1 wt% stock solution. Then, 0.25 ml of the hot agarose solution was added together with 0.20 ml water to dissolve 50.0 mg NIPAm, 1.0 mg photoinitiator Irgacure 2959 and 4.4 mg PEGDA crosslinker. The resulting solution, thus, contained 10.0 wt% NIPAm, 0.5 wt% agarose and 0.1 mol% PEGDA relative to NIPAm. The thoroughly mixed solution was then degassed by nitrogen bubbling for 5 min in a 40 °C water bath to prevent the gelation of agarose. The degassed solution was transferred to fill half of a silanized glass tube, which was sealed with Parafilm and kept in a nitrogenated vial. The tube was then stored in a fridge at 4 °C for 30 min for the gelation of agarose and then irradiated in a UV reactor (8 × 14 W lamps, 350 nm, Rayonet) for 20 min for the polymerization of NIPAm. Afterwards, the glass tube was filled with a degassed aqueous solution containing 10 wt% AAm, 1 mol% BIS and 1 mol% photoinitiator relative to AAm and AuNPs. The concentration of AuNPs corresponds to an optical density of 2 at 532 nm for an optical path of 2 mm (Supplementary Fig. 2), calibrated by a UV–visible spectrometer (Cary 5000, Agilent). Finally, the polymerization of AAm gel was carried out in a UV reactor (8 × 14 W lamps, 350 nm, Rayonet) for 20 min. The resulting hydrogel in the tube was purified by incubation in a water bath at 60 °C for 30 min and then kept overnight in pure water at room temperature. In this way, the agarose network was removed to form channelled PNIPAm. The tubes were stored in deionized water before use. Oscillators with other compositions were prepared following the same protocol.

Simulation of oscillator

The simulation of the oscillator was carried out in COMSOL Multiphysics 5.5 using a heat transfer module. The geometry parameters used to set up the three-dimensional model are summarized in Supplementary Table 1 and Supplementary Fig. 10. It is assumed that the two gels possess the same physical parameters (heat capacity, thermal conductivity and density) and the effect of the interface between the two gels on heat transfer is negligible. Therefore, a single piece of gel was created in the model with heating and transmission spots added. The physical properties of the materials are summarized in Supplementary Table 2. A time-dependent study was used to acquire the simulation data with a step length of 0.2 s .

Optical setup

A continuous laser beam (532 nm) was focused on the gel capillary surface with a plano-convex lens with 12.5 cm focal length. The transmitted beam through the channelled PNIPAm was reflected by an angle-adjustable mirror and projected on the PAAm. A linear translation stage was used to change the sample position for tuning the delay distance between the transmission and reflection (heating) spots. A marker was put on the PAAm side to ensure an identical heating position by the reflected beam with respect to the PNIPAm–PAAm interface. A metal block was attached to the PAAm side as the sample holder, as well as an effective heat sink to assist with heat dissipation near the heating spot. The oscillation requires sharp transition of the gel at the transmission spot and time delay between the heating and inhibition processes. Therefore, no oscillation was observed upon non-focused light through PNIPAm, direct photoheating of PNIPAm without delay or using conventional PNIPAm without nanochannels (Supplementary Figs. 46, control experiments).

Fabrication of LCE film actuator

The LCE actuator for the mechano-thermo-mechanical signal transduction was fabricated using a chain extension reaction41. Here 0.16 mol of 1,4-bis-[4-(6-acryloyloxyhexyloxy)benzoyloxy]-2-methylbenzene, 0.05 mol of 6-amino-1-hexanol and 0.05 mol of dodecylamine were mixed by magnetic stirring at 85 °C. Then, 2.5 wt% of initiator 2,2-dimethoxy-2-phenylacetophenone was added into the mixture. The mixture was infiltrated into to a cell at 85 °C via capillary force. The cell was prepared by gluing two coated glass substrates, one with a homeotropic alignment layer (JSR OPTMER, 4,000 rpm for 1 min, followed by baking at 100 °C for 10 min and 180 °C for 30 min) and the other with unidirectionally rubbed polyvinyl alcohol (5% water solution, 4,000 rpm for 1 min, baked at 100 °C for 10 min). Then, 50 µm microspheres (Thermo Scientific) were used as spacers to determine the film thickness. The cell was cooled down to 63 °C at 5 °C min−1 and kept in the oven for 24 h at 63 °C to allow the aza-Michael addition reaction for oligomerization (Supplementary Fig. 17). Then, the sample was irradiated with UV light (360 nm, 180 mW cm−2, 20 min) for polymerization. Finally, the cell was opened by a blade, and strips were cut from the film.

