Dynamic Optimization of Proton Exchange Membrane Water Electrolyzers Considering Usage-Based Degradation
Extracted source text · arXiv 2405.06766
\documentclass[12pt]{article}
\usepackage[margin=1.0in]{geometry}
\usepackage[utf8]{inputenc}
\usepackage{siunitx}
\sisetup{per-mode=fraction}
\usepackage{tabularx}
\renewcommand\tabularxcolumn[1]{m{#1}}
\usepackage{ltablex}
\usepackage{lscape}
\usepackage{booktabs}
\usepackage{array}
\newcolumntype{x}[1]{>{\centering\let\newline\\\arraybackslash\hspace{0pt}}p{#1}}
\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{xcolor}
\usepackage{chemformula}
\usepackage{soul}
\usepackage{authblk}
\usepackage{parskip}
\newcommand*{\plimsoll}{{\ensuremath{-\kern-5pt{\ominus}\kern-5pt-}}}
\usepackage[normalem]{ulem}
\usepackage{svg}
\usepackage{bm}
\usepackage{makecell}
\usepackage{placeins}
\usepackage{subcaption}
\usepackage{setspace}
% Using \doublespacing in the preamble
% changes the text to double-line spacing
\usepackage[style=numeric,sorting=none,bibstyle=ama,style=numeric-comp,maxbibnames=99]{biblatex} %Imports biblatex package
\addbibresource{paperpile.bib} %Import the bibliography file
\DeclareUnicodeCharacter{2212}{-}
\DeclareUnicodeCharacter{0300}{`}
\newcommand{\beginsupplement}{%
\setcounter{table}{0}
\renewcommand{\thetable}{S\arabic{table}}%
\setcounter{figure}{0}
\renewcommand{\thefigure}{S\arabic{figure}}
\setcounter{equation}{0}
\renewcommand{\theequation}{S\arabic{equation}}%
}
\title{Dynamic Optimization of Proton Exchange Membrane Water Electrolzyers Considering Usage-Based Degradation}
\author[1]{Landon Schofield}
\author[2]{Benjamin Paren}
\author[3]{Ruaridh Macdonald}
\author[4]{Yang Shao-Horn}
\author[5]{Dharik Mallapragada}
\date{}
\affil[1]{Department of Chemical Engineering, Massachusetts Institute of Technology}
\affil[2]{Department of Chemical Engineering and Materials Science, Stevens Institute of Technology}
\affil[3]{MIT Energy Initiative}
\affil[4]{Department of Mechanical Engineering, Massachusetts Institute of Technology}
\affil[5]{Department of Chemical and Biomolecular Engineering, New York University}
\begin{document}
\begin{refsection}
\maketitle
\begin{abstract}
We present a techno-economic optimization model for evaluating the design and operation of proton exchange membrane (PEM) electrolyzers, crucial for hydrogen production powered by variable renewable electricity. This model integrates a 0-D physics representation of the electrolyzer stack, complete mass and energy balances, operational constraints, and empirical data on use-dependent degradation. Utilizing a decomposition approach, the model predicts optimal electrolyzer size, operation, and necessary hydrogen storage to satisfy baseload demands across various technology and electricity price scenarios. Analysis for 2022 shows that including degradation effects raises the levelized cost of hydrogen from \$4.56/kg to \$6.60/kg and decreases stack life to two years. However, projections for 2030 anticipate a significant reduction in costs to approximately \$2.50/kg due to lower capital expenses, leading to larger stacks, extended lifetimes, and less hydrogen storage. This approach is adaptable to other electrochemical systems relevant for decarbonization.
\end{abstract}
\doublespacing
\section{Introduction}
\input{Introduction}
\section{Methods}
\input{Methods}
\section{Results and Discussion}
\input{ResultsandDiscussion}
\section{Conclusion}
\input{Conclusion}
\newpage
\printbibliography
\end{refsection}
\clearpage
\section{Supporting Information}
\begin{refsection}
\input{SupplementaryInformation}
\clearpage
\printbibliography
\end{refsection}
\end{document}
Policies emphasizing economy-wide decarbonization are increasingly focused on promoting production of low-carbon hydrogen (H$_2$) to address emissions from difficult-to-electrify end-uses. For example, the emissions-indexed production tax credit (PTC) for \ch{H2} as part of the Inflation Reduction Act in the U.S. provides up to \$3/kg \ch{H2}, which has led to several project announcements for H$_2$ production using electrolyzers powered by low-carbon electricity \cite{Yarmuth2021-wu,UnknownUnknown-ci,Arjona2023-bc}. Proposed electrolyzer projects include both “islanded” configurations, where renewable electricity is co-located with the electrolyzer and is the sole source of electricity input, as well as “grid-connected” systems involving using grid electricity (typically from the wholesale market) as well as on-site or contracted (via power purchase agreements) renewable electricity supply. In both cases, cost-effective operation of the electrolyzer will likely involve operating at much less than 100\% capacity utilization to manage fluctuations in variable renewable energy (VRE) availability as well as grid electricity prices (that are also increasingly influenced by VRE availability) \cite{Stansberry2017-eo,Laoun2016-mb,Sarrias-Mena2015-vy,Saadi2016-xg,Zhao2015-bk}. This part load operation not only has implications for capital utilization, which has been extensively studied, but also stack lifetime through accelerated degradation which is less well studied and could impact the overall economics of H$_2$ production \cite{Kalinci2015-re,Shibata2015-qt,Hurtubia2021-jo}. Here, we focus on the answering this question in the context of grid-connected proton exchange membrane (PEM) electrolyzers.
As of 2022, global installations of electrolyzers totalled 700 MWe \footnote {Unlike many technologies, electrolyzer capacity is typically reported on the basis of nameplate electricity input rather than \ch{H2} output}, mostly composed of alkaline electrolyzers \cite{International_Energy_Agency2023-wq}. While the share of PEM electrolyzers is relatively small at 30\% of total installed capacity, several appealing attributes compared to alkaline systems are expected to drive their deployment in the future \cite{International_Energy_Agency2023-wq}. In particular: a) their ability to operate at higher current densities ($\geq$ 1 \si{\ampere\per\square \centi\meter}), which allows for smaller stack areas (and possibly lower areal footprint) to produce the same amount of H$_2$, and b) ability to operate with a differential pressure between the anode and cathode, which enables production of high pressure H$_2$ product \cite{Carmo2013-bq, International_Energy_Agency2023-wq}. The greater range of current density operation also creates the potential for increased operational flexibility which is an important criteria for cost-effective electrolyzer projects to manage fluctuations in temporal attributes, such as emissions or cost, of electricity supply \cite{Guerra2019-ua,Corengia2022-ha,Chung2024-vr,Tsay2023-se}. The importance of such operational flexibility for power systems balancing is likely to grow with an increasing share of VRE supply in future low-carbon grids, owing increasing instances of near-zero prices and very high prices compared to today's fossil-fuel dominant systems \cite{Mallapragada_undated-iq}.
Several techno-economic optimization studies have highlighted the importance of dynamic operation in minimizing the operating cost of other electrochemical processes when utilizing electricity sourced from VRE that is co-located or supplied via connection to the electric grid \cite{Otashu2019-ge,Hofmann2022-ti,Weigert2021-ik}. Studies on PEM systems reveal that cost-optimal dynamic operation involves operating at high current densities (say $\gg$ 2 \si{\ampere\per\square \centi\meter}) during times of abundant VRE supply or low electricity prices, as well as periods of idling or near-zero current densities during high electricity prices or low VRE supply periods, that collectively results in smaller stack areas when compared to an equivalent steady state electrolyzer operating at current densities near 1-2 \si{\ampere\per\square \centi\meter} \cite{Espinosa-Lopez2018-hb,Stansberry2020-zm,Chung2024-vr,Corengia2020-on,Corengia2022-ha}.
