A compact harmonic-suppressed Metasurface rectifier for efficient RF-to-DC power conversion in wireless power transfer systems Authors: Dunyu Chang, Chenchen Wang, Jinling Zhang, Xiongzhi Zhu, Shuobing Bai, Lifeng Duan Published: 2026 ABSTRACT SummaryThis study presents a compact harmonic-suppressed metasurface rectifier (HS-MSR) designed for efficient RF-to-DC power conversion in wireless power transfer (WPT) systems. By introducing four symmetrical T-slot resonant branches into the metasurface unit cell, high-order harmonic components generated by the nonlinear rectifying diode are effectively suppressed, improving conversion efficiency without additional filters. The metasurface employs a single-layer, fully integrated structure that allows direct impedance matching between the rectifier and the receiving surface, reducing circuit complexity and insertion loss. Both full-wave and co-simulation results verify an RF-to-DC efficiency of 75.6% under arbitrary polarization and wide incidence angles (0°–60°). Experimental validation with 2 × 2 and 4 × 4 prototypes further confirm the scalability and robustness of the proposed design. The HS-MSR provides a promising platform for system-level RF energy harvesting and high-efficiency WPT applications in compact electronic and internet of things (IoT) devices.Subject areas: Applied sciences, Engineering, Materials science Introduction Wireless power transfer (WPT) has attracted increasing attention in power electronics and energy systems, offering a promising solution for supplying energy to distributed and battery-free devices. In far-field WPT systems, the rectifying receiver plays a critical role in converting incident radio frequency (RF) energy into usable direct current (DC) power. The overall performance of such systems strongly depends on the RF-to-DC conversion efficiency of the rectifier, which is affected by impedance mismatch, harmonic re-radiation, and circuit complexity. Therefore, the design of compact, high-efficiency rectifiers with wide angular and polarization stability remains a key challenge in power electronics based wireless power conversion systems.1,2,3 The key performance metric to evaluate rectifying receiver is efficiencyRF-DC (i.e., radiation to DC efficiency), which combines the efficiencyRF-AC (i.e., radiation to AC efficiency), efficiencyAC-DC (i.e., rectification efficiency), and power transfer coefficient (k).4 Conventional rectifying antennas (rectennas) and metasurface-based energy harvesters have demonstrated remarkable potential in improving electromagnetic absorption and miniaturization.5,6,7 Metasurfaces have been proposed as efficient receivers for RF energy harvesting, achieving efficiencyRF-AC values exceeding 95%.8,9 Various types of metasurface receivers with loaded resistors have been proposed, mainly focusing on absorption characteristics and bandwidth enhancement. However, these designs primarily focused on electromagnetic absorption characteristics without addressing efficient power transfer from the metasurface to the rectifier.10,11,12,13 Some studies proposed metasurface receivers based on electric ring resonator, bow-tie dipoles, and slot patch.14,15,16 The integrated feed network guided the collected AC energy into a single rectifier circuit, thus achieving the efficiencyRF-DC of 40%,60%, and 72%. Nevertheless, most existing designs rely on multi-layer configurations or external matching networks, which lead to additional losses and limit system integration. Several efforts have been devoted to improving the robustness and angular tolerance of rectifying receivers. A cross-dipole-based design achieved 74% RF-to-DC efficiency with complete polarization independence.17 Similarly, a UC-PBG-based metasurface realized dual-band operation (2.45/5.8 GHz) with wide incident-angle response and efficiencies of 70% and 59%, respectively.18 These studies demonstrate that properly engineered metasurface symmetry can significantly enhance polarization and angular robustness in wireless power conversion systems. Moreover, the high-order harmonics generated by nonlinear Schottky diodes not only degrade RF-to-DC conversion efficiency but also induce unwanted re-radiation that interferes with adjacent circuits.19,20 One solution to suppress harmonics is to add a low-pass filter between the antenna and the rectifier,21 the other is to design a defective grounding structure or additional open circuit branch.22,23 Due to the high integration and compactness of the metasurface, these two harmonic suppression methods are not applicable. Hence, there remains a need for a new receiver architecture capable of efficiently harvesting energy from arbitrarily polarized and obliquely incident waves while mitigating higher-order harmonics. Consequently, it is highly desirable to realize a rectifying receiver that inherently combines harmonic suppression and impedance self-matching, enabling a compact, low-loss architecture with enhanced robustness and system-level energy conversion efficiency. In this study, a compact harmonic-suppressed metasurface rectifier (HS-MSR) is proposed for efficient RF-to-DC power conversion in WPT systems. The HS-MSR introduces four symmetrical T slot resonant branches into each metasurface unit cell to suppress high-order