Chemical-diffusion Metamaterials with “Plug and Switch” Modules for Ion Cloaking, Concentrating and Selection: Design and Experiments


 The outstanding abilities of metamaterials to manipulate physical fields have been extensively studied in wave-based fields. Recently, this research has been extended to diffusion fields. Chemical diffusion behavior is crucial in a wide range of fields including the transportation of various matters, and metamaterials with the ability to manipulate diffusion with practical applications associated with chemical and biochemical engineering have not yet been proposed. In this work, we propose the idea of a “plug and switch” metamaterial to achieve the switchable functions of ion cloaking, concentrating and selection in liquid solvents by plugging modularized functional units into a functional motherboard. The respective modules are theoretically designed based on scattering cancellation, and the properties are verified by both simulations and experiments. Plugging in any module barely affects the environmental diffusion field, but the module choice impacts different diffusion behaviors in the central region. Cloaking strictly hinds ion diffusion, and concentrating promotes a large diffusion flux, while cytomembrane-like ion selection permits the entrance of some ions but blocks others. In addition to property characterization, these functions are demonstrated in special applications. The concentrating function is experimentally verified by catalytic enhancement, and the ion selection function is verified by protein protection. This work not only demonstrates the effective manipulation of metamaterials in terms of chemical diffusion behavior but also shows that the "plug and switch" design is extensible and multifunctional, and facilitates novel applications including sustained drug release, catalytic enhancement, bioinspired cytomembranes, etc.


Introduatcon
Metamaterials have become the hottest topic in this century because of their exotic electromagnetic properties not found in naturally occurring materials. These properties originate from the outstanding ability of metamaterials to manipulate the physical fields. In particular, the idea of an invisibility cloak, which can render an object invisible to observers, has attracted significant attention. It controls the path of electromagnetic wave transport to achieve invisibility through an inhomogeneous medium artificially designed by the theories of transformation optics 1 (TO) by Pendry and scattering cancellation 2 (SC) by Engheta. In addition to the ideal cloaking proposed by theory and demonstrated by experiments, the design is further extended to metamaterials with diversified functions such as concentrators 3,4 , rotators 5,6 and illusion metamaterials 7,8 . After great success in manipulating electromagnetic waves [9][10][11]TO and SC were also proven to be powerful tools to manipulate not only other wave-based fields like acoustic waves [12][13][14] and water waves 15,16 but also diffusion fields such as thermal flux [17][18][19][20][21][22][23][24][25] , electric fields [26][27][28][29] , magnetic fields [30][31][32][33][34] , diffusive light [35][36][37] and chemical fields [38][39][40][41][42][43] . For comparison, metamaterials designed by TO usually exhibit substantial inhomogeneity and anisotropy, while SC provides a simpler way to achieve cloaking for diffusion fields by obeying the Laplacian equation, i.e., Laplacian fields.
As a fundamental phenomenon in nature, chemical diffusion is a spontaneous behavior that depends on the concentration gradient and follows the Laplacian equation, i.e., a typical Laplacian field. For years, researchers have devoted their efforts to seeking a more effective method to control the behavior of diffusion for applications in chemical and biological engineering. Although the idea of manipulating chemical flow by metamaterials has been proposed for years 40,41 , related research is still rare [38][39][40][41][42][43] . In particular, most studies are limited to theoretical design and numerical simulation rather than experiments, and those that do involve experimentation deal only with the tractable solid samples instead of the more general liquid environments. Additionally, the proposed chemical-diffusion metamaterials are limited to a single role, cloak or concentrator, and they have never been connected to practical applications. Compared to metamaterials used in electromagnetic fields and thermal fields that NATURE COMMUNICATIONS ARTICLE 4 have seen massive advances, chemical diffusion metamaterials are still prototypes and need great improvements.
As metamaterials are different from traditional materials, the key feature of cloaking and concentrating by metamaterials is to realize the minimum and maximum concentration gradient in a certain region while not altering the environmental concentration distribution. This feature is crucial to prevent specific objects from being detected through concentration variation in some chemical and biological applications. To achieve cloaking and concentrating, tailored anisotropic parameters (specifically, the diffusion coefficient) are required, which are proposed to be resolved by various passive and active components. Although the solid host is much easier to model and simulate, the extremely low diffusion efficiency is not appropriate for practical applications of chemical engineering and bioengineering, including ion separation, bioinspired devices, and drug delivery. In this work, we design chemical diffusion metamaterials with different functions for manipulating ion diffusion in liquids, including a bilayer cloak based on the SC theory, a concentrator with a fan-shaped structure and an ion selector. In addition, we propose the idea of "plug and switch" in a metamaterial device, in which different functions are switchable by plugging modularized functional units into the same motherboard without affecting the environmental concentration, as illustrated in Fig. 1. Furthermore, the designed functions are experimentally demonstrated in special applications of catalytic enhancement and protein protection that will benefit catalytic engineering and bioengineering.

