We used only commercially available materials in our experiments, following the thiol-acrylate Michael addition and photopolymerization approach. As shown in fig. S1, the reaction involves one reactive rod-shaped mesogen with dual acrylate end groups, code-named RM257 (Wilshire Technologies); one linear dithiol, HDT (Sigma-Aldrich); and one cross-shaped tetrathiol, PETMP (Sigma-Aldrich). The former thiol acted as a spacer, and the latter acted as a cross-linker for the initial sparse network. These components were dissolved in dichloromethane (Sigma-Aldrich) together with 10% Irgacure 819 photoinitiator (Ciba). The standard composition of LCE precursor solution had RM257, HDT, and PETMP in a molar ratio of 1:1:0.2. While this always resulted in ground-state negative order parameter shells, 〈P2〉 < 0, removal of the cross-shaped tetrathiol from the solution led to shells with positive order parameter, 〈P2〉 > 0. Depending on the targeted shell type, this mixture composition was adjusted, with varying amounts of linear HDT and cross-shaped PETMP. For confocal fluorescence imaging, we dissolved 3 weight % (wt %) dichroic azo dye named DDR536 (Color Synthesis Solutions) into the LCE precursor solution.

Following the principle of Utada et al. (33), nested square and cylindrical glass capillaries were used for shell production, as shown in fig. S2. The cylindrical injection tube of 1-mm outer diameter was tapered using a pipette puller (P-1000, Sutter Instruments) and cut to an orifice diameter of 90 μm using a microforge (Narishige). The collection tube, also cylindrical with a 1-mm outer diameter, was likewise tapered and cut to an orifice diameter of around 400 μm. Both tapered cylindrical capillaries were inserted into a square capillary of 1-mm inner side length, facing each other, thus creating a flow-focused junction. The interstitial space in the square capillary outside the cylindrical capillaries supported the outer and middle fluid flow, directed against each other. All three fluids met at the junction, in such a way that the middle fluid encapsulated the inner one, the former making a shell around the latter, suspended in the outer fluid. The resulting multiple emulsion was harvested through the collection tube (fig. S2A). The injection capillary was given a hydrophobic surface before assembly using an aqueous solution of 2 wt % fluorinated silane (nonafluorohexyl-triethoxysilane, Gelest). The square capillary and collection tube were treated with 2 wt % methoxytriethyle-neoxypropyltrimethoxysilane (SIM 6493.4, Gelest) in water for ensuring strongly hydrophilic surfaces. We used 5 wt % solutions of PVA (average molar mass, 13 to 18 kg/mol; 87% hydrolyzed) in a water-glycerol mixture of 1:1 volume ratio for inner and outer fluids. Various compositions of LCE precursor solution were used as the middle fluid.

To initiate the first step polymerization, we added 1 mM aqueous solutions of the catalyst triethylamine (Sigma-Aldrich) to the outer aqueous phase. Specifically, the outer phase was diluted by replacing 5 ml of it with the same amount of 1 mM aqueous triethylamine solution. This initiated the thiol acrylate reaction while the shell was undergoing osmosis at the same time. We repeated this step every 24 hours for 3 days. At this stage of the experiment, we had an isotropic sparse thiol-acrylate network throughout the shell. Further osmosis reduced the shell wall thickness and stretched the network in the shell plane, thus inducing negative order. The second-stage polymerization was carried out using a UV light-emitting diode (UVATA LED UV system; peak wavelength, 370 nm; full power, 8800 mW cm−2), kept at a distance of 10 cm above the sample vial. This step initiated the second polymerization of the unreacted excess acrylate groups; thus the configuration obtained through osmosis and solvent removal processes was locked in place. As the shell contained an unusually large quantity of photoinitiator from the start, extra care was necessary not to expose the shells to light with wavelengths shorter than 550 nm, as this would initiate the second-step photocrosslinking reaction prematurely. The reason for the unusually high concentration of photoinitiator is that much photoinitiator was lost during the osmosis and solvent removal stage, as were, to some extent, the two thiols. Therefore, these were also present at higher mole fractions than would normally be the case for this LCE chemistry.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.