IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL. 11, 2012 973 Design and Implementation of a Broadband Single Layer Circularly Polarized Reectarray Antenna Reza Shamsaee Malfajani, Student Member, IEEE, and Zahra Atlasbaf, Member, IEEE Abstract—The design and implementation of a broadband single-layer circularly polarized reectarray antenna is presented in this letter. Each element in the proposed reectarray consists of a circular microstrip patch attached to four variable-length phase delay lines to achieve more than 650 of linear phase for frequen- cies in the range 9.6–11.2 GHz. Measurement results show 15.5% 3-dB gain bandwidth and 14.6% 3-dB axial-ratio bandwidth. Index Terms—Circular microstrip patch, circular polarization, phase delay lines, reectarray antenna. I. INTRODUCTION T HE REFLECTARRAY technology combines concepts of phased arrays with reectors. Flat surface, light weight, and relatively low fabrication cost make microstrip reectarray antennas a suitable replacement for the conventional parabolic reectors in radar and satellite systems. One drawback of reectarray antennas is their narrow band- width, which is due to the narrowband nature of the microstrip patches and difference in spatial phase delays between the feed and elements in the array. The latter is more considerable in large-sized reectarrays [1]. Using elements with a large-range linear phase response increases the reectarray bandwidth and reduces the sensitivity to fabrication tolerance [2], [3]. The phase range can be achieved by using phase delay lines [4], thick substrates, multilayered structures [5], or elements, which are variable in size or rotation [6], [7]. Among all, the phase delay line technique is appropriate for obtaining the large linear phase response and broad bandwidth with low fabrication cost. It is common for satellite communication systems to use cir- cularly polarized antennas with no sensitivity to transmitter and receiver orientation. In previous works, narrow-bandwidth rect- angular patches or rings with different rotation angles have been used for circular polarization operation [7], [8]. In those inves- tigations, a circularly polarized feed was needed for circular po- larization operation of the reectarray. In this letter, we present a simple and small-sized element with a large linear phase range that can be employed for circular polarization operation even with a linearly polarized feed. The linear phase range of the element is more than 650 , which has not been achieved in the previous works on circular polariza- tion with single-layer and multilayered structures [9], [10]. The large linear phase range enables the element to be used in large Manuscript received December 29, 2011; revised March 12, 2012; May 06, 2012; and July 09, 2012; accepted August 02, 2012. Date of publication August 16, 2012; date of current version August 30, 2012. The authors are with the Department of Electrical and Computer Engineering, Tarbiat Modares University, Tehran 14115, Iran (e-mail: r.shamsaee@modares.ac.ir; atlasbaf@modares.ac.ir). Digital Object Identier 10.1109/LAWP.2012.2213570 Fig. 1. Element structure. mm, mm, mm, mm, mm. reectarrays. A special arrangement of the elements in the array is also presented for decreasing the axial ratio and improving the circular polarization performance of the reectarray. II. DESIGN PROCEDURE The proposed element is a single-layer circular microstrip patch connected to four variable-length phase delay lines with angular distance of 90 as shown in Fig. 1. The required phase shift is obtained by adjusting the lengths of the phase delay lines. At the rst step, the required phase shift for each element is calculated. The phase shifts should compensate the difference in spatial phase delays between the feed and each element in the reectarray. For a planar reectarray with elements illuminated by a low-gain feed located at distance of from the reectarray, the required phase delay for each element is calculated by [7] (1) where is the required phase shift for the th element. According to Fig. 2, is the position vector for the th element, and is the direction of the main beam. is the free-space wavenumber. In fact, (1) results in a cophasal plane for the reected wave at the aperture. The reected eld in the observation direction of at the far eld is calculated by the array theory as [7] (2) 1536-1225/$31.00 © 2012 IEEE