Rmsu ts
The experimental setup contains a functional motherboard with optimized diffusivity and some pluggable modules to switch the function of metamaterials, as shown in Fig. 1. In this work, Modules 1, 2, and 3 represent the cloak, concentrator and ion selector, respectively. They have identical geometric configurations of a 3 cm annulus with a 1 cm concentric annulus and can be plugged into the center of the same motherboard. The proposed idea of "plug and switch" with modularization design can achieve minimum perturbation to the concentration distribution in the motherboard after plugging in any module and switch the device to functions such as cloaking, concentrating, ion selection and other possible applications.

NATURE COMMUNICATIONS
First, we design a bilayer cloak based on SC theory rather than TO theory to avoid complex anisotropic parameters. A diffusion process without convection can be expressed according to Fick's second law: ( ) Similarly, we have the concentration distribution for the bilayer cloak: For the two-dimensional case (corresponding to the bilayer annulus): It is concluded that a bilayer cloak requires an inner layer with an index of zero and a complementary outer layer with an index associated with the background media according to For a 2D chemical diffusion cloak, we use a 3D printed resin annulus (functional unit 1) to realize the zero index, which leaves only the diffusivities of the outer layer of cloak and the background media to be engineered. Then, we use the effective medium theory (EMT) to match the diffusivity of the outer layer of the cloak with that of the background media. The radii of the inner layer and the outer layer are set to 2 cm and 3 cm, respectively. The outer layer is pure water, and the internal layer is a 2 mm annulus made of resin. Therefore, the diffusivity ratio of the outer layer of the cloak and the background media is calculated as 13/5 according to Equation (5). Similar to the calculations in thermal physics, we expand the Bruggeman approximation of EMT to acquire specific diffusivity. For a two-phase system composed of water and homogeneously distributed resin pillars, the effective diffusivity is obtained as: ( ) we focus on the concentrating feature more than the invisibility feature in the design. The fraction of resin is 20% in the whole volume of Module 2, so that the diffusivity of r D decreases to be approximately 80% of that in pure water and D   to be 0. The simulated concentration distribution for the metamaterial device with the concentrator module demonstrates that although the parameters do not strictly obey the formula, the scattering is significantly reduced compared to the blank reference, and a significantly concentrated gradient is developed in the central region, as shown in Fig. 2f. The measured concentration along the line at y= -3.5 cm in Fig. 2e is almost identical to the simulated results.
Previous studies were always limited to model design and property characterization but ignored connecting the obtained function to practical applications. Here, we use the functions NATURE COMMUNICATIONS ARTICLE 9 of cloaking and concentrating in some practical applications, i.e., catalytic boost and protein protection. The concentrating feature of Module 2 significantly increases the concentration gradient in the central region, which can be used to promote catalytic efficiency. We select the typical Fenton reagent as an example, which is widely used in chemical engineering to remove refractory organic pollutants. As a strong oxidation system composed of Fe 2+ and H2O2, the Fenton reagent generates hydroxyl radicals and superoxide radicals with extremely strong oxidization through a chain reaction:  Fig. 3a. Here, the above chain reaction is simply treated as the disproportionation of hydrogen peroxide. We plug Module 2 into the motherboard and inject FeSO4 solution and deionized water into the two tanks. The concentrations in the central region with and without Module 2 are measured, and a 150% increase in the concentration of Fe 2+ is found at t = 120 s. We take the solution inside and outside the concentrator every 30 s to evaluate the concentration change with time, as shown in Fig. 3b. At every measured timepoint, the concentration inside the concentrator is obviously higher than that outside. The reduction in concentration outside of the concentrator is because the solution concentration is more susceptible to being disturbed by sampling as it approaches equilibrium.
To evaluate the catalytic efficiency of the Fenton reaction in dye degradation, we add the Fe 2+ solution taken from inside the concentrator into the precursor solution of Fenton reagent with organic dye (Specimen 1) and add the solution from outside the concentrator to another precursor solution (Specimen 2). In addition, Specimen 3 was made by adding the deionized water to precursor solution as a blank control. The comparison of the degradation effect for different specimens is shown in Fig. 3c, and all of the results are shown in Supplementary Fig.   6. The blank control indicates that organic dye cannot be degraded by pure water. The organic dye is completely degraded after 6 min in Specimen 1, while it takes approximately 36 min to fully degrade in Specimen 2. This result indicates that the catalytic efficiency is 6 times higher when the metamaterial device is used. The experimental results prove that the concentrator greatly promotes catalytic applications, which can be further extended to built-in metamaterials for catalytic enhancement.