Dynamic operation of PEM electrolyzers via modulation of the operating current density brings with it several challenges -- namely, heat management, gas crossover and associated safety concerns, and stack degradation. Heat management in PEM systems is typically undertaken via excess feed water on the anode side \cite{Stansberry2020-zm,Holst2021-nd}. In an effort to model a 46 kWe electrolyzer connected to renewable energy, Espinosa et. al developed a model of the temperature response of PEM systems at varying current density with this excess feed water strategy \cite{Espinosa-Lopez2018-hb}. With increasing current densities, the additional heat generated due to the higher overpotential will increase the water flow rates needed for heat management, and could create create implicit constraints on maximum current density levels. At the same time, in the absence of sufficient heat evacuation through increasing feed water flow rate, the cell temperature could rise and and begin vaporizing feed water as well as damage the membrane electrode assembly.
Another operational challenge of dynamic operation is hydrogen crossover to the anode under differential pressure operation, which represents both an energy loss and safety concern. The pressurized cathode provides a driving force for hydrogen to cross the membrane and mix with the oxygen at the anode. At 4\% hydrogen in oxygen, the mixture reaches its lower flammability limit (LFL), presenting serious safety concerns. This crossover is generally greater under differential pressure operation and has been shown to be most problematic at low current densities when there is not enough oxygen production at the anode to keep the hydrogen concentration low \cite{Bernt2020-af}. While the crossover phenomenon has been well documented experimentally, most dynamic techno-economic models of PEM systems exclude any discussion of safety, crossover, or mitigation mechanisms (e.g. recombination catalyst on the anode side) \cite{Christopher_Bryce_Capuano_Morgan_Elizabeth_Pertosos_Nemanja_Danilovic2018-lw,Nicolas_Guillet2014-tq}.
It is common for PEM systems to undergo degradation over time, which requires an increased voltage to the cell for the same current as a result of the increased high frequency resistances in the stack \cite{Hartig-Weis2021-mi}. This degradation is a result of dissolution of catalyst, chemical degradation of the membrane, degradation of the bipolar plate, degradation of the current collector, and manufacturing defects \cite{Feng2017-kz,Weis2019-vj}. The industry standard model for PEM electrolysis techno-economics made by the National Renewable Energy Laboratory and U.S. Department of Energy assumes that the rate of degradation (increase in cell voltage) is independent of operation \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz}. For electrolyzers operating statically (i.e. at constant current density), this constant rate of degradation may hold true. However, experimental studies have shown that degradation rate will be impacted by changes in operating current density \cite{Suermann2019-kz,Rakousky2017-te,Papakonstantinou2020-hj,Alia2019-lw,Li2021-nj,Sun2014-iz,Frensch2019-ff}. To our knowledge, no techno-economic systems of PEMH$_2$ production systems account for this usage-based degradation though it plays a large roll in stack lifetime estimates and therefore replacement costs in these systems. Such operation-dependent degradation may alter the incentives for dynamic operation and consequently impact the levelized cost of hydrogen (LCOH).
Our work seeks to address the above gaps in the techno-economic modeling of PEM electrolyzer systems. Our approach is based on a dynamic optimization framework that co-optimizes the design (stack area and on-site \ch{H2} storage) and operation of the PEM electrolysis-based \ch{H2} production process, with the following unique model elements: a) accounting for dynamic mass and energy balances at 15 min resolution at each electrode as well as flux across the membrane, b) development and use of an empirical correlation for characterizing degradation as a function of current density, which allows us to co-optimize stack replacement times, and c) accounting for safety and temperature related operational constraints and their impact on cost-optimal design and operation of the system. We use the model to investigate the cost-optimal sizing and operation of the PEM system under a range of electricity price scenarios representative of present low-VRE grids and future high-VRE penetration grids, as well as capital cost assumptions for the PEM electrolyzer. Through a systematic scenario-based framework, we isolate the impact of degradation on the levelized cost of hydrogen production. For instance, under 2022 capital cost assumptions for PEM systems, accounting for degradation could increase calculated LCOH by about 45\% due to the use of larger stack sizes (2.3x) to minimize operation at high current densities (and hence degradation) and more frequent stack replacements (every 2 vs. 7 years) also due to degradation. Moreover, accounting for degradation after fixing the design of the system to that obtained without accounting for degradation could increase LCOH by a greater amount of 52\%, which highlights the value of the proposed integrated design and scheduling (IDS) framework.
In general, scenarios that do not consider degradation tend to have a wider current density distribution, size larger \ch{H2} storage, and utilize the electrolyzer more than scenarios that account for degradation. Applying the model to future electrolyzer capital cost and electricity price scenarios reveals several interesting findings. First, levelized costs are estimated around \$2.50/kg dominated by cost of electricity and other fixed operating costs. Second, as stack costs decrease, the initially installed stack size increases, which further reduces instances of high current density operation and results in: a) increase time for stack replacement from 2 to 3-5 years depending on the electricity price scenario and CAPEX assumption and b) reduced capacity of installed \ch{H2} storage (0.1 days of \ch{H2} demand vs. 0.5 or greater in the 2022 scenarios).
Finally, we also investigated the impact of enforcing a safety constraint that limits \ch{H2} build up in the anode to avoid an explosive mixture. We found that without this constraint, the electrolyzer's cost-optimal operation profile tends to favor lower current densities which lowers the overall levelized cost but could result in \ch{H2} concentration in anode exceeding 2\% threshold and 4\% LFL in a few periods.
\subsection{Model Overview}
In order to explore dynamic operation in the context of heat management, safety, and stack degradation, we develop a 0-D model of the electrolyzer process shown in Figure \ref{flowsheet}, which accounts for electrode-level mass balances, cell-level energy balances as well as electrochemical conversion described by the bottom-up characterization of the polarization curve. Below we provide a brief description of the model, with the complete mathematical description given in the supporting information (SI), along with the model nomenclature in Tables \ref{setnomenclature} - \ref{variablenomenclature}.
\begin{figure}
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/Process_Flow.eps} % calls your figure and set the width dimension
\caption{Flowsheet for the current work. Cell level details with flows across the membrane are shown in the cutout. Downstream treatment of the anode gas stream is not considered. The \ch{H2} stream is dried and can either be compressed and stored to satisfy demand at a later time, or directly used to satisfy demand. Stream 11 is a nitrogen purge stream available for the model to use at a cost as a fail-safe should the concentration of \ch{H2} in \ch{O2} reach the 2\% safety limit.}% caption to the figure
\label{flowsheet}
\end{minipage}
\end{figure}
Figure \ref{flowsheet} shows the process flow diagram for the electrolyzer and balance of plant which has been adapted from the literature and is consistent with industrial PEM installations \cite{Holst2021-nd,Espinosa-Lopez2018-hb,Caparros_Mancera2020-rv}. Deionized water is pumped into the electrolyzer where the electrolytic reaction takes place. Gases and excess water leave in streams 3 and 4 and are sent to flash drums to separate the vapor from the liquid. It is assumed that a negligible amount of gas is dissolved in the liquid water leaving the flash drum. Water from the flash drum is cooled and recycled back into the electrolyzer stack. The \ch{H2} in stream 6 is dried and then either compressed into \ch{H2} storage or used to satisfy the \ch{H2} demand. Since the focus of the study is on H$_2$ production, the treatment of the anode gas stream, primarily composed of O$_2$, is not considered for this study. Stream 11 is a nitrogen purge stream that is modeled as a backstop to ensure \ch{H2} concentration in the anode stays well below the lower flammability limit. This stream is only used as a fail safe to make sure an explosive mixture is not formed at the anode.
