Abstract
In this communication, a dual-band, orthogonal polarization transmission metasurface antenna operating at 6.5 GHz and 9.5 GHz is proposed. The element of this transmitarray is composed of only two identical layers of metal plates, with air medium filling the space between the upper and lower layers. Each layer has four sets of staggered I-slots, thus ensuring dual-frequency orthogonal polarization operation. Moreover the proposed element overcomes the challenge of achieving full 360-degree phase shift in dual-layer metal-only structures without any additional connecting elements. By altering the length of the slots, it can achieve a continuous full 360-degree phase shift while maintaining a transmission amplitude above −2 dB and exhibits co-polarized transmission in both bands. Then a prototype containing 16 × 16 elements is designed and fabricated to verify the feasibility of the design scheme. The measurement results, which are consistent with the simulation, demonstrate that the maximum gains of 24.85 and 27.64 dBi are obtained at 6.4 GHz and 9.4 GHz, with corresponding aperture efficiencies of 23.18% and 20.4% and a 3-dB gain bandwidth of 15.6% and 22.34%. The good transmission coefficients of the element and the dual-band capability make the design a competitive candidate for satellite and high-power applications.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Metasurface has been widely investigated as a new generation of high gain antennas [1], combining the advantages of the optical lens principle and the array antenna. Compared to reflectarray antennas (RAs) [2], transmitarray antennas (TAs) have been studied more intensively due to the elimination of feed obstruction. Until now, areas such as folded TAs [3], dual-band antennas [4–6], multi-beam antennas [7–9], beam-scanning antennas [10], orbital angular momentum antennas [11], reconfigurable antennas [12,13] and highly transparent antennas [14–17] have been proposed in the development of metasurface.
Despite the various research being conducted on TAs, the design for achieving high performance remains a difficult task deserving of more in-depth research. The prevailing design approach is usually to use multilayer frequency selective surface structures (MFSS) [18–21] and receiver-transmitter designs [22–25], where the cell is constructed with 3 or more layers, due to the fact that fewer layers make it more difficult to design. According to calculations [26], a 3-layer structure is required for a common TAs element to achieve a phase shift range of more than 360 at an S21 magnitude of 1 dBi. Many studies have been proposed in order to reduce the thickness of the transmission metasurface elements, and one of the most representative approaches is to utilize vias structures [27–29]. When the size of the metal patch is too small, resulting in a lower transmission coefficient, the vias can effectively improve the transmission performance of the element. In [28], a continuous 360 phase shift is realized by using four vias and cross-patches with two degrees of freedom. The achieved 1-dB gain bandwidth of the TA is 9.6%. In [27], elements consist of two multi-resonant dipoles and two metal vias achieved wide 1-dB gain bandwidth of 10%(12.9-14.35 GHz) and 6%(19.7-20.9 GHz). Moreover, the use of the Huygens’ principle [30–32] can also reduce the thickness of the TAs and enable dual-band operation [33] .
However, due to the dielectric layers in most of the elements, both the extreme conditions in the space environment and the temperature changes due to the heat generated by the dielectric losses when faced with high power beams can cause changes in the dielectric constant of the material, which in turn affects the performance of the antenna. To overcome these drawbacks, metal-only metasurface have received much attention. Currently, one of the current design approaches for dual-layer metal-only elements is to use metal posts to connect the upper and lower structures [34–36]. However, the addition of metal posts not only makes the antenna more difficult to manufacture, but also introduces more insertion loss. There are also elements that rely on their own special structure to obtain excellent performance [37–39]. The element presented in [38] relies on a metal-only Huygens unit consisting of a double anti-symmetric multi-slot structure, which successfully achieves a full 360 phase shift and high transmission amplitude (2 dB). However, a large number of parameter sweeps are required to obtain the performance of this element, which makes the design process extremely complex and time-consuming. Among the published articles, the double-layered pure metal element presented in [39] is the most outstanding element, and this element is capable of providing a phase shift of 180 degrees each in two states with a total phase shift of 360 degrees at 10 GHz by varying the opening angle of the upper and lower C-slots. However, the structure still has a drawback: it will lead to polarization rotation of the transmitted electromagnetic wave. And due to its circular grooves, when applied to dual-band designs, the different structures will exhibit performance coupling that is difficult to ignore.
