The paper proposes a novel approach to enhance physical layer security (PLS) in wireless networks by combining reconfigurable intelligent surfaces (RIS) and artificial noise (AN). The proposed aerial RIS (A-RIS) concept involves using a RIS-attached unmanned aerial vehicle (UAV) to improve signal quality for legitimate users while jamming signals for illegitimate users. The authors introduce a method to virtually partition the RIS, configuring phase shifts to enhance the intended signal for legitimate users and increase the impact of AN on illegitimate users. Closed-form expressions for the ergodic secrecy capacity (ESC) of legitimate and illegitimate users are derived and validated. Optimization problems are formulated to maximize network ESC by optimizing the 3D deployment of the A-RIS and RIS portions, subject to predefined quality-of-service (QoS) constraints. Simulation results demonstrate that the proposed joint A-RIS deployment and partitioning framework significantly improves network security compared to benchmarks where RIS and AN are used separately without optimization. Additionally, the proposed deployment approaches converge in less than a second, making it suitable for dynamic A-RIS deployment.The paper proposes a novel approach to enhance physical layer security (PLS) in wireless networks by combining reconfigurable intelligent surfaces (RIS) and artificial noise (AN). The proposed aerial RIS (A-RIS) concept involves using a RIS-attached unmanned aerial vehicle (UAV) to improve signal quality for legitimate users while jamming signals for illegitimate users. The authors introduce a method to virtually partition the RIS, configuring phase shifts to enhance the intended signal for legitimate users and increase the impact of AN on illegitimate users. Closed-form expressions for the ergodic secrecy capacity (ESC) of legitimate and illegitimate users are derived and validated. Optimization problems are formulated to maximize network ESC by optimizing the 3D deployment of the A-RIS and RIS portions, subject to predefined quality-of-service (QoS) constraints. Simulation results demonstrate that the proposed joint A-RIS deployment and partitioning framework significantly improves network security compared to benchmarks where RIS and AN are used separately without optimization. Additionally, the proposed deployment approaches converge in less than a second, making it suitable for dynamic A-RIS deployment.