\subsection{General Optimization Formulation}
Given a fixed demand for \ch{H2}, the system as described and the model presented in the SI can be used to minimize the total cost of a PEM electrolyzer system. The general form for such an optimization is shown in Equation \ref{gen_opt_func}.
\begin{equation}
\begin{aligned}
\min _{x,y} \quad & PV = C^{stack}(x) + C^{storage}(x) + C^{vOPEX}(x,y) \\
\textrm{s.t.} \quad & f(x,y) =0 \\
& g(x,y) \leq 0 \\
& x \in \mathbb{R}^{n}, y \in \mathbb{R}^m
\end{aligned}
\label{gen_opt_func}
\end{equation}
The objective function is the present value of the system which is a sum of the cost of the electrolyzer stack, $C^{stack}$, the cost of \ch{H2} storage, $C^{storage}$, and the variable operating costs, $C^{vOPEX}$, which are primarily the cost of electricity and deionized water. The decision variables can be separated into two classes, $x$ which represents the decision variables related to electrolzyer and storage sizing, and $y$, representing the time-dependent operational variables calculated from the relations in $f$ and constraints represented by $g$. A detailed description of the optimization problem is presented in the following sections.
\subsection{Reformulated Model}
Rather than solve a single monolithic non-convex optimization model which is computationally challenging, we apply a decomposition strategy similar to Roh et. al and Chung et. al \cite{Roh2019-vj, Chung2024-vr}. This approach decouples the sizing of the electrolyzer (the outer problem) and the operational optimization given a fixed electrolyzer area and storage capacity (the inner problem) as illustrated in Figure 2 (explained in a later section).
\subsubsection{Outer Problem}
The outer problem iterates over the total number of cells in the electrolyzer set up, $N_c$, and number of days of \ch{H2} storage $\Gamma_{\ch{H2},stor}$. Each cell is assumed to have an area of 450 \si{\square\centi\meter} represented by $A_{cell}$. The cost of the stack and storage can then be calculated as
\begin{align}
C^{stack} &= N_{c}A_{cell}p^{stack} + C^{mBoP} + C^{eBoP} \\
C^{storage} &= \Gamma_{\ch{H2},stor} \dot{m}_{\ch{H2}}p^{storage}
\end{align}
where $C^{mBoP}$ mechanical balance of plant capital cost, $C^{eBoP}$, electrical balance of plant capital cost, $p^{stack}$ is the capital cost of the stack in \$/\si{\square\centi\meter}, $\dot{m}_{\ch{H2}}$ is the daily \ch{H2} production rate in kilograms as calculated by the inner problem, and $p^{storage}$ is the cost of storage in \$/\si{\kilo\gram} \ch{H2}.
\begin{align}
C^{mBoP} &= \alpha^{mBoP} \dot{m}_{\ch{H2}} \label{mBoP} \\
C^{eBoP} &= \alpha^{eBoP}P^{max} \label{eBoP}
\end{align}
The mechanical and electrical balance of plant are scaled based on the rate of \ch{H2} production and peak power consumption for the facility, as shown in Equations \ref{mBoP} and \ref{eBoP}. The mechanical balance of plant represents the compressors, pumps, DI water system, and \ch{H2} separation system while the electrical balance of plant includes all electrical components necessary to make the interconnection to the grid including a rectifier. Here, $\alpha^{mBoP}$ is a cost factor in \$/\si{\kilo\gram} \ch{H2}, $\alpha^{eBoP}$ is a cost factor in \$/\si{\kilo\watt}e, and $P^{max}$ is peak power consumption of the system in \si{\kilo\watt}e as calculated by the inner problem. $P_{max}$ is then a parameter passed to the outer problem to calculate the CAPEX for a given outer problem iteration.
\subsubsection{Inner Problem}
The inner problem takes a fixed $N_c$ and $\Gamma_{\ch{H2},stor}$ and minimizes operating cost while meeting the daily production target of $50,000 \si{\kilo\gram\per\day}$. The inner problem can be written as a minimization of the operating cost as defined in Equation \ref{innerproblem}.
\begin{equation}
\begin{aligned}
\min_{\bf{x}} \quad &C^{vOPEX} = C^{elec} + C^{BoP, \:elec} + C^{water} + C^{\ch{N2}} \\
\textrm{s.t.} \quad & \textrm{electrochemical model, Equations \ref{totvolt} -\ref{conduct}}\\
& \textrm{mass balances, Equations \ref{wateranmol1}-\ref{hydcat2}}\\
& \textrm{energy balance, Equations \ref{energybalance} - \ref{energyloss}} \\
& \textrm{\ch{H2} storage constraints, Equation \ref{deltaequation}-\ref{massrep}}\\
& \dot{m}_{\ch{H2}} \geq 50,000 \si{\kilo\gram\per\day} \\
& 60^{\circ}\text{C} \leq T \leq 80^{\circ}\text{C} \\
& 0.1 \si{\ampere\per\square\centi\meter} \leq i \leq 4 \si{\ampere\per\square\centi\meter} \\
& y_{\ch{H2},anode} \leq 0.02
\label{innerproblem}
\end{aligned}
\end{equation}
where \textbf{x} represents the primary decision variables, $C^{elec}$ is the annual cost of electricity used by the electrolyzer, $C^{BoP,elec}$ is the annual cost of electricity consumed by the balance of plant, $C^{water}$ is the annual cost of the deionized water, and $C^{\ch{N2}}$ is the annual cost of the nitrogen gas used at the anode for purging.
The model is subject to the mass and energy balances outlined in the SI as well as a demand constraint of 50,000 \si{\kilo\gram\per\day} of \ch{H2}. An operating temperature range of $60^{\circ}$C - $80^{\circ}$C was imposed on the stack. Additionally, a minimum and maximum current density were applied in this study. The minimum current density was chosen based on minimizing degradation at low voltages and the high current density was chosen based on the limit of commercialized PEM today \cite{Weis2019-vj}. In order to ensure 4\% \ch{H2} in \ch{O2} at the anode is never reached a safety factor of 2 was applied making another constraint of a maximum of 2\% \ch{H2} in \ch{O2} at the anode.
Evaluating each variable in the system for 15 minute time periods over the course of a year quickly becomes computationally intractable. To maintain computational tractability, we approximate annual operating costs through modeling operation over representative days at 15 min resolution assuming the hourly price of electricity holds for the entire hour. Time domain reduction using k-means clustering was performed for each electricity price time series data in order to find 7 representative days. Further details of the clustering method can be found in SI \ref{clustering description}.
The clustering created a mapping $f: d \rightarrow r$. Each real day, $d$, is mapped to a representative day, $r$. To distinguish between variables for real or representative days, any variable with a bar above it is a variable that is indexed by the set of representative days.
Beginning with the cost of electricity for the electrolyzer,
\begin{align}
C^{elec} &= \sum_{d=1}^{365} C^{elec}_d\label{cost of elect}\\
C^{elec}_d &= \int_{0}^{\tau} p^{elec}_{d}(t) i_{d}(t) A \left(V^{undeg}_d(t) + V^{deg,cuml}_d(t)\right) \: dt
\end{align}
where $\tau$ represents the end of a representative day, $p^{elec}$ represents the price of electricity, $i$ is the current density, $A = N_{c}A_{cell}$ is the total electrolyzer area, $V^{undeg}$ is the voltage without considering degradation (based on polarization curve - see Equations \ref{totvolt} -\ref{conduct}), and $V^{deg,cuml}$ is the cumulative increase in voltage due to degradation.