Fabrication of colour display

Thermochromic dyes (1 wt%) were mixed with polydimethylsiloxane precursor followed by drop casting in a Petri dish mould. The sample was thermally cured in an oven at 80 °C (24 h) to form an elastic film of about 0.5 mm thickness. The polydimethylsiloxane film was cut into 1 mm × 1 mm squares and used as thermochromic stickers. Two types of thermochromic powder pigment from Atlanta Chemical Engineering were used: TP-BP35 (transition from black to pink upon heating above 35 °C) and TP-RC45 (transition from red to white upon heating above 45 °C). Before attaching the thermochromic stickers, a kitchen aluminium foil (10 µm thickness) was cut and stuck onto the top surface of the gel tube, to block the scattered laser from the gel.

Fabrication of cargo transport system

Three LCE fins (6.00 mm × 1.50 mm × 0.05 mm) were vertically glued on top of a rectangular plastic sheet (4.00 mm × 2.00 mm × 0.10 mm) cut from an Optiazure transparency film. The plastic sheet was placed on top of the capillary tube with LCE fins near the heating spot. Two ends of the plastic sheet were fixed on the capillary surface by double-sided tapes. A U-shaped cargo was made by folding a piece of paper, which was placed on top of the LCE fins. The weight of the paper was 8.5 mg.

Fabrication of Mimosa- and flytrap-mimic assembly

The gel capillary was fixed on a mechanical stage that can provide a mechanical-trigger-induced displacement (0.3–0.5 mm) along the tube. The stage was connected to a spring, which allows it to return to the original position when the mechanical trigger is removed. For Mimosa-inspired gel–LCE assembly, an LCE strip was placed on top of the heating spot of the capillary. Due to heat-induced softening, the soft LCE spontaneously sticks to the capillary surface. A lightweight fibre is glued on the LCE strip, as an extended rod for better visualization of the bending angle. The LCE strip bends after sensing the heat conducted from the bottom gel capillary. For the flytrap-inspired gel–snapper assembly, a 100-µm-thick plastic strip was first glued with a glass sphere (2 mm diameter) at the centre position. The strip was supported by two objects (3 mm height) on both sides on top of the gel capillary. The sphere was then glued on top of the heating spot of the gel capillary via a liquid crystal ester as a temperature-sensitive adhesive. The plastic strip was pre-bent, and the release of elastic energy was induced by melting of the ester glue, which sharply occurs at 55 °C.

Optical characterization

Photographs and videos were recorded using a digital camera (Canon 5D Mark III, 100 mm lens), and the camera was equipped with an optical filter with a cut-off wavelength of <500 nm. Thermal images/videos were recorded with an infrared camera (FLIR T420BX, close-up lens with 50 µm resolution). The positions of the LCE strip, sample stage and snapper were tracked by a video analysis software (Kinovea, version 0.9.5).

Measurement of light transmission

A cover glass slide (0.2 mm thick) was placed between the mirror and gel tube to reflect about 5% of the total power of the light beam transmitted through the channelled PNIPAm gel. The reflected light was measured by a power meter (OP-2 VIS power sensor, Coherent; 1 Hz sampling rate) to determine light transmission (Itrans/I0) through the gel, where Itrans is the measured light power during temperature oscillation and I0 is the initial light power through the PNIPAm gel below its LCST.

spot_img

Latest Intelligence

spot_img