harmonics while maintaining polarization independence. The single-layer structure achieves direct impedance matching with the rectifier diode, resulting in minimized losses and simplified circuit topology. Both simulation and experimental results demonstrate an RF-to-DC efficiency exceeding 75.6% under wide incident angles and arbitrary polarization conditions. The main contributions of this work are summarized as follows: (1)A compact metasurface rectifier integrating harmonic suppression and rectification functions within a single layer, eliminating external filters and matching networks; A compact metasurface rectifier integrating harmonic suppression and rectification functions within a single layer, eliminating external filters and matching networks; (2)A system-level energy conversion model and experimental verification achieving high RF-to-DC efficiency with angular and polarization robustness; A system-level energy conversion model and experimental verification achieving high RF-to-DC efficiency with angular and polarization robustness; (3)A scalable design approach suitable for high-efficiency, low-profile wireless power conversion applications. A scalable design approach suitable for high-efficiency, low-profile wireless power conversion applications. Results and discussion Design of the metasurface structure Figure 1 shows the proposed metasurface cell structure. The selected operating frequency is 5.8 GHz, and the dielectric substrate is Rogers 5880. The dielectric substrate was chosen for its low loss, allowing the absorbed energy to go into the rectifier diode instead of being consumed in the dielectric plate. The energy harvester is a single-layer structure with a thickness of 1.56 mm, where the media thickness H = 1.524 mm and the metal layer thickness T = 35 μm. The basic structure of the metasurface unit is a circular patch with a radius of R1 = 4.16 mm. Four arrowhead shaped branches are loaded diagonally to extend the bandwidth, where the arrow spacing is L1 = 0.8 mm, the arrow bottom width is L2 = 1.5 mm, and the arrow length is R2 = 5 mm. The 5.8 GHz transmission characteristics, such as resonant frequency, bandwidth, and depth, can be adjusted by changing the shape of the arrow. Four rectangular microstrip lines are loaded on the axis to connect the diodes between the cells. The width of the microstrip line is W = 0.8 mm and the length is J = 1.36 mm. The impedance at the gap between the metasurface cells can be changed by adjusting the width and length of the rectangular microstrip lines, thus eliminating the matching circuit between the metasurface and the lode. A set of centrally symmetric T slots are etched at the center of the patch, where the transverse length and width of the T slots are T1 = 1.84 mm, M1 = 0.72 mm, and the longitudinal length and width are T2 = 2.67 mm, M2 = 0.1 mm. The transmission characteristics at higher harmonics, such as suppression frequency, bandwidth, and depth, can be adjusted by the transverse and longitudinal lengths of the T slots. The metasurface array is spaced with metallized holes to connect to the ground. The metallized hole is located in the center of the unit with diameter D = 0.6 mm. Structural topology of HS-MSR Circuit analog absorber approach has been widely used to explain the absorption mechanisms of periodic absorber metasurfaces.24,25 As shown in Figure 2, the transmission line equivalent circuit is used to simulate the interaction of a plane wave with a periodic absorbing metasurface. In the circuit model, Y0 denotes the free-space conductance, Y0 = 1/377 S. Ys is the characteristic conductance of the substrate equivalent transmission line. Rogers5880 substrate corresponds to d = 1.524 mm and dielectric constant εr = 2.2, YS = √εr/377 = 1/254 S. As shown in Figure 2, the equivalent circuit of the proposed HS-MSR cell is modeled with lumped elements. Where R is the lumped resistance placed at the cell terminal, C1 is the capacitance between the two arms of the metasurface cell, C2 is the capacitance introduced by the etching gap, C3 is the capacitance introduced by considering adjacent cells, and L is the total inductance of the metasurface cell.26 The reflection coefficient of the equivalent circuit model is Γ(w), which can be written as follows: Where, Y0, YA, and Ys are free-space admittance, metasurface equivalent admittance and substrate transmission line admittance. Figure 3 compares the S parameters of the absorbing metasurface simulated by computer simulation technology (CST) and the equivalent circuit simulated by advanced design system (ADS), and they are consistent. Schematic showing an equivalent circuit model of the proposed HS-MSR The parameters of the circuit are C1 = 0.41 pF, C2 = 0.02 pF, C3 = 0.13 pF, L = 7.5 nH, R = 169 Ω. (A) Top-view geometry, (B) Side-view configuration, (C) Equivalent circuit model. Reflection coefficient of the absorbing metasurface simulated by CST and the equivalent circuit simulated by ADS Absorptivity, efficiency and harmonic suppression characteristics of metasurface We further analyze the absorptivity and efficiency of the HS-MSR. The periodic boundary conditions are used to simulate the infinite metasurface in which the metasurface elements behave like those placed in an infinite plane. Floquet