The first term in the integral can be mapped to its respective representative day via $f(d) \rightarrow r$. However, the second term that accounts for the cumulative degradation that changes over the 365 day range and is inter-temporally coupled. So $C^{elec}_d$ becomes:
\begin{equation}
C^{elec}_d = \int_{0}^{\tau} p^{elec}_{f(d)}(t) i_{f(d)}(t) A \left(V^{undeg}_{f(d)}(t) + V^{deg,cuml}_{d}(t)\right) \: dt
\end{equation}
In accordance with the degradation correlation based on experiential literature values seen in Figure \ref{deg_curve}, we use the following model to represent the change in degradation over a representative day $r$:
\begin{equation}
\frac{d\bar{V}_{deg,r}}{dt} = \begin{cases}
a \label{deg} \quad &\bar{i}_r \leq 1 \\
a \left[\bar{i}_{r}(t)\right]^2 \quad &\bar{i}_r > 1
\end{cases}
\end{equation}
Cumulative degradation for real day $d$ can then be written as
\begin{equation}
V_{d}^{deg,cuml}(t) = V_{d-1}^{deg,cuml}(t) + \delta \bar{V}_{f(d)}^{deg}(t)
\end{equation}
where $\delta \bar{V}_{deg,f(d)}(t)$ comes from solving the differential equation in Equation \ref{deg} up to the time point of interest. Initial degradation at the beginning of the year for a fresh stack is assumed to be 0 \si{\volt}.
Using the mapping $f(d)\rightarrow r$, $C^{elec}_d$ can be written
\begin{equation}
C^{elec}_d = \int_{0}^{\tau} \bar{p}^{elec}_{r}(t) \bar{i}_{r}(t) A \left(\bar{V}^{undeg}_{r}(t) + V_{d-1}^{deg,cuml}(t) + \delta \bar{V}_{r}^{deg}(t)\right) \: dt
\end{equation}
The annual cost of electricity is then written:
\begin{equation}
C^{elec} = \sum_{r=1}^7 w_{r} \int_{0}^{\tau} \bar{p}^{elec}_{r}(t) \bar{i}_{r}(t) A \left(\bar{V}^{undeg}_r(t) + \delta \bar{V}_{r}^{deg}\right) \: dt + \sum_{d=1}^{365} \int_{0}^{\tau} \bar{p}^{elec}_{r}(t) \bar{i}_{r}(t) A V_{d-1}^{deg,cuml}(t) \: dt
\end{equation}
where $w_r$ represents the weight of representative day $r$ as determined by the clustering.
$C^{BoP,elec}$, $C^{water}$, and $C^{\ch{N2}}$ can be calculated in a similar manner using the associated representative days and mapping $f$.
\begin{align}
C^{BoP,elec} &= \sum_{r=1}^7 w_{r} \int_{0}^{\tau} \left[ \alpha^{mBoP} \dot{m}_{\ch{H2},r} \bar{p}^{elec}_{r}(t) + W^{compress}\bar{p}^{elec}_{r}(t) \right] dt \\
C^{water} &= \sum_{r=1}^7 w_{r} \int_{0}^{\tau} \left[ \dot{n}_{\ch{H2O},consum,r} + \dot{n}_{\ch{H2O},vap,5,r} + \dot{n}_{\ch{H2O},vap,6,r}\right]p^{water} dt \\
C^{\ch{N2}} &= \sum_{r=1}^7 w_{r} \int_{0}^{\tau} \dot{n}_{\ch{N2},11,r} p^{\ch{N2}} dt
\end{align}
where $W^{compress}$ is the work of the \ch{H2} storage compressor, $p^{water}$ is the price of deionized water, and $p^{\ch{N2}}$ is the price of nitrogen. Compressor work is evaluated using a single stage isentropic compression with an isentropic factor of 0.7 as outlined in Kahn et. al and Chung et. al \cite{Chung2024-vr,Kahn2021-jv}.
\subsubsection{H$_2$ Storage}
\ch{H2} storage is necessary to enable meeting base load \ch{H2} demand with flexible electrolyzer operation. We model the state of charge variation in \ch{H2} storage considering both the short-term (intra-day) and long-term (inter-day) changes resulting from hour-to-hour discharging/charging activity. Following the method of Narayanan et. al \cite{Narayanan2022-uj} and Kotzur et. al \cite{Kotzur2018-rx}, the storage state of charge variation throughout the year is approximated as as a super-position of: a) inter-day state of charge variations modeled over all days of the year and b) intra-day state of charge variations modeled only for the representative days. This approach enables us to respect inter-temporally coupled nature of energy storage operations without the need to model all time periods of the year.
Long duration storage as well as intra-day storage are formulated as follows:
\begin{align}
\delta_{r} &= \bar{\Lambda}^{\ch{H2},stor}_r(\tau) - \bar{\Lambda}^{\ch{H2},stor}_r(0) \quad &\forall r \in D_{r} \label{deltaequation}\\
\Lambda^{\ch{H2},stor}_1(0) &= \Lambda^{\ch{H2},stor}_{365}(\tau) + \delta_{f(365)} \label{wraparound}\\
\Lambda^{\ch{H2},stor}_{d+1}(0) &= \Lambda^{\ch{H2},stor}_{d}(0) + \delta_{f(d)} \label{realdaycarry}\\
\Lambda^{\ch{H2},stor}_{d}(0) &= \bar{\Lambda}^{\ch{H2},stor}_{r}(\tau) - \delta_{f(d)} \quad &\forall d \in D \label{realtorep}\\
\frac{d\Lambda^{\ch{H2},stor}_{d}}{dt} &= \dot{n}_{15,f(d)}(t) - \dot{n}_{16,f(d)}(t) \label{massreal}\\
\frac{d\bar{\Lambda}^{\ch{H2},stor}_{r}}{dt} &= \dot{n}_{15,r}(t) - \dot{n}_{16,r}(t) \label{massrep} \\
\Gamma_{\ch{H2},stor} \dot{m}_{\ch{H2}} \label{max_storage} &\geq \max\left(\bar{\Lambda}^{\ch{H2},stor}_r,\Lambda^{\ch{H2},stor}_{d}\right)
\end{align}
where $\delta_{r}$ is the positive or negative change in storage levels that occurs over representative day $r$ and allows for \ch{H2} storage to carry over between representative periods. Equation \ref{wraparound} represents a wrapping constraint between the last day in the year and the first day in the year. Equation \ref{realdaycarry} allows \ch{H2} storage to carry over between real days. Equation \ref{realtorep} relates the beginning of a real day of storage to the associated beginning of the representative day. Equations \ref{massreal} - \ref{massrep} ensure the mass balance across the \ch{H2} storage is closed over real and representative days. Finally, Equation \ref{max_storage} ensures all storage state of charge variables are less than the maximum amount of \ch{H2} storage. A visual representation of the storage formulation is included in the SI in Figure \ref{storage_form}.
\subsubsection{Implementation Approach}
The outer and inner problems can be combined in an iterative manner using the algorithm described in Figure \ref{gss}.
\begin{figure}
\centering
\begin{minipage}[c]{0.7\textwidth}
\includegraphics[width=\textwidth]{Figures/GSS_Flow_Diagram.eps} % calls your figure and set the width dimension
\caption{Summary of the bilevel optimization. The outer problem iterates over number of cells and \ch{H2} storage capacity. The inner problem solves the cost optimal variable operation problem given a fixed number of cells and storage from the outer problem.}% caption to the figure
\label{gss}
\end{minipage}
\end{figure}
Upper and lower bounds for the electrolyzer area and number of days of \ch{H2} storage are chosen based on physical limits of the electrolyzer in order to satisfy the 50,000 \si{\kilo\gram\per\day} \ch{H2} demand. For this study, an area range of $40,000 - 300,000$ 450 \si{\square\centi\meter} cells and a storage range of $0.1-14$ days were chosen where one day of storage is equivalent to 50,000 \si{\kilo\gram} of \ch{H2}. From the GSS algorithm two trial areas and two trial storages are chosen as is done by Sandhya et al assuming unimodality in each of the search directions \cite{Sandhya-Rani2019-pw}. This creates a total of four trial points in this 2D search space. The operational optimization is run for those 4 trial points which results in four different variable operating costs. For the PEM system, in the limit of very large electrolyzer areas, high capital costs would result in a high present value. In the limit of small areas, the system would need to operate at high current densities which would result in more stack replacement and a high present value. Storage also exhibits a similar unimodal trend at the extremes.