In this communication, we present for the first time an innovative dual-layer metal-only TA, which not only achieves a continuous 360-degree phase shift range, but also possesses a dual-frequency transmission capability. By etching two sets of independent and complementary I-shaped narrow slots, we have achieved a component with a full 360-degree phase-shift range without the need for polarization rotation. This breakthrough design avoids the use of additional connectors, such as metal posts, and drastically simplifies the structure and reduces the difficulty of machining. Although these narrow slots take up more space and may cause structural crossover problems in dual-frequency designs, in this study the narrow slots operating at different frequencies are designed to be arranged orthogonally to each other in order to minimise the coupling effect between them. To the best of our knowledge, this structure is the only metal-only unit that has successfully realised a dual-band functional design with an outstandingly high transmission coefficient and phase compensation covering the full 360 degrees.
2. Unit cell design
Dual independent units are typically employed to achieve dual frequency functionality in most articles. And in this communication we endeavour to make the dual independent units work at a same frequency to obtain a larger phase shift range. It is observed that a dual-band metal-only unit with similar slot placements is proposed in [19] and [20], respectively, with the slots situated at the corners and the center. However, the unique H-shaped slot in [20] exhibits a wider passband. In addition, the surface currents at resonance are concentrated around the wider slots on either side, while the centre-connected slot have almost no current. This indicates that the primary resonance structure of the H-shaped slot is situated in the longer branches on both sides, with the distinctive feature of narrowing the period of the branches on both sides in the corresponding polarization direction.
Based on the above analysis, the proposed dual-band element operates in an orthogonal linearly polarization (LP) mode. As illustrated in Fig. 1, the element consists of two layers of metal plate etched with two sets of orthogonally aligned I-shaped narrow slots. To ensure that the device operates in dual-band orthogonal polarization mode, the narrow slots are arranged orthogonally. Structures with the long side of the I-notch parallel to the Y-direction operate at 6.5 GHz and structures parallel to the X-direction operate at 9.5 GHz. Separated by an air medium with a thickness of 8 mm, both layers of sheet metal have a thickness of 1 mm and the spacing of the elements is set to 30 mm (P = 30 mm, 0.65 at 6.5 GHz and 0.95 at 9.5 GHz). The material of the element in this communication is chosen to be copper with high electrical conductivity ( 43.48), which gives the element a better transmission coefficient.
Fig. 1.
To obtain further data on the transmission performance of the element, the full-wave simulation software CST Microwave Studio is used to model the element and obtain the transmission response under periodic boundary conditions. As shown in Fig. 1(c), the element has two sets of I-shaped narrow slots in the Y-direction: Part A and Part B. When setting L1 = 26 mm, L11 = 2 mm, L2 = 22.1 mm, the element is simulated and the curves shown in Figs. 2(a), (b), and (c) can be obtained, where the black curve indicates the transmission coefficient of the element of the X-polarized wave, and the red curve indicates its phase.
Illustrated in Fig. 2(a), (b), when it exists only Part A and Part B, the element exhibits band-pass at both frequencies. Based on this property, it is common to increase or decrease the length of the grooves to shift the pass-band range of the element towards lower or higher frequencies, and thus obtaining different phase compensations at the frequency we need while maintaining a high transmission performance. However, in the range where the transmission amplitude is higher than 2 dB, the difference in phase shift produced by the two part for the lowest and highest frequency is only 222.1 and 177.5. It is obvious that a sufficient phase shift range cannot be obtained by Part A or Part B alone by the above method. However, when both Part A and Part B exist at the same time, as in Fig. 2(c), the cell has a much wider pass-band in this state, and the phase shift difference for high and low frequencies reaches 575.5, which makes it possible for the element to be able to achieve a continuous 360-degree phase compensation by adjusting length of the grooves. Combining the above results, in Fig. 2(d) we give the transmission properties of the designed cell in four different states as a feasibility verification.
By analysing the surface current distribution of the proposed TA element, the operation of the cell at the two frequency points can be understood more clearly, as shown in Fig. 3. When the cell operates at 6.5 GHz, the surface currents are mainly concentrated in the larger sized I-shaped slits in the y-direction, while at 9.5 GHz, the surface currents are mainly concentrated in the smaller sized I-shaped slits in the x-direction. By adjusting the sizes of the two sets of orthogonally arranged slits separately, phase tuning at the two frequency points can be achieved. Then the cell was investigated around the thickness of the air layer. The simulation results of the transmission performance at various heights h at 6.5 GHz are depicted in the in Fig. 4. It can be observed from the figure that while the variation in the air layer height has a minimal impact on the battery’s transmission phase, excessively high or low values of h lead to a significant decrease in the transmission coefficient of the cell. It is worth noting that Fig. 2(d), Fig. 4, and Fig. 5 all show the results obtained by simulating the complete unit structure, except for Fig. 2(a), (b) and (c) where the orthogonally aligned slots are not included in the simulation in order to better demonstrate the performance of the individual structure.