port is used to simulate plane wave incidence of 1 W. Electromagnetic waves with different polarization directions and incident directions are simulated by changing the polarization angle-(φ) and incident angle-(θ) of the port (Figure 1 for the definition of angle). The absorptivity reflects the ability of the HS-MSR to capture electromagnetic waves in a specific area. It can be calculated by A(w) = 1-|S11|2–|S21|,2 where S11 and S21 are the reflection and transmission coefficients of the Floquet port, respectively. The metasurface element is supported by the metallic ground, so the transmission coefficient is essentially zero. Considering that the energy absorbed by the metasurface may be dissipated in resistive loads or dielectric substrates,27 efficiency is used to demonstrate the ability of the metasurface to transfer electromagnetic waves in a specific area to the available power in the load (where the diodes are placed). It is defined as the conversion efficiency of radiation to AC, expressed as efficiencyRF-AC = PAC/Pin, where Pin is the total time-averaged power under the physical area of the metasurface. PAC is the time-averaged AC power available at the load. Figure 4 shows the absorptivity and efficiency of the metasurface element when the electromagnetic wave is incident vertically in the Z direction. The results show that the absorptivity of the metasurface element is 99%, 96.8% of which is dissipated in the load, and the other 2.2% is dissipated in the form of dielectric loss and conductor loss. Absorptivity, efficiency, reflection coefficient, and loss of HS-MSR As shown in Figure 5, we can observe the change of the electric field before and after entering the metasurface absorber. As shown in Figure 5A, the electric field intensity at the position λ/2 above the metasurface element is 1 × 102 V/m, and the electric field distribution of the incident field is uniform. As shown in Figure 5B, the electric field is redistributed at the position close to the metasurface cell (about λ/20). It is observed that the electric field mainly focuses on the position where the diode is placed, and the maximum electric field intensity is 6.0 × 103 V/m. Finally, Figure 5C shows the electric field distribution on the surface of metasurface. At this time, the electric field is almost all concentrated in the unit gap (where the diode is placed), and the intensity is 1.0 × 104 V/m. Field simulation results show that HS-MSR can capture the incident plane wave and concentrate almost all the energy in the diode. Simulated electric-field distributions reveal energy concentration in the HS-MSR cell (A) Simulated electric-field distribution of the HS-MSR cell at a distance of λ/2 above the cell. (B) Simulated electric-field distribution of the HS-MSR cell at a distance of λ/20 above the cell. (C) Simulated electric-field distribution on the surface of the metasurface. (D) Magnified view of the diode-mounted gap region, highlighting the localized electric-field concentration at the rectifying diode position. The simulation results verify the efficiency of HS-MSR in collecting different polarized electromagnetic waves at vertical incidence. Figure 6 shows that at different polarization angles, the metasurface can capture RF energy efficiently with a peak efficiency of more than 95%. It has polarization insensitivity. From Figure 7, it can be observed that the absorbed energy is mostly concentrated in the gap in the X direction when the polarization angle is 0°. When the polarization angle is 90°, the absorbed energy is mostly concentrated in the gap in the Y direction. When the polarization angle is between 0° and 90°, the absorbed energy is distributed in the gap in both X direction and Y direction. We also analyze the efficiency of the metasurface in capturing electromagnetic waves with different incidence angles under two polarizations. Figure 8A shows that under transverse electric (TE) polarization, the peak efficiency is greater than 90% when the incident angle ranges from 0° to 60°. But further increasing the incidence angle will reduce the absorption rate. Figure 8B shows that under transverse magnetic (TM) polarization, the peak efficiency is greater than 96% when the incident angle ranges from 0° to 60°, but the optimal efficiency point is offset. The reason is that when metasurface responds to electromagnetic wave, magnetic resonance is relatively difficult and resonant strength is weak. Notably, at oblique incident angle of 60° under TM polarization, a local dip around 6.2 GHz and a recovery near 6.4 GHz are observed, which are mainly caused by a rapid change in the metasurface’s equivalent impedance and phase response at large angles. This yields a coupling-mismatch condition at specific frequencies temporarily higher reflection and reduced absorption, followed by re-matching at higher frequencies. The optimal absorption point can shift away from 5.8 GHz and may even exceed the efficiency at 5.8 GHz under certain polarization conditions. For the electromagnetic wave under TE polarization, the Y-component of the incident magnetic field decreases as the incidence angle increases. Finally, it is difficult to maintain a certain strength of magnetic resonance. However, the electromagnetic wave under TM polarization can