With the size of each system and the variable operating cost estimated, the present value for each of the four trial points can be calculated. If the trial areas and number of storage days are within the user defined tolerance (0.1\% in this study), then the algorithm terminates and the final solution corresponds to the solution of trial point problems with the lowest present value. The variable operating routine from the associated minimum present value is then the optimal operation routine and the area and number of days of storage from the minimum present value are the optimal sizing of the system.
If still outside the tolerance, regions in the 2D space where the minimum cannot lie are eliminated. For example, for $A$ and $\Gamma_{\ch{H2},stor}$ where $A_{lb} \leq A \leq A_{ub}$ and $\Gamma_{\ch{H2},stor,lb} \leq \Gamma_{\ch{H2},stor} \leq \Gamma_{\ch{H2},stor, ub}$ and given four present values that are functions of the two trial areas and two trial storages $PV_A (A_{1},\Gamma_{\ch{H2},stor,1}), PV_B (A_{2},\Gamma_{\ch{H2},stor,1}), PV_C (A_{1},\Gamma_{\ch{H2},stor,2}), PV_D (A_{2},\Gamma_{\ch{H2},stor,2})$, we can eliminate a portion of the search region by finding the minimum of the present values. If $PV_A$ is the smallest present value, we know the minimum will not lie in $A_{2} \leq A \leq A_{ub}$ and $\Gamma_{\ch{H2},stor,2} \leq \Gamma_{\ch{H2},stor} \leq \Gamma_{\ch{H2},stor, ub}$. Thus $A_2$ and $\Gamma_{\ch{H2},stor,2}$ become the new upper bounds. Similar eliminations can be made if the smallest present value is one of the other three values. The algorithm repeats until converged on a locally optimal solution. Due to the non-linearity of the problem, a global optimum is not guaranteed.
The NLP to solve the operational problem was implemented in Python using Pyomo, specifically Pyomo DAE with IPOPT as the non-linear solver \cite{Nicholson2018-mw,Wachter2006-la}. Linear solver MA86 was chosen to use in IPOPT for its ability to handle large problems well \cite{UnknownUnknown-hg}. The bilevel optimization approach with the degradation correlation results in a problem with $\approx 200,000$ continuous variables, $\approx 200,000$ equality constraints, and $\approx 2,500$ inequality constraints. Full details of each run can be found in Table \ref{results} and in the SI in Table \ref{convergence_time}. The problem was run using resources on MIT's SuperCloud HPC \cite{Reuther2018-xn}.
Table \ref{algo_scaling} shows how the bi-level optimization approach scales with the number of representative days. As we increase the number of representative days we expect convergence to a value for storage, number of cells, and LCOH, although this will not be monotonic due to the different effect of each added day to the optimization. Although increasing number of representative days allows for capturing greater day-to-day variability in the underlying electricity price time series, it increases the overall problem size and the computational burden as indicated by the time to convergence and avg. run time per GSS iteration. We choose 7 representative days to strike a balance between computational tractability and accuracy.
\subsection{Data Inputs}
\subsubsection{Techno-economic Data}
For the 2022 case, capital and operating cost assumptions were made consistent with the 2019 version of NREL's H2A model adjusting for inflation from 2019 to 2022 \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz}. The 2030 case capital costs come from Fraunhofer ISE's cost forecast for PEM and alkaline electrolysis \cite{Holst2021-nd}. Summary of the cost assumptions can be found in Table \ref{technoecon}.
\begin{table}
\caption{Techno-economic parameters for model. Consistent with NREL's H2A Model, Chung et. al, and Fraunhofer ISE \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz,Chung2024-vr,Holst2021-nd}. Planned replacement costs are used as the stack replacement cost when a replacement is necessary. }.
\centering
\begin{tabular}{lcccc}
\toprule
Techno-economic Parameter & 2022 & 2030 & Units \\ \midrule
Stack CAPEX & \$2.37 & \$0.79 & \$/cm$^2$ \\
BoP CAPEX & 289 & 103 & \$/kWe \\
Storage CAPEX & \$500 & \$300 & \$/kg \ch{H2} \\
Site Prep & \multicolumn{2}{c}{2\%} & of direct capital\\
Engineering & \multicolumn{2}{c}{10\%} & of direct capital \\
Contingency & \multicolumn{2}{c}{15\%} & of direct capital \\
Permitting & \multicolumn{2}{c}{15\%} & of direct capital \\
Planned Replacement & \multicolumn{2}{c}{15\%} & of direct capital \\
Unplanned Replacement & \multicolumn{2}{c}{0.5\%} & of direct capital \\
Overhead & \multicolumn{2}{c}{20\%} & of labor cost \\
Tax/Insurance & \multicolumn{2}{c}{2\%} & of total CAPEX \\
BoP Electricity Usage & \multicolumn{2}{c}{5.1} & kWh/kg \ch{H2} \\ \bottomrule
\end{tabular}%
\label{technoecon}
\end{table}
Using the parameters in Table \ref{technoecon}, capital cost ($C^{CAPEX} = C^{stack} + C^{storage}$), unplanned replacement costs ($C^{UPR}$), planned replacement costs ($C^{PR}$), fixed operating cost ($C^{fOPEX}$), and variable operating cost ($C^{vOPEX}$) can be estimated. Assuming a plant life of 40 years and a discount rate of 8\%, the present value of the project ($PV$) can be estimated as shown in Equation \ref{pv}.
\begin{equation}
PV = C^{stack} + C^{storage} + \sum_{y=1}^{40}\left[ C^{UPR} +C^{PR} + C^{fOPEX} + C^{vOPEX}\right] \frac{1}{(1+d)^y} \label{pv}
\end{equation}
The direct capital needed to calculate the above costs is the sum of the cost of the stack and cost of the balance of plant. For the scenario run without useage-dependent stack degradation, the replacement rate for the stack was fixed at 7 years \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz}. For the scenarios where degradation is accounted for as shown in Equation \ref{deg}, a degradation rate-informed replacement rate is calculated as follows. Based on the fixed H2A degradation rate, approximately $\Delta^{max}_{deg} \approx 1 \si{\volt}$ of degradation is allowed to occur before the stack needs to be replaced \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz}. The operational optimization of the problem gives the degradation of a fresh stack after one year of operation ($\Delta^{1}_{deg}$). The replacement rate can then be defined as:
\begin{equation}
r_{stack} = \begin{cases}
\left\lfloor \frac{\Delta^{max}_{deg}}{\Delta^{1}_{deg}}\right\rfloor \quad &\Delta^{1}_{deg} \leq 1 \\
1 \quad &\Delta^{1}_{deg} > 1
\end{cases}
\end{equation}
For implementation purposes, if the degradation of the stack exceeds 1 V in a given year, the stack is allowed to run until the end of the year before being replaced.
Fixed operating cost is the sum of the cost of labor, overhead, tax and insurance, and material. The labor rate was assumed to be \$70 per hour per worker where 10 workers are needed for a system of this size, and the plant is assumed to operate 350 days per year.
Variable operating cost for the first year is simply obtained from the objective function of the operational optimization. Post-convergence of the operational optimization, the variable operating cost is updated to include degradation for the years leading up to replacement. This assumes that degradation in each future year is the same as the first year after stack replacement.
The amount of \ch{H2} produced in one year ($M_{\ch{H2}}$) can be discounted in a similar way as the present value of the project as seen in Equation \ref{h2pv}. The levelized cost of \ch{H2} is then calculated as the ratio of the two present values.
\begin{equation}
PV_{\ch{H2}} = \sum_{y=1}^{40} \frac{M_{\ch{H2}}}{(1+d)^y}
\label{h2pv}
\end{equation}
\begin{equation}
LCOH = \frac{PV}{PV_{\ch{H2}}}
\end{equation}
\subsubsection{Electrolyzer Performance and Model Assumptions}
Validation of the electrochemical model without degradation at different temperatures and operating pressures is shown in Figure \ref{polarcurves}. The model fits well across varying current densities, pressures, and temperatures.
\begin{figure}
\subcaptionbox*{}[0.45\linewidth]{
\includegraphics[width=\linewidth]{Figures/1bar76C.eps}}
\hfill
\subcaptionbox*{}[0.45\linewidth]{
\includegraphics[width=\linewidth]{Figures/HighP.eps}}
\caption{Left Figure: Current voltage relationship at 1 bar and 76$^{\circ}$C. Data from \cite{Lee2020-db}. Fitted parameters $\alpha_{an} = 0.58$, $\alpha_{cat} = 1.28$. Right Figure: Current voltage relationship at high pressure. Data from \cite{Marangio2011-jc}. Fitted parameters at 40$^\circ$C: $\alpha_{an} = 1.9$, $\alpha_{cat} = 0.1$, fitted parameters at 55$^\circ$C: $\alpha_{an} = 1.38$, $\alpha_{cat} = 0.11$ }
\label{polarcurves}
\end{figure}
Many degradation studies have been done on PEM cells that attempt to characterize the degradation rate of the cell as a function of the applied current density \cite{Suermann2019-kz,Rakousky2017-te,Papakonstantinou2020-hj,Alia2019-lw,Li2021-nj,Frensch2019-ff}. We use data from these studies to construct a usage-based degradation rate. The operating current density and degradation rate have been averaged over the lifetime of the cell and only square and hold wave patterns were taken from the degradation studies \cite{Suermann2019-kz,Rakousky2017-te,Papakonstantinou2020-hj,Alia2019-lw,Li2021-nj,Frensch2019-ff}. Additionally some studies exhibited a negative degradation rate likely to a decrease in ohmic losses from membrane thinning. For this work, those data points were not considered.
Figure \ref{deg_curve} illustrates the points taken from literature, and Equation \ref{deg_corr} shows the correlation found. The degradation rate increases linearly with the square of the current density for $i > 1 \si{\ampere\per\square\centi\meter}$. The experiments at low current density do not follow this linear trend but instead have a relatively constant degradation rate. The degradation rate was thus modeled as a piece-wise function shown in Equation \ref{deg_corr}. The constant in the equation is a fitted parameter from the data.
Pushing electrolyzers to higher and higher current densities past the 4 \si{\ampere\per\square\centi\meter} maximum presented in this work makes the economics more favorable when degradation is ignored \cite{Chung2024-vr}. Bench scale experimentalists have already begun to push up to 10 \si{\ampere\per\square\centi\meter} \cite{Martin2022-ek}. However, operating at higher and higher current densities could potentially result in increased degradation rates if one assumes the extrapolation of the trend in Figure \ref{deg_curve}.
\begin{figure}
\centering
\begin{minipage}[c]{\textwidth}
\centering
\includegraphics[width=0.7\textwidth]{Figures/deg_correlation.eps} % calls your figure and set the width dimension
\caption{Degradation rate as a function of the square of the current density plotted on a log-log axis. Data taken from \cite{Suermann2019-kz,Rakousky2017-te,Papakonstantinou2020-hj,Alia2019-lw,Li2021-nj,Frensch2019-ff}. Current density and degradation are time averaged over the operation lifetime. Points shown in the graph here are for square and hold wave patterns. Some cells showed negative degradation rates likely due to membrane thinning; these negative degradation rates were not considered for this study.}% caption to the figure
\label{deg_curve}
\end{minipage}
\end{figure}
\begin{equation}
\frac{dV_{deg}}{dt} = \begin{cases}
30 \quad &i \leq 1 \\
30 \left[i(t)\right]^2 \quad &i > 1 \label{deg_corr}
\end{cases}
\end{equation}
\subsubsection{Electricity Price Scenarios}
Corpus Christi, Texas was used as a case study for this work due to its proximity to existing \ch{H2} demand from the petrochemical sector on the Texas Gulf cost which has prompted commercial interest to deploy GW-scale electrolyzers in that region \cite{Arjona2023-bc}. In order to consider the impact of current and future electricity price scenarios, historical electricity prices from the years 2022 and price projections available for 2030 were used. Electricity prices were taken from 2022 ERCOT South load zone historical pricing \cite{noauthor_undated-so}. For the 2030 scenario, prices were taken from NREL's Cambium future electricity prices model mid-case for Texas \cite{Gagnon_undated-so}. These projected future prices represent a moderately decarbonized grid by 2030. Prices are shown in the left panel of Figure \ref{price_comparison_future}. The average electricity price is higher in 2022 than in 2030. However the 2030 price scenario exhibits more volatile prices.
\begin{figure}
\subcaptionbox*{}[0.5\linewidth]{
\includegraphics[width=\linewidth]{Figures/prices_mid_south_west.eps}}
\hfill
\subcaptionbox*{}[0.5\linewidth]{
\includegraphics[width=\linewidth]{Figures/prices_future_scenarios.eps}}
\caption{Left Figure: Duration curves for 2022 and 2030 prices used in this study. The south load zone (avg. price: \$62.55) and west load zone (avg. price: \$63.83) prices are for 2022. The 2030 case is the mid-case for NREL's Cambium model for 2030 Texas (avg. price: \$21.43). Right Figure: Duration curves for 2030 prices used in this study. The low decarbonization scenario projects 95\% decarbonization of the grid by 2050 (avg. price: \$21.97). The high decarbonization scenario projects 100\% decarbonization of the grid by 2030 (avg. price: \$22.67).}
\label{price_comparison_future}
\end{figure}
In order to show the model's sensitivity to 2022 electricity prices, a case with electricity prices from the West Texas load zone in 2022 (another location with many planned electrolyzer projects) was run. For sensitivity to 2030 electricity prices, the Cambium model gives a scenario where the grid is 100\% decarbonized by 2030 (high decarbonization) and 95\% decarbonized by 2050 (low decarbonization). The duration curves for these price series in comparison to the mid-case are shown in the right panel of Figure \ref{price_comparison_future}.
\subsubsection{Limitations of Model}
While this model accounts for many of the aspects of dynamic operation including usage-based degradation, it does have some limitations. In this work we solve a non-linear, non-convex optimization with a local solver IPOPT. There is no guarantee we have converged on the global solution, so a lower cost solution could exist. However, any local minimum will be better than operating naively or statically and a global solution is not strictly necessary in practice.
Additionally, we model the economics of the system as a price-taker. In other words, we must accept the price of electricity, and our operation has no influence on the prices of electricity. In reality, especially with increasing demand on the grid for electrified industrial processes, dynamic operation to best utilize cheap electricity prices would influence the local cost of electricity, as shown by other studies \cite{Tsay2023-se,He2021-kv}.