The previously discussed cloak and concentrator modules remarkably change the field distribution in the central region but do not disturb the surrounding field, so they do not destroy the environmental field distribution and avoid detection through field fluctuations when they work. However, the process becomes much more complex for chemical diffusion metamaterials because diffusion may involve a variety of ions. This is a problem not faced by electromagnetic or other metamaterials. Similar to leukocytes, which detect viruses or bacteria through antigenic determinants, chemical diffusion metamaterials have the important ability to screen and selectively manipulate different ions. Here, we design the third module for our metamaterial device, a bioinspired cytomembrane-like ion selector. It enables specific ion penetration or shielding and maintains acid-base equilibrium.
Module 3 has a fan structure filled with an ion exchange resin. The ion-exchange resin contains sulfonic acid groups, carboxyl groups and phenol groups that exchange with high valent cations. As a result, the high-valent cations of Cu 2+ are shielded out of the central region, but low-valent ions such as K+ and Na + are still permitted. We demonstrate the design of a potassium ion channel. When the metamaterial device with an ion selection module is in the CuSO4/K2SO4 gradient, the cation exchange resin preferentially reacts with cupric cations and allows potassium ions to achieve the separation of CuSO4/K2SO4. The measured concentration inside and outside the module at 120 s is shown in Fig. 3d. The ion selector limits nearly all cupric cations while promoting the concentration of potassium ions. The concentration distribution along line y=-3.5 cm is also measured and shown in Supplementary Fig. 7. The overall concentration is lower than that of the cloak, concentrator and reference since the ion exchange resin absorbs major cupric cations.
As shown in Fig. 3e, when 100 g ion-exchange resin was mixed with 200 mL 13.2 g/L CuSO4 solution, the concentration approached a saturation of 8.6 g/L after 2 hours. This result indicates that the ion exchange resin acts as the active component to absorb the cupric cation.
In addition, the absorption speed is fast enough to shield the cupric cation in the CuSO4 gradient.
Since a certain cation is selectively shielded, the metamaterial device with Module 3 can be used as a cytomembrane-like ion selector to prevent chemical harm in bioengineering. Protein protection is selected as a special application to demonstrate the function in the experiment. As shown in Fig. 3f, a dialysis bag sealed with bovine serum albumin (BSA) solution was placed in the center of the motherboard in a CuSO4 gradient. BSA has a molecular weight of 66.5 kDa, while that of the dialysis bag is 12 kDa, which means that CuSO4 can penetrate the dialysis bag and cause the denaturation of BSA. If Module 3 is absent, BSA is exposed to CuSO4 and aggregates because of the conformational change and the loss of solubility. If Module 3 is plugged in, the BSA solution is unaltered after the same amount of time, demonstrating an excellent protein protection performance. Moreover, Module 3 can also be applied to anion shielding by filling the apparatus with anion exchange resin.
Note that although Module 1 of the cloak can also accomplish ion shielding, the strategy is quite different. The ion selector is an open system that selectively blocks some specific ions so that it provides more flexibility in complicated scenarios. Additionally, the ion selection function is limited by the saturation of the cation exchange resin. As more cation exchange resin is used, the length of time that the high-valent cations are blocked increases, but the effective diffusivity along the radial direction is smaller for other chemicals. This is because the ion exchange resin does not contribute to the diffusion of chemicals.

Dcsausscon
In this paper, we proposed a design paradigm for a "plug and switch" metamaterial with

Experimental setup of cloak and concentrator:
The experimental equipment has two tanks at left and right end with CuSO4 solution and deionized water respectively to produce a certain concentration gradient, and then a diffusion platform with engineered diffusivity is connected to both tanks through a port sealed by dialysis membrane, which is adopted to slow the diffusion at the port and prevent convection. After the liquid is added to both tanks and the diffusion platform between the tanks, the dialysis membrane is removed slowly.