always produce effective magnetic response because its magnetic field direction remains unchanged.28 Therefore, HS-MSR can maintain a high absorption at a wide range of incidence angles under TE polarization and TM polarization. Efficiency of HS-MSR varies with frequency at different polarization angles at vertical incidence When the incident wave is irradiated vertically, the HS-MSR achieves an absorption maximum at 5.8 GHz. Simulation of electric field strength of HS-MSR at different polarization angles. The highest electric field strength (red) corresponds to 1.6 × 104 V/m, the lowest (blue) corresponds to 0 V/m (A) φ = 0°. (B) φ = 30°. (C) φ = 60°. (D) φ = 90°. Efficiency of simulated HS-MSR varies with frequency at different incident angles (A) TE polarization. (B) TM polarization. For the metasurface rectifier, high-order harmonic suppression allows the energy to be returned to the diode and rectified again, thus improving system efficiency. We propose to load T slots on the metasurface to achieve harmonic suppression in the system, which does not require additional branches and harmonic suppression networks, resulting in a simple and compact structure. To illustrate the role of the T slot structure at harmonic suppression, we show the surface current distribution of MSR and HS-MSR at 5.8 GHz, 11.6 GHz, and 17.4 GHz. As shown in Figure 9, both metasurface can radiate normally at 5.8 GHz. However, the surface current of HS-MSR is mainly concentrated near the T slot at 11.6 GHz and 17.4 GHz, which results in the higher harmonics are suppressed. Figure 10 shows the S parameter of the metasurface unit with or without T slot structure. The S parameter is seen from the position of the diode. The reflection coefficient of HS-MSR at higher harmonics is better than −2 dB with the help of T-slot. The surface current distribution of MSR and HS-MSR at 5.8 GHz, 11.6 GHz and 17.4 GHz (A) HS-MSR. (B) MSR. S parameter of HS-MSR and MSR (viewed from the position where the diode is placed) (A) MSR (without harmonic suppression structure). (B)HS-MSR (with harmonic suppression structure). Impedance matching and equivalent DC channel circuit design of the rectifier Unlike traditional rectenna designs, the proposed HS-MSR does not employ a matching network between the diode and the antenna. In contrast, the impedance of the metasurface can be directly matched to the input impedance of the rectifier, so the Schottky diode can be installed in the internal metasurface gap. The HSMS-2860 Schottky diode was selected as the rectifying element because of its low on-off voltage of 0.28 V and fast switching speed.29 The simulator ADS is used to analyze the input impedance of the rectifier. Due to the nonlinear characteristics of the diode, the input impedance of the rectifier varies with the input power as shown in Figure 11A. Under the condition of optimal rectification efficiency, when the input power is 11 dBm, the input impedance of the rectifier Zd = 105-j255Ω. As shown in Figure 11B, the impedance at the metasurface element gap is obtained by using the full-wave simulator CST. The impedance at the feed gap of the metasurface cell needs to be conjugate matched with the rectifier to achieve the maximum power transmission. The lumped port is placed at the two feed gaps with an impedance of Zd∗ = 105 + j255Ω, which is the conjugate impedance of the rectifier input impedance. Figure 12 shows the scattering parameters for each feed gap when the metasurface element is matched to the rectifier impedance, where the ports at the transverse and longitudinal gaps are defined as port-1 and port-2 (as show in Figure 1). The reflection coefficients S11 and S22 in the feed gap show that the metasurface element and the rectifier are completely impedance matched. Input impedance of HS-MSR and rectifier (A) Rectifier. (B) HS-MSR. Reflection coefficients of HS-MSR at transverse and longitudinal gap positions (the input power of the diode is 11dBm) To illustrate the DC path, Figure 13 shows the circuit model of a 16-cell metasurface array. The metasurface unit can capture the electromagnetic wave energy and transmit it to the feed gap, so an equivalent RF voltage source is formed in the feed gap. The voltage of the voltage source depends on the operating frequency and the angle and polarization of the incident wave. The diodes are connected in parallel to these gaps. The HS module is to show that the metasurface unit has harmonic suppression function. In general, the source acts as an “output port for the energy harvester” and is connected in parallel with the rectifier diode, while harmonic suppression is integrated into the system. The connection modes of the rectifier branch in the rectifier antenna array include series connection, parallel connection, and cascade connection.17,18,30 We constructed the entire HS-MSR array in parallel. As shown in Figure 13A, the HS-MSR array is spaced by cells with metallized holes and cells without metallized holes, where the cells without metal holes are connected to the positive electrode of the diode, and the cells with metallized holes are connected to the negative electrode of the diode. As shown in Figure 13B, all cells with metallized holes are connected to the ground, forming an equipotential surface-DC–. Other