The performance of the final designed element at two frequencies is shown in Fig. 5(a): the phase shift of the cell reaches a full 360-degree range when the parameter L1 is varied from 26 to 23 mm and the parameter L11 is varied from 5 to 2 mm at 6.5 GHz. And a full 360-degree range is also reached at 9.5 GHz shown in Fig. 5(b) when the parameter L33 is varied from 7 mm to 11 mm. It is clear that the device designed in this communication has a phase compensation capability with a range of more than 360 degrees at both frequencies and a transmission coefficient higher than 2 dB, which provides a good basis for the design of the dual-band transmission metasurface.
However, it should be noted that the design employs two parameters for regulating the performance of the element at 6.5 GHz, rather than a single parameter. This approach is taken into account because under certain parameters, the grooves may intersect, leading to the formation of isolated island structures on the array surface that are disconnected from the surrounding structure. Obviously, this structures are impractical and would result in a fragmented overall structure. With consideration for the coherence and integrity of the array, we chose to rely on the variation of two parameters, and , to modify the performance of the element.
In order to explore the interaction between the two orthogonal structures, Fig. 6 gives the transmission performance of the unit at 6.5 GHz and 9.5 GHz, respectively, with the same parameter changes. From Fig. 6(a) and (b), it can be seen that the unit have good transmission amplitude at both frequencies, which are mostly above -2 dB. In Fig. 6(c) and (d), it can be seen that the unit has a very stable phase shift at 9.5 GHz, and the variation of L33 hardly affects it. At 6.5 GHz, there is a small fluctuation in the phase shift capability of the unit, which is difficult to avoid due to the intersection of the two orthogonal structures. For the design of the array, such fluctuations do not have a significant impact on the overall structure, and the unit remains an excellent choice for a dual-band transmitarray.
Since the boundary of the cells passes through part of the slit, this may result in some positions where the slits may not be in a standard I-shape when forming an array. However, in the periodic structure, this will not affect the performance of the whole array too much. That is because it can be noticed that the error only appears at the longer slits on the sides of the cell. For the slits at the corners, the length of L4 is fixed, so that it is still a more complete I-shape when it meets the surrounding cells. Further, since the phase compensation ability of the cells varies linearly with the size of the slits, most of the neighbouring cells have small or no gap in the slit backbone length at the boundary, and the performance of the cells is less affected, and serious structural misalignment occurs only at the positions where the phase compensation has jumped. Combining the above reasons, serious structural anomalies will only occurs in a very small number of positions when composing the array, which does not have a large impact on the results.
To validate the design, a double-layer metal-only transmission metasurface containing 16 16 elements (480 480 mm) is designed and simulated. Since the rectangular slot structure is naturally weak in the face of obliquely incident waves, in order to reduce this drawback, the phase centers of the two horns are located 514.7 mm away from the surface of the array, corresponding to an FD of 1.07, and the distribution of the cell phase compensation for each position of the array can be found by the following equation
where is the propagation constant, is the focal distance, and is the distance from the central point of the TA to the element of the TA. Based on the relationship between the transmission phase and the parameters illustrated in Fig. 5, we can then obtain the size distribution of the cells at different locations in the array.
3. Design and measurements of transmitarray
Moreover, to ensure the feasibility of practical applications, a prototype has been fabricated for testing. An acrylic frame, custom-designed to match the size and contours of the array, acts as a support structure to which the metal array can be securely fastened, but the central portion of the array may deform when the copper sheet is placed vertically. Therefore, hollow nylon columns with a length of 8 mm are placed between the upper and lower metal plates and fixed with nylon screws. This arrangement effectively maintains a stable thickness of the air layer while preventing relative displacement of the metal plates, thus ensuring the stability and reliability of the overall structure. Finally, to facilitate precise and accurate manipulation of the positions of the horn and the array, the acrylic frames supporting each two structures attached to distinct sliders and subsequently mounted on metal tracks, which not only maintains the perpendicularity of the horn to the array but also allows for a more precise adjustment of the distance between them. It should be noted here that since the structure designed in this paper operates at a low frequency, the errors introduced by the use of laser cutting in the processing are acceptable, and if the structure is extended to higher frequency TA designs, a more delicate processing method is required.