In order to demonstrate the sensitivity of the model to capital cost assumptions, electricity price assumptions, and other constraint assumptions, various test cases were run. All scenarios contain the degradation correlation unless ``no use-dependent degradation" is indicated. Capacity factor for the scenarios is calculated by taking the ratio of the energy actually used by the system and the energy used by the system if it were to operate at the maximum current density of 4 \si{\ampere\per\square\centi\meter}. A summary of the tests run with high level results are in Table \ref{results}.
\begin{landscape}
\begin{table}
\centering
\caption{High level results of electrolyzer design and scheduling optimization. One day of storage is defined as 50,000 \si{\kilo\gram} of \ch{H2}. Replacement rate is defined as the maximum degradation allowed (1 V) divided by the degradation after 1 year. Annual electrolyzer utilization defined based on maximum current density operation for the 350 working days of the year considered in this model. Convergence time is reported in Table \ref{convergence_time}.}
\resizebox{\columnwidth}{!}{
\begin{tabular}{@{}lccccccccc@{}}
\toprule
Case & Price Series & Total CAPEX (in thousands) & vOPEX (in thousands) & LCOH (\$/kg) & Days of Storage & Number of Cells (in thousands) & Degradation after 1 Year & Replacement Rate (Years) & Utilization \\ \midrule
Degradation 2022 Case & 2022, South LZ & \$ 366,800 & \$ 45,400 & \$ 6.60 & 0.51 & 116.2 & 0.45 & 2.2 & 25.8\% \\
No Use-Dependent Degradation & 2022, South LZ & \$ 203,900 & \$ 50,900 & \$ 4.56 & 1.39 & 50.1 & -- & 7 & 70.1\% \\
Degradation, West Texas & 2022, West LZ & \$ 430,800 & \$ 46,700 & \$ 7.08 & 0.82 & 141.8 & 0.31 & 3.2 & 20.5\% \\
Degradation, Fixed CAPEX & 2022, South LZ & \$ 239,100 & \$ 67,300 & \$ 6.92 & 1.39 & 50.1 & 1.97 & 1.0 & 67.5\% \\
High Temperature & 2022, South LZ & \$ 362,000 & \$ 45,600 & \$ 6.56 & 0.84 & 110.5 & 0.50 & 2.0 & 27.0\% \\
No Safety Constraint, Fixed CAPEX & 2022, South LZ & \$ 315,200 & \$ 47,500 & \$ 6.21 & 0.51 & 116.2 & 0.41 & 2.4 & 25.2\% \\
Degradation 2030 Mid-Case & 2030, Texas Mid & \$ 200,000 & \$ 9,400 & \$ 2.47 & 0.20 & 156.8 & 0.25 & 4.0 & 18.2\% \\
2030 High CAPEX & 2030, Texas Mid & \$ 238,300 & \$ 9,400 & \$ 2.76 & 0.65 & 159.1 & 0.24 & 4.1 & 18.1\% \\
2030 Low CAPEX & 2030, Texas Mid & \$ 181,900 & \$ 9,400 & \$ 2.28 & 0.11 & 177.8 & 0.19 & 5.1 & 15.8\% \\
2030 High Decarbonization & 2030, Texas High & \$ 202,300 & \$ 9,400 & \$ 2.48 & 0.13 & 161.3 & 0.25 & 4.0 & 17.8\% \\
2030 Low Decarbonization & 2030, Texas Low & \$ 180,400 & \$ 11,400 & \$ 2.50 & 0.11 & 132.1 & 0.33 & 3.0 & 22.1\% \\ \bottomrule
\end{tabular}
}
\label{results}
\end{table}
\end{landscape}
\subsection*{Degradation vs No Use-Dependent Degradation}
\begin{figure}
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/present_deg_no_deg.eps} % calls your figure and set the width dimension
\caption{Impact of modeling use-dependent degradation on optimal electrolyzer operation and temperature profile. Results correspond to 2022 ERCOT South Load Zone price scenario. Two representative days, one from early in the year and one from late in the year, of operation are shown. In general, the case with degradation operates at lower current densities than the case without degradation. During the highest price periods of the day both cases idle at 0.1 \si{\ampere\per\square\centi\meter}.}% caption to the figure
\label{present_results}
\end{minipage}
\end{figure}
Figure \ref{present_results} highlights the operational impact of incorporating current density dependent degradation in the techno-economic optimization model. Both scenarios produce optimal operating routines that ramp up current density (and therefore production of \ch{H2}) during periods of low prices and switch to idling at 0.1 \si{\ampere\per\square\centi\meter} (the minimum assumed current density) during periods of high electricity prices \cite{Weis2019-vj}. However, because the model with degradation is penalized by accumulating degradation, it operates at much lower current densities in order to minimize the cost of electricity while meeting the demand constraint of 50,000 \si{\kilo\gram\per\day} of \ch{H2}. When electricity prices start to rise, both models switch to the idling state.
The overall effect of degradation on the operation profile can be seen from the current density distribution in Figure \ref{present_currents}. The accumulating degradation causes an increased power demand which forces the model with degradation to favor lower current densities on average, including idling as much as possible while meeting the demand constraint. The net result is that electrolyzer stack utilization (or capacity factor) with degradation is 26\% as compared to 70\% in the case without degradation.
\begin{figure}
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/present_deg_no_deg_hist.eps} % calls your figure and set the width dimension
\caption{Cost-optimal current density, storage, and temperature distributions for the cases with and without use-dependent degradation. Accounting for degradation moves the distribution leftwards towards lower average current densities. Results corresponds to model optimization based on 2022 ERCOT South Load Zone scenarios.}% caption to the figure
\label{present_currents}
\end{minipage}
\end{figure}
Additionally, due to the increased overpotential from degradation, the case that accounts for degradation operates for more many more hours at the maximum temperature limit than the case without degradation. This is for two reasons: 1) the increased overpotential generates more heat in the system and would result in a higher temperature for the same inlet water flowrate, and 2) operating at a higher temperature lowers the activation and ohmic overpotentials of the cell that partially offset the impact of degradation.
Given today's electricity prices and capital costs for the electrolyzer stack, accounting for degradation results in a 45\% higher LCOH than when use-dependent degradation of the stack is ignored (see Figure \ref{LCOH_final}). The case with no use-dependent degradation has a LCOH that agrees well with the work done by Chung et. al although our model accounts for crossover, static degradation, and the energy balance over the stack \cite{Chung2024-vr}. The case without use-dependent degradation also sized 2.7 times more \ch{H2} storage than the case with it. Intuitively, when use-dependent degradation is not accounted for, the model is incentivized to maximize current density (and production) during periods of low electricity prices and turn down production during periods of high prices. This "bang-bang" operating pattern results in larger need for energy storage. With degradation, there is a reduced incentive for operating at high current densities and thus storage requirements are lower. Overall use-dependent degradation results in an overpotential of 0.45 V/year that shortens stack replacement to 2.2 years vs. 7 years without modeling use-dependent degradation. A second location in West Texas was also run to show this dynamic operation approach works in more than one electricity market (see Table \ref{results}).
To demonstrate the value of the IDS optimization approach, we ran an operational optimization with use-dependent degradation using the size of stacks and storage from the no use-dependent degradation case. The operational profile can be seen in Figure \ref{fixed_case}. In this highly constrained scenario, the model with degradation is forced to operate at higher current densities much like the case without degradation. Yet the effect of degradation is seen by the model avoiding operation at maximum current density of 4 A/cm$^2$. As a result, the system accumulates 1.97 \si{\volt} of degradation in one year. Consequently, stacks need to be replaced every year at the end of the year. This increases the levelized cost to \$6.92/kg which is a 4.6\% increase over the case with degradation and a 52\% increase over the case without usage-based degradation.