cells without through-holes are connected by microstrip lines to form an equipotential surface-DC+. Then all equivalent rectification branches pass through the equal potential surface DC+ into the DC filter, and output a smooth DC waveform. As shown in Figure 13C, the final array can be equivalent to N rectification branches in parallel, and N depends on the number of diodes in the array. Note that under the condition of plane wave irradiation, the metasurface element in the array can receive the same RF power, at which time the system can obtain the highest output power. However, if they are different, the output power will be reduced.31 DC system connection diagram of HS-MSR (A) System principle topology. (B) DC-path diagram. (C) Equivalent circuit diagram. We adopted CST and ADS co-simulation to simulate the common effect of field and road. Firstly, the Z parameter of metasurface cell in full-wave simulation are extracted and imported into ADS. The Z parameter is inserted into the data access component and then specified as the impedance of the source.32 Each metasurface element is modeled as an equivalent voltage source in the circuit. Figure 14 shows the simulation results of the circuit with a single voltage source. When the input power is 11.4 dBm and the load resistance is 933 Ω, the best rectifier efficiency is 73.7%. The scalable architecture of the proposed energy harvester is further demonstrated in Figure 15A, where simulations are performed for arrays N = 2, N = 4, and N = 16. Figure 15B shows that the scalability of HS-MSR enables it to obtain higher output power by increasing the size of the array. When the input power is 11 dBm, the system efficiency of HS-MSR with N = 4 and N = 16 is 72.4% and 76.6%. It is worth noting that the HSMS-2860 diode was selected for the high power WPT system applications. In practice, different diodes can be flexibly used depending on the power level of the WPT, such as HSMS-2850 for medium power and SMS-7630 for low power. In order to obtain the desired frequency and efficiency behavior after diode replacement, the element structure and size can be optimized. Simulation of a rectifier (A) EfficiencyAC-DC varies with frequency for different input power. (B) EfficiencyAC-DC varies with input power for different load. Simulation of HS-MSR with different number of elements (A) EfficiencyRF-DC and output voltage varies with frequency for different number of elements. (B) EfficiencyRF-DC and output voltage varies with input power for different number of elements. Setting and calculation method of the measurement system The measurement setup is shown in Figure 16. The whole experiment was performed in a microwave darkroom. The fabricated HS-MSR is placed at a distance of 1.2 m from the horn antenna and illuminated by a linearly polarized plane wave generated by the horn antenna. The feed-in power of the horn antenna is determined by the excitation source assembly. The excitation source components are a point frequency source (outputting a 5.8 GHz local oscillation signal), an adjustable attenuator (adjusting the input power level) and a power amplifier (amplifying the local oscillation signal). The point frequency source can be replaced with a signal generator to adjust the desired power level and frequency. Table 1 gives the basic parameters of the system link at 5.8 GHz. In addition, considering the effect of power level and frequency on the amplifier gain, the feed-in power of the horn antenna is measured by a power meter each time. When calculating the power provided by the horn antenna, the effect of the test frequency on the gain was also carefully considered. The voltage across the load resistor was measured with a multimeter. The physical picture of HS-RMS is shown in Figure 17. Measurement system (A) System setup. (B) Excitation source. Table 1Link parameters of the 5.8 GHz transmission systemTransmitter parametersFrequency (f)5.8 GHzSource out power (Ps)0-30 dBmAmplifier gain (Ga)10 dBMeasured cable loss (Lc)1.4 dBHorn antenna gain (Gt)20 dBmPropagation parametersDistance(D)1.2 mRF Power density at receiving surface (Sr)0-4 mW/cm2Metasurface parametersPhysical area of the 2 × 2 metasurface array5.66 cm2Physical area of the 4 × 4 metasurface array24.2 cm2 Link parameters of the 5.8 GHz transmission system 2 × 2 and 4 × 4 HS-MSR physical samples (A) 2 × 2 HS-MSR array. (B) 4 × 4 HS-MSR array. In this work, we follow the same procedures for efficiency measurements and calculations as those provided in previous publications.17,18,24,25 Where Pin is the total time-averaged power irradiated onto the metasurface, and Pout is the available time-averaged power received across the load resistance. Pin can be calculated using the Friis equation, and Pout can be calculated by measuring the voltage across the load resistor. Where Pt is the actual power fed to the horn antenna, Gt is the actual gain of the horn antenna, D is the distance between the horn antenna and HS-MSR, and Ar is the effective receiving aperture of the HS-MSR. According to previous reports,14,17,18 Ar can be expressed as the physical surface area of the HS-MSR. It should be noted that the effective aperture of an electric small antenna cell may be larger than its physical area. However, as the element becomes