Research on heat integration has made significant advances in reducing utility consumption in chemical plants. However, the idea of work exchange between high and low-pressure process streams to reduce the consumption of the relatively expensive electricity has received limited attention. In this article, we present a more efficient mixed-integer nonlinear programming (MINLP) formulation to synthesize work heat exchanger networks (WHENs). We propose a superstructure that explicitly considers constant pressure streams for heat integration and enables an optimized selection of end-heaters and end coolers to meet the desired temperature targets. Using a few examples, we demonstrate that simultaneous integration of work and heat in a chemical plant can offer significant savings in total annualized cost. In a case study from the literature, our approach yields a network with 3.1% lower total annualized cost, 10.6% more work exchange, and 81.0% more heat exchange than the best solution obtained from the existing literature approach. Furthermore, our approach successfully solves two case studies that previous literature approaches fail to solve. (C) 2016 Elsevier Ltd. All rights reserved.