Elsevier

Desalination

Volume 476, 15 February 2020, 114230
Desalination

Multistage pressure-retarded osmosis configurations: A unifying framework and thermodynamic analysis

https://doi.org/10.1016/j.desal.2019.114230Get rights and content

Highlights

  • A thorough thermodynamic analysis was applied to multistage PRO.

  • A unified framework was developed to compare various multistage configurations.

  • Comprehensive multistage design taxonomy was developed.

  • Optimal multistage design from a practical perspective was identified.

Abstract

Pressure-retarded osmosis has enjoyed increasing research interest over the last decade. Recent studies focusing on single-stage PRO designs have raised doubts regarding the long-term economic viability of the technology. While most of the analyses are based on single-stage operation, comprehensive analysis of multistage PRO which shows promise for better energetic performance is absent. Previous studies on multistage PRO differ in their design philosophies and performance metrics, leading to an incomplete assessment regarding the potential benefits of multistaging. In this paper, we develop a unifying framework to classify several existing multistage configurations. In addition, we analyze the multistage PRO system from a thermodynamic perspective. Among the two major multistage design strategies, namely interstage pressure control and independent feed inputs to each stage, we found the latter to be more effective towards increasing net power density. In comparison to a single-stage device, a 10-stage system achieves around 9% higher net power density while using the same membrane area.

Introduction

As recent studies [1,2] have found potential links between climate change and natural disasters, the need for renewable energy technologies is rapidly growing. Mature renewable technologies such as wind or solar have an inherent disadvantage since they are intermittent and cannot be used as a reliable source of base load power, without additional energy storage. Salinity-gradient power technology, in contrast, is non-intermittent renewable technology. Two major technologies for capturing salinity-gradient power are pressure-retarded osmosis (PRO) and reverse electrodialysis (RED). Yip et al. [3] found that PRO has both higher efficiency and power density.

The minimal components required for the PRO process are shown in Fig. 1. Draw stream is more concentrated than the feed stream. The PRO module is effectively a counterflow mass exchanger where the two channels are separated by a membrane that allows water to pass through while rejecting significant portion of the solutes. The draw stream is pressurized to a pressure less than its osmotic pressure before entering the module. Because of the salinity difference, water is drawn to the draw side. The draw stream is run through a turbine after exiting the PRO module. Because of the increased flow rate, the energy extracted from the turbine can be larger than what was required to pressurize the draw stream in the pump, hence generating net positive energy.

PRO has received significant interest from researchers around the world as evidenced by the rapidly increasing number of publications on the topic [4]. Notwithstanding this research interest, PRO has not yet been successfully commercialized. The only major attempt at commercialization was by Statkraft [5,6], which had closed operations in 2014. In order to develop PRO as a commercially viable technology, researchers should strive toward improving a performance metric that directly relates to the economics of PRO as a power production technology. While power density is widely used as a performance metric [[7], [8], [9], [10]], Chung et al. [11] recently found that net power density is a more useful performance metric because it captures both the economic and energetic aspects of a PRO system. In this manuscript, we evaluate multistage operation as a means to improve net power density of PRO.

The vast majority of the studies and tests have focused on single-stage PRO (including large pilot plants such as Japan's Megaton Project). Multistage operation results in improved energetic performance in related processes such as reverse osmosis [12] and has the potential to outperform conventional single-stage PRO. Research into multistaging of PRO for improved performance is still in its infancy, with no clear multistage configurations identified for maximizing PRO performance. This is because researchers [[13], [14], [15], [16]] have used different performance metrics and analyzed contrasting multistage configurations without sufficient theoretical justification or a common design philosophy. This makes it difficult to compare different multistage designs available in the literature in a coherent manner. The goals of this paper are therefore to theoretically analyze multistage systems and to develop an optimal design strategy. We establish a unified design taxonomy for multistage PRO system and apply thermodynamic analysis to three different multistage configurations. We then compare the three configurations to a baseline single-stage model to determine the optimal design.

Section snippets

Model description

We follow the modeling approach described in detail in Chung et al. [17]. The major building blocks of the model are summarized here. Since we want to capture property variations in the flow direction (i.e., the x direction), we define a differential control volume as shown in Fig. 2. The water mass balance for the control volume in the draw channel yields

ṁdx+Δxṁdx=JwdchΔx.where Jw is water flux and dch is the channel depth (into the page). By Taylor series expansion, we have ṁdx+Δx=ṁdx+dṁ

Multistage PRO

In this section, we investigate multistage PRO in more detail. First, existing studies on multistage systems are reviewed in Section 3.1, focusing on the multistage flow configurations considered therein and the performance metrics used to evaluate competing designs. Then, in Section 3.2, we define two strategies for designing multistage systems. Based on these two strategies, we define a unified multistage design taxonomy and classify existing systems using this general framework in Section 3.3

Thermodynamic analysis

In order to understand how interstage pressure and input exergy affects the multistage system, we first analyze the systems from a thermodynamic perspective. We use a single-stage system as our baseline model and make frequent comparisons to it. In doing so, we use the same total area for both single-stage and dual-stage systems, i.e., the sum of the area of stage 1 and stage 2 is equal to the area of the single-stage system. Therefore, comparing these two systems in terms of net power is

Optimal multistage design

From the analysis in Section 4, we suggest the optimal multistage configuration shown in Fig. 11.

This system uses an independent feed stream in each stage. Although we focused on two-stage systems in Section 4, we can extend the analysis to N stages with N > 2. Fig. 12 shows how net power density changes as the number of stages increases. For this plot, we fixed the total system size to be the same for all points. For example, if single-stage system has length of 5 m, then 10-stage system has

Conclusion

A comprehensive analysis of various multistage PRO configurations has been presented, starting with a survey of the literature followed by a thermodynamic analysis of two-stage systems. The primary conclusions are as follow:

  • Inconsistent multistage design philosophies and evaluation metrics exist in the literature, which have prevented the identification of an optimal PRO design.

  • Two major design strategies with respect to multistaging of PRO were identified: one based on adjusting the input

CRediT authorship contribution statement

Hyung Won Chung: Conceptualization, Methodology, Software, Writing - original draft. Jaichander Swaminathan: Conceptualization, Validation, Investigation. John H. Lienhard V: Supervision, Writing - review & editing, Conceptualization, Formal analysis.

Declaration of competing interest

The authors declare no conflict of interest.

Acknowledgments

The authors would like to thank Kuwait Foundation for the Advancement of Sciences (KFAS) for their financial support through project no. P31475EC01.

References (1)

    Cited by (19)

    • A systematic methodology for targeting the thermodynamic limit of pressure-retarded osmosis with non-zero driving force

      2022, Journal of Cleaner Production
      Citation Excerpt :

      Li (2018) investigated the effects of multi-stage PRO design, and the results indicated that the multi-stage configuration could improve the system's normalized specific energy production and power density by progressively reducing hydraulic pressure. Chung and Swaminathan (2020) proposed a unified taxonomy for multi-stage PRO design. Their thermodynamic analysis of a two-stage PRO showed that managing exergy input was crucial to increasing energy density.

    View all citing articles on Scopus
    View full text