larger and larger, its physical size and effective aperture will be almost the same, which is different from conventional antenna array theory.8 VL is the voltage across the load, and RL is the load resistance value. The efficiencyRF-DC here can be seen as the product of three efficiencies. where ηRF-AC denotes the radiation-to-AC conversion efficiency, which shows the percentage of available radiated power captured by the HS-MSR at its effective area versus the incident power. The simulation results are expressed as efficiencies in Figure 4 ηM indicates the matching efficiency, which shows the actual AC power delivered to the diode as a percentage of the captured available AC power. A simulation of their matching performance is given in Figure 12 ηAC-DC indicates the AC to DC conversion efficiency, showing the percentage of available DC power at the load resistor versus the total AC power in the input diode. During the measurements, we focused on measuring the system efficiency, which is a measure of the overall HS-MSR performance and characterizes the ability of the HS-MSR to capture RF energy and convert it to DC. Measurement results and discussions of the fabricated HS-MSR First, the load impedance and input power density of the HS-MSR are scanned at 5.8 GHz. The purpose of the scan is to obtain the optimal DC load and power level to maximize the efficiencyRF-DC of the HS-MSR. The efficiencyRF-DC is then recorded at multiple frequency points as the HS-MSR is terminated by the optimal load. The power level is further adjusted to maximize efficiency. The efficiency profiles of the 2 × 2 and 4 × 4 finite arrays are shown in Figures 18 and 19. The measured results show that the 2 × 2 HS-MSR obtained an efficiencyRF-DC of 71.2% at an input power density of 0.075 mW/mm2 and a load resistance of 500 Ω. The 4 × 4 HS-MSR obtained an efficiencyRF-DC of 75.6% at an input power density of 0.075 mW/mm2 and a load resistance of 75 Ω. Measurements of the 2 × 2 metasurface array (A) Measured efficiencyRF-DC and output voltage varies with frequency at different input powers. (B) Measured efficiencyRF-DC and output voltage varies with input power at different loads. Measurements of the 4 × 4 metasurface array (A) Measured efficiencyRF-DC and output voltage varies with frequency at different input powers. (B) Measured efficiencyRF-DC and output voltage varies with input power at different loads. The sensitivity of HS-MSR to polarization angle and incidence angle was further measured. Here we take a 4 × 4 metasurface array as an example. Figure 20 shows that HS-MSR achieves almost the same system efficiency under the polarization angle of 0°–90°, indicating that HS-MSR has polarization insensitivity. Figure 21 shows the variation of HS-MSR system efficiency as a function of incident angle under TE polarization and TM polarization. When the E-field vector of the incident electromagnetic wave is parallel to the X axis, the optimal system efficiency is greater than 60% in the incidence angle range from 0° to 60°. When the E-field vector of the incident electromagnetic wave is parallel to the y axis, the peak efficiency is greater than 65% in the incidence angle range from 0° to 60°, and the optimal receiving frequency point is slightly offset. It is proved that HS-MSR has the characteristic of wide incidence angle. Measured efficiencyRF-DC of the 4 × 4 metasurface array (A) Measured efficiencyRF-DC varies with frequency at different polarization angles at vertical incidence. (B) Measured efficiencyRF-DC varies with input power at different loads. Measured efficiencyRF-DC varies with frequency at different incident angles (A) TE polarization. (B) TM polarization. Finally, we verify the effect of harmonic suppression on efficiency. Figure 22 shows the efficiency of the rectified metasurface with and without harmonic suppression. The results show that the harmonic suppression structure can improve the system efficiency by 2%–4%. ▪▪▪ Images show (A) measured and simulated efficiencyRF-DC and output voltage varies with input power at the optimal resistance of the 4 × 4 metasurface array. (B) Measured efficiencyRF-DC varies with input power at different temperature. Based on the measurement results, the manufactured HS-MSR is more sensitive to changes in frequency and power. This is because the design is based on an infinite periodic structure, while the finite array we fabricated leads to potentially slightly different results. The performance optimization can be solved by using a larger panel to minimize the effect of peripheral cells.17,18 Table 2 summarizes the recent work on rectifying antennas and provides the key metrics compared to our design. It should be noted that in some recent works,33,34,35,36 a new GaAs Schottky barrier diode (SBD) based bridge rectifier circuits was developed to obtain rectification efficiencies of more than 90%. Our goal from the beginning was to develop a scalable harmonic suppression metasurface antenna receiver that could accommodate different polarization and incidence angles. Therefore, we choose the commercial Si SBD that is consistent with other rectifying antennas.14,15,16,17,18 So the comparison of efficiency alone is not a fair assessment. Our novelty lies in the fact that we propose a method