To test the sensitivity of the levelized cost on the degradation correlation and the replacement voltage threshold, two other sensitivity cases were run and compared with the 2022 degradation case: one where the coefficient in Equation \ref{deg_corr} is reduced by half to 15 and one where the maximum allowed replacement threshold voltage is decreased from 1 V to 0.5 V. Results from these runs are summarized in Figure \ref{deg_sensitivity}. Reducing the degradation correlation coefficient results in a 5.2\% decrease in LCOH due to the longer lifetime of these cells (3.04 years), reducing the planned replacement costs. Reducing the replacement threshold voltage results in an 8.5\% increase in LCOH mainly due to more frequent stack replacement (every 2.00 years) due to the lower threshold. Additionally, as seen in Figure \ref{diff_deg_scenarios}, changes in the underlying degradation assumptions do influence operation as well. When the coefficient is lowered to 15, the model can operate at higher current densities than the base case. However, when the voltage threshold is lowered, the model must operate at lower current densities to keep degradation (and therefore replacement rate) low.
\subsection*{Impact of Higher Temperature}
Since the temperature limit was a binding constraint, particularly with use-dependent degradation (Figure \ref{present_results}), we ran a scenario where the upper bound for the temperature of the cell was increased from $80^{\circ}$C to $90^{\circ}$C. The purpose of altering this was to determine if the upper bound of the temperature constraint included in the model was binding. As seen in Figure \ref{high_temp_operation}, operation of the system with the higher temperature bound marginally increases the current density during periods of low electricity prices. This is due to the fact that operating at a higher temperature lowers both activation overpotentials and ohmic overpotentials, which reduces operating costs and incentivizes higher current density operation (see Figure \ref{polarcurves}). For example, at 1 \si{\ampere\per\square\centi\meter} and 60$^\circ$C the voltage is 1.78 \si{\volt} and at 1 \si{\ampere\per\square\centi\meter} and 80$^\circ$C the voltage is 1.7 \si{\volt}.
When looking at the distribution of operating current densities over the year shown in Figure \ref{high_temp_current}, it is clear the higher temperature upper bound case operates at slightly higher current densities generally. These hours of higher current density operation can take advantage of the extra \ch{H2} storage sized in the high temperature upper limit case (See Table \ref{results}). The smaller number of cells in the high temperature upper bound case combined with a greater use of storage results in a \$0.04/kg lower LCOH when compared with the 2022 degradation case (see Figure \ref{LCOH_final}). However, higher current density operation results in more degradation and a more frequent stack replacement rate of 2 years in comparison to the base case replacement rate of 2.2 years (See Table \ref{results}).
\subsection*{Impact of Recombination Catalyst and Safety Constraint}
Industrial PEM membranes come embedded with a gas recombination catalyst on the anode side \cite{Christopher_Bryce_Capuano_Morgan_Elizabeth_Pertosos_Nemanja_Danilovic2018-lw,Nicolas_Guillet2014-tq}. This catalyst lowers the concentration of \ch{H2} at the anode by reacting the \ch{H2} with oxygen to form water. In the original base model, we included a safety constraint. However, in practice, it may be difficult to enforce a safety constraint implying that the recombination catalyst might be the primary mitigation strategy. To evaluate whether the catalyst with 90\% conversion is sufficient to ensure safe operation, we ran the model with the area and storage fixed from the cost-optimal base case with the catalyst present but without a safety constraint. Results are summarized in Figures \ref{no_safety} and \ref{no_safety_hist}.
The distribution of current densities with the relaxed safety constraint shifts to lower current densities. Intuitively, with the constraint relaxed, the model does not need to keep production of \ch{O2} at the anode high and regardless of the rate of crossover can operate in the most cost optimal manner. Relaxation of the safety constraint results in a lowered levelized cost of \$6.21/kg. Though the levelized cost is lower, the \ch{H2} lost to crossover increases from 11,300 kg/year in the base case to 30,400 kg/year in the case without a safety constraint, nearly a 3x change. This \ch{H2} represents lost product as it is not recovered. Replacement rate increases slightly from 2.2 years to 2.4 years due to lower current density operation without the safety constraint, further contributing to the decrease in LCOH.
More important than the change in LCOH is the change in safe operating conditions. Without the safety constraint and with the recombination catalyst, the model exceeds the LFL of 4\% at 1\% of the time points. While only a small percentage of the time operating, crossing into the flammability region of the mixture of gases presents a serious safety concern with dynamic operation. In practice, the composition of the outlet gas of the anode should be monitored continuously. This composition measurement could potentially be integrated into a controller that varies the flowrate of \ch{N2} into the anode to keep \ch{H2} concentrations well below the LFL.
\subsection*{Future Electricity Prices and Future CAPEX}
To test the sensitivity of these results to future electricity prices and capital costs, we evaluated the model based on a representative 2030 electricity price scenario for Texas and three possible capital cost projections. Capital cost projections in the mid-range case are \$0.79/\si{\square\centi\meter}, consistent with Fraunhofer ISE \cite{Holst2021-nd}. The low capital cost case is \$0.39/\si{\square\centi\meter}, and the high captial cost case is \$1.00/\si{\square\centi\meter}.
The operational profile in Figure \ref{future_operation}, shows that the 2030 mid-range CAPEX and high CAPEX case operate at similar current densities due to their similarly sized stacks and storage (see Table \ref{results}). Due to the increased cost of stacks in the high CAPEX case, the LCOH is 12\% more than the mid-case (see Figure \ref{LCOH_final}). For the lowest capital cost scenario, the number of cells increases by 13\% and the amount of storage decreases by 47\% compared to the mid-range case. With the greater number of cells, the scenario with the lowest capital cost can operate at lower current densities to avoid high degradation rates and keep operating costs lower. This results in a 7.7\% decrease in LCOH from the mid-range case.
\begin{figure}
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/2030_diff_capex_no_temp.eps} % calls your figure and set the width dimension
\caption{Two representative days with projected 2030 prices and projected 2030 capital costs. The mid-range CAPEX case has a stack CAPEX of \$ 0.79/cm$^2$, the high CAPEX case is \$1.00/cm$^2$, and the low CAPEX case is \$0.39/cm$^2$. The two cases with the higher capital costs operate quite similarly. The case with the lowest capital cost sizes many more cells and is able to therefore operate at lower current densities. Temperature profile included in Figure \ref{Temp_Only}.}% caption to the figure
\label{future_operation}
\end{minipage}
\end{figure}
In order to test sensitivity to future electricity prices, another two scenarios with differing electricity prices were run with the mid-range capital costs scenario as shown in Figure \ref{price_comparison_future}. Figure \ref{decarb} shows the differences in operation for the three different electricity price scenarios: the mid-decarbonization case, low decarbonization, and high decarbonization. While we do see the lowest decarbonization case able to operate at higher current densities, and a much larger sized storage for the high decarbonization case, ultimately, the LCOH is quite similar between scenarios indicating a weak relationship between LCOH and electricity price for this particular location (see Figure \ref{LCOH_final}).
\begin{figure}
\centering
\includegraphics[width=\textwidth]{Figures/LCOH.eps} % calls your figure and set the width dimension
\caption{Levelized cost of \ch{H2} for each of the cases. CAPEX Total is the capital cost of the stacks and downstream equipment, Planned Replacement is the cost for planned replacements of the stack as they age, Unplanned Replacement is the unplanned replacement cost as defined in Table \ref{technoecon}, fOPEX is the fixed operating cost, and vOPEX is the variable operating cost which is the cost of electricity and the cost of water used to feed and cool the stacks. Scenarios reported here are the same as reported in Table \ref{results}.}% caption to the figure
\label{LCOH_final}
\end{figure}
We developed a techno-economic optimization model for integrated design and scheduling of a PEM electrolyzer that respects temperature constraints, safety constraints, and models degradation of the electrolyzer stack as a function of its use. The model relies on an empirical correlation linking current density with degradation rate derived from experimental data in the literature, and is solved to co-optimize electrolyzer stack, balance of plant and storage capacity along with their operation. To enable computational tractability we employ number of techniques including approximating annual operations via operations over representative days, as well as 2-D GSS