Once the transmission metasurface had been fabricated, it is measured in a far-field anechoic chamber to ensure accurate and precise test results. As shown in Fig. 7, the array is supported by acrylic material and positioned on a slide to facilitate the adjustment of the distance from the feed to the array. Considering that the self-weight of the horn could easily deform the support structure, acrylic panels with a thickness of 10 mm is selected for support to minimize the error caused by the offset of the horn beam. The whole structure was fixed on a platform and sampled using a near-field test system with a sampling interval of less than half a wavelength and a sampling plane 4 wavelengths from the transmission array. The far-field pattern can be obtained by Fourier transform. By replacing the test metasurface with a standard gain horn and then comparing the two measurements, the gain of metasurface can be determined. The measured and simulated radiation patterns of the transmitting antenna array at 6.5 GHz and 9.5 GHz are shown in Fig. 8(a) and (b), from which we can find that the normalised side-lobe level measured at these two frequencies are both below 15 dB, with a cross-polarization level of 40 dB. And the measured and simulated radiation patterns at two additional frequency in the C/X bands are also given in Fig. 8(c) and (d). Moreover, the measured and simulated gain and aperture efficiency of the TAs are shown in Fig. 9. The aperture efficiency [40] is defined as the ratio of the effective area of the antenna to the aperture area, and is given by , where is the antenna gain and is the aperture area of the antenna. It can be found that in the C-band, the maximum gain is 24.85 dBi at 6.4 GHz, with a calculated aperture efficiency of 23.18% and a 3-dB bandwidth of about 15.6%. In the X-band, the maximum gain is 27.64 dBi at 9.4 GHz, the aperture efficiency is 20.4%, and the 3-dB bandwidth is about 22.34%. Compared with the simulation results, the measured gain in the C-band is consistent with it, and the measured gain in the X-band is slightly reduced. This may be due to the fact that higher frequency electromagnetic waves are more susceptible to assembly errors in the test structure.
To clearly demonstrate the merits and novelty of the proposed design, Table 1 compares this work with some other reported works. In contrast to the conventional multilayer structures proposed in [20] and [27], this communication exhibits a full 360-degree phase shift range using only a two-layer structure. Although the addition of metal posts can make the element have a better transmission performance as proposed in [34], it will also increase the complexity of the array fabrication. The Huygens’ unit proposed in [29] also have excellent performance, but it is extremely complex and time-consuming to design, requiring a large number of parameter sweeps to build up a database to satisfy the performance requirements, and the grooves occupies too much space in the element, which makes it difficult to be applied to the design of the dual-band function. Even though the better-performing element proposed in [39] successfully removes the metal connecting posts and the design process is simpler, it brings unavoidable polarization rotation, which is also a drawback of the element proposed in [34]. And the C-slot grooves [41] will be more prone to coupling when applied to a dual-band design. However, the element proposed in this communication not only avoids the need for the metal post structure and avoids the polarization rotation of the transmitted wave through a special structural combination, but also succeeds in achieving a dual-band functionality with its orthogonal arrangement, achieving a huge optimisation in terms of both design complexity and processing cost.
One final point worth exploring is that focusing only on the low-frequency structures part A and B in Fig. 1(a), it can be seen that the periodic structure delineated by the black dashed line is not the minimal periodic structure. By dividing the square into two identical rectangular structures on the left and right, it is the rectangular structure that is the smallest independent periodic structure. The width of the rectangle is only , and similarly for the high-frequency structure the width of the minimum period structure is . However, the long side is still larger than half the wavelength, and there is no gate flap on the H-plane of the far-field radiation patterns, which is indeed a noteworthy problem. We speculate that this may be due to the peculiarities of the unit cell structure: since it is almost impossible to find a minimal periodic structure that contains both the complete Part A and B substructures, the slit is necessarily split into two parts by the boundary, which is significantly different from a conventional unit, and thus traditional methods of period determination may not be applicable to this cell.
4. Conclusion
In this communication, a novel double-layer double-layer metal-only transmission metasurface operating in the C/X band is proposed. To the authors’ best knowledge, this is the first proposed dual-band metal-only element design, which consists of only two metal plates etched with I-shaped slots, and a full 360 phase shift range can be achieved while maintaining a transmission amplitude greater than 2 dB by adjusting the length of the slots. The measurement results show that the maximum gains of the proposed transmission metasurface at 6.4 GHz and 9.4 GHz are 24.85 dBi and 27.64 dBi, with corresponding aperture efficiencies of 23.18% and 20.4%, respectively. Compared with existing studies, this work successfully realises the dual-band functionality of a double-layer metal-only transmission metasurface with a simpler element design: it only needs to change the slot length to obtain a phase shift range up to 360 degrees, and does not require complex and time-consuming sweeping of a large number of parameters. This design eliminates the need for any additional metal connections, without adding cost and complexity to the fabrication process, and the device’s good performance and dual-frequency capability provide additional options for space and high-power applications.