to suppress the higher harmonics in the metasurface antenna receiver. This is beneficial to improve the conversion efficiency of the system without adding additional complexity. Compared with the rectenna, proposed by Yang et al.19 and Sim et al.,23 HS-MSR has smaller size and higher integration. Research by Erkmen et al.25 and Ashoor et al.30 proposed a rectified metasurface based on FSS and dipoles, which can only capture a single polarized electromagnetic wave. Our proposed HS-MSR has characteristics of polarization independence and wide incidence angle, which solves the problem that the efficiency of WPT system is greatly reduced due to polarization or angle mismatch. The rectifier metasurface receivers proposed in some researches15,17,25,30 all require two dielectric layers and an air layer. Our proposed HS-MSR requires only one dielectric layer, which enables it to obtain the minimum profile thickness. Compared with the single-layer rectifying metasurface in Li et al.,18 we obtain a wider half-power bandwidth. Our proposed HS-MSR is a polarization-independent, wide-angle, and scalable high-efficiency energy harvesting system. Table 2Performance comparison between our work and other latest workReferencesFrequency (GHz)EFFRF-DCand input power (dBm)Cell size and number of array unitsMulti-polarization and wide angleHalf-power bandwidth (MHz)Harmonic suppressionNumber of layers and section thickness (mm)Yang et al.195.862% &180.8λ×0.8λ&5 × 5Yes/YesSingle substrate and 0.87 mmYang et al.235.861.4%&180.8λ×λ&5 × 5Yes/YesSingle substrate and 0.87 mmErkmen et al.252.4561%& 150.3λ×0.5λ&5 × 6No20%NoSingle substrate + air layer&6.52 mmItoh and Sakai363.476% & 150.2λ×0.45λ&9 × 3No12.5%NoDouble substrate and 6.42 mmAshoor and Ramahi17374% & 160.2λ×0.2λ&7 × 7Yes10%NoDouble substrate + air layer&4.44 mmLi et al.182.45&5.858% (2.45 & 0)50% (5.8 & 0)0.3λ×0.3λ& 4 × 4Yes2.45/12%5.8/5%NoSingle substrate and 1.27 mmErkmen and Ramahi152.4561%& 150.5λ×0.6λ& 3 × 4No8%NoSingle substrate + air layer and 8.05 mmKikkawa et al.342.4591% & 37only rectifierNO/YesSingle substrate and 0.4 mmLiu et al.372.3–2.960.5% & 25208 mm × 208 mmφ (0–45°)θ (0–60°)23.1%YesDouble substrate + air layer&12 mmOur work5.871.2%& 11(2 × 2)75.6%& 11(4 × 4)0.2λ × 0.2λ& 2 × 2/4 × 4φ (0–90°)θ (0–60°)15%YesSingle substrate and 1.56 mmBold values are the results of the proposed HS-MSR in this work. Performance comparison between our work and other latest work Bold values are the results of the proposed HS-MSR in this work. This study demonstrates that integrating harmonic suppression and impedance matching directly into the metasurface topology can significantly improve the robustness and efficiency of WPT receivers. The proposed harmonic-suppressed metasurface rectifier achieves stable RF-to-DC efficiencies of 71.2% and 75.6% for 2 × 2 and 4 × 4 prototypes at an input power of 11 dBm, respectively. Beyond achieving wide-angle and polarization-independent operation, the HS-MSR verifies that structural co-design between electromagnetic and rectifying domains is an effective approach to enhance system-level energy conversion. These results provide a practical foundation for developing compact, reliable, and high-integration wireless power modules. Future work will focus on enhancing system efficiency using high-performance GaN Schottky diodes and integrating adaptive impedance-tuning control for intelligent and reconfigurable WPT systems. Limitations of the study Current research is limited to small-scale arrays fabricated in the laboratory. Future work could enhance this by increasing scalability, exploring larger array performance of HS-MSR, and integrating them with end devices for practical applications, such as combining them with smartphones or drones for long-distance wireless power supply. So far, the RF power acquired using silicon-based diodes is limited. Further development can expand the system’s application scenarios by employing new generation semiconductors such as GaN Schottky diodes to achieve higher-power wireless microwave power transfer. Furthermore, combining this with intelligent metasurface design methods can enhance the system’s reconfigurability. Resource availability Lead contact Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Dunyu Chang (changdunyu@bupt.edu.cn). Materials availability This study did not generate new unique reagents. Data and code availability •The data supporting the findings of this study are available within the manuscript and the supplemental information. The data supporting the findings of this study are available within the manuscript and the supplemental information. •This study does not generate original code. This study does not generate original code. •Any additional information required is available upon reasonable request to the lead contact. Any additional information required is available upon reasonable request to the lead contact. Acknowledgments This work was supported in part by supported by “the Natural Science Foundation of Beijing (grant no. L254051),” “the National Natural Science Foundation of China (grant no. 62271063),” and “the Director Fund of Beijing Key Laboratory of Space-ground Interconnection and Convergence.” Author contributions D.C.: methodology, investigation, simulation, measurement, and writing; C.W.: methodology, investigation, and writing; J.Z.: review, editing, supervision, funding acquisition, and project administration; X.Z.: methodology and validation; S.B.: data curation and investigation; L.D.: methodology and validation. Declaration of interests The authors declare no competing interests. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERSoftware and algorithmsOrigin 2023bOriginLabhttps://www.originlab.com/2023bCST Studio SuiteSIMULIA-Dassault Systèmeshttps://www.3ds.com/products/simulia/cst-studio-suiteAdvanced Design System (ADS)Keysighthttps://www.keysight.com.OtherRogers 5880 substrateCommercial SupplierN/AHSMS-2860 Schottky diodeCommercial SupplierN/A Experimental model and study participant details The experimental models in this study mainly include metasurface structure and rectifier. The design and simulation of the metasurface was achieved using CST, while the design and simulation of the rectifier was achieved using ADS. Method details Full-wave EM simulation and calculation of absorptivity/radiation-to-AC efficiency To evaluate the capture capability of the HS-MSR unit cell under different incidence angles and polarization states, we performed full-wave simulations using periodic boundary conditions to emulate an infinite periodic metasurface. A Floquet port was used to excite a normally/obliquely incident plane wave with 1 W incident power. The incidence angle θ and polarization angle ϕ were swept to generate the datasets for different angular and polarization conditions. Diode modeling and impedance self-matching strategy The HSMS-2860 Schottky diode was selected as the rectifying device owing to its low turn-on voltage and fast switching behavior. Because the rectifier is nonlinear, its input impedance depends on the input power. We used ADS to perform large-signal simulations of the rectifier branch and extracted the power-dependent input impedance. To realize direct impedance matching between the metasurface feed gap and the rectifier, a lumped port was placed across the feed gaps in CST with the reference impedance set to the conjugate target. Co-simulation and scalability analysis To account for the coupled “field–circuit” behavior, we adopted CST–ADS co-simulation. Specifically, the Z-parameters of the metasurface unit cell were extracted from CST and imported into ADS. The Z-parameter dataset was connected through a data access component (DAC) and used to specify the source impedance, while each cell was represented as an equivalent RF voltage source whose amplitude depends on frequency, incidence angle, and polarization. Diodes were connected in parallel at the feed gaps and followed by the DC routing and filtering network. Prototype fabrication The HS-MSR was fabricated on a single-layer Rogers 5880 substrate with dielectric thickness H =H1.524 mm and copper thickness T = 35 μm (total thickness 1.56 mm). The unit-cell geometry parameters are as follow: R1 = 4.16 mm; arrow-branch parameters L1 = 0.8 mm, L2 = 1.5 mm, R2 = 5 mm; T-slot parameters T1 = 1.84 mm, M1 = 0.72 mm, T2 = 2.67 mm, M2 = 0.1 mm; and via diameter D = 0.6 mm. The DC routing adopts a parallel-branch architecture: via-connected cells form an equipotential DC− plane, while non-via cells are connected by microstrip lines to form an equipotential DC + plane. All equivalent rectifying branches are combined into a single DC filter to produce a smooth DC output. Quantification and statistical analysis Replicates Key efficiency datasets (corresponding to Figures 18, 19, 20, 21, 22, and 23) were collected from 5 independent fabricated prototypes, with 10 repeated measurements per condition on each prototype. Power-setting and angular-position uncertainties were recorded (typical ±0.5 dB and ±5°). Measured efficiencyRF-DC of the rectified metasurface with and without harmonic suppression (except for the presence of the harmonic-suppression structure, all other key conditions are kept identical) (A)Measurement results. (B) Physical image. Randomization To mitigate systematic drift, the order of power/angle/polarization sweeps was interleaved or randomized across runs and prototypes Inclusion/exclusion criteria No data were excluded unless a clearly documented measurement failure occurred (e.g., unstable power-meter readings, loose connections causing discontinuous voltage readings, or diode open/short failure). Data For each condition, the reported value corresponds to the mean across prototypes; uncertainty was summarized by combining repeatability and calibration tolerances to provide traceable confidence in the reported efficiency trends. Contributor Information Dunyu Chang, Email: changdunyu@bupt.edu.cn. Jinling Zhang, Email: zhangjl@bupt.edu.cn. References Associated Data Data Availability Statement •The data supporting the findings of this study are available within the manuscript and the supplemental information. The data supporting the findings of this study are available within the manuscript and the supplemental information. •This study does not generate original code. This study does not generate original code. •Any additional information required is available upon reasonable request to the lead contact. Any additional information required is available upon reasonable request to the lead contact.