Engineering efficient desalination systems: Materials development and systems analysis in capacitive deionization

Hand, Steven

Permalink

https://hdl.handle.net/2142/106412 Copy

Description

Title

Engineering efficient desalination systems: Materials development and systems analysis in capacitive deionization

Author(s)

Hand, Steven

Issue Date

2019-07-24

Director of Research (if dissertation) or Advisor (if thesis)

Cusick, Roland

Doctoral Committee Chair(s)

Cusick, Roland

Committee Member(s)

• Guest, Jeremy
• Espinosa-Marzal, Rosa
• Mariñas, Benito

Department of Study

Civil & Environmental Eng

Discipline

Environ Engr in Civil Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Capacitive Deionization
• Desalination
• Electrochemistry
• Sensitivity Analysis
• Techno-Economic Analysis

Abstract

Climate change and population growth are increasingly rendering traditional water sources inadequate in many regions. Accordingly, desalination of unconventional water sources such as brackish aquifers, which has been historically cost prohibitive, will be necessary to provide more sustainable water resources. While reverse osmosis (RO) is currently utilized in most desalination processes worldwide, capacitive deionization (CDI) — wherein salt ions are removed from solution and stored in electric double layers — is a class of alternative electrochemical desalination technologies that can operate at energy consumption near the thermodynamic limit of separation without membranes and at far lower pressures. However, the viability of CDI systems is limited by the low charge efficiency and influent concentration reduction of carbon electrodes. Numerous material modifications and architectural strategies have been individually proposed to overcome these limitation. The most common method of increasing charge efficiency in CDI systems is to add ion exchange membranes (IEMs) over the electrodes in membrane CDI (MCDI). Replacing the stationary electrodes of CDI with flowable carbon slurries in a configuration termed flowable CDI (FCDI) has been proposed as a means of increasing concentration reduction. Therefore, specific objectives of this dissertation were to: (1) improve CDI energetic and salt removal performance without IEMs; (2) assess performance and input sensitivity in CDI configurations across the operational and material design landscape; and (3) determine performance and cost benchmarks necessary for cost-effective, full-scale CDI systems. To that end, new faradaic material modifications were developed to improve CDI performance and concentration reductions. Global sensitivity analysis across the CDI design space was conducted to evaluate the performance, sensitivity, and tradeoffs in design and operation. A parameterized framework was developed for sizing and costing full-scale CDI system from commonly reported material and operational parameters. Firstly, faradaic MnO2 was deposited onto carbon aerogels via cyclic voltammetry (CV) and electroless deposition (ED) at varying deposition times. Deposition techniques and conditions were evaluated according to common CDI performance metrics. SEM imaging and XRD analysis of the different deposition conditions reveal uniform, homogeneous deposition of MnO2 under ED, while CV deposition produced localized dendritic growth leading to less complete aerogel coverage. MnO2 cathodes were then paired with a carbon slurry flow-anode in a hybrid CDI (HCDI) system to allow for operation MnO2 electrodes without counter electrode capacity limitations. ED MnO2 absorbed approximately 1.5 to 3 times more salt than bare carbon aerogels. Similarly, observed charge efficiency increased from 0.2 for bare carbon electrode to 0.43–0.91 for ED MnO2, depending on deposition time and operating current. When operated at 5 A m-2, the HCDI system consumed 0.45 kWh kg-NaCl-1 without energy recovery, far below the mean reported values of 0.90 kWh kg-NaCl-1 for standard CDI. If 100% of the energy released during the desalination discharge stage was recovered, net energy consumption would be 0.26 kWh kg-NaCl-1. These results suggest that the incorporation of pseudocapacitive materials into CDI systems can greatly increase desired characteristics for desalination of brackish water. To assess performance across the CDI design space, a one-dimensional porous electrode (M/F)CDI model, was used to measure the sensitivity of CDI performance to material selection, design, and operating choices. Monte Carlo and Morris methods were used to conduct sensitivity analysis of eight common input parameters across multiple electrode geometries and influent salt concentrations with respect to six commonly used performance metrics (salt adsorption capacity [SAC], average salt adsorption rate [ASAR], energy-normalized adsorbed salt [ENAS], charge efficiency [λ or η¬CE], and round-trip efficiency/energy recovery [η¬RTE], and thermodynamic efficiency [η¬T]). The sensitivity analysis shows that operating current density, electrode specific capacitance, and contact resistance were the parameters which most significantly dictated (M/F)CDI performance. MCDI outperformed both CDI and FCDI across all performance metrics. The simulations highlighted the importance of maintaining charge efficiency and round-trip-efficiency which limited the performance of CDI and FCDI, respectively. Lastly, a parametrized system sizing and costing framework for capacitive materials in CDI was developed. Using this framework, a techno-economic analysis (TEA) of capacitive CDI systems was conducted, including both sensitivity and uncertainty analyses. The results of the TEA show that capital costs are generally greater than operating costs for both MCDI and CDI at relevant system lifetimes. Notably, the additional costs of incorporating IEMs in MCDI systems significantly outweighs the performance improvements of that architecture. Using these results, necessary materials lifetimes and costs for economical operation CDI were evaluated. In general, CDI systems must achieve lifetimes beyond two years in order to favorably compete with RO, depending on target concentration reduction. For IEMs or performance enhancing electrode modifications to be economically viable, they must achieve costs under $10 m-2. These findings suggest that CDI systems could be a cost-effective alternative to RO for brackish water desalination if target system lifetimes and material costs can be met. Alternatively, because capital cost is both the majority of total CDI system price and proportional to concentration reduction, CDI may be competitive in ion separation applications which occur at low influent salinity and concentration reduction (such as NO3- removal from drinking water). Using treatment water quality set by existing NO3- removal case studies, CDI systems were sized to selectively remove NO3- below maximum contaminant levels (MCLs) set by the US EPA and WHO. Initial results indicate that CDI can remove NO3- to below MCLs in all case studies at less than $0.3 m-3, below the $15.87 m-3 target suggested for the State of California.

Graduation Semester

2019-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/106412

Copyright and License Information

Copyright 2019 Steven Hand

Owning Collections