In network functions virtualization (NFV) environments, network security is provided with the deployment of virtual security functions (VSF) on general-purpose servers to filter and monitor the incoming traffic. However, the traffic handled by VSFs may have different security requirements in a network. On the other hand, there are some operational objectives that differ according to the needs of the network such as minimizing cost, latency, and energy consumption etc. Therefore, the issue of placing VSFs considering both security requirements and operational objectives is an important research challenge. In this paper, we tackle the problem of energy-efficient VSF placement (EE-VSFP) to minimize energy consumption while meeting flow-level security requirements as well as resource constraints. We develop an integer linear programming (ILP) model for this problem to minimize server energy consumption, and also propose a novel heuristic algorithm to solve the problem for larger scale network instances within practical time limits. Evaluation results show that our heuristic reduces energy consumption by up to 48 % compared to baseline placement solutions. In addition, we demonstrate that in most cases our heuristic can provide optimal solutions while running much faster than the ILP, and drastically reduces the average flow path length by up to 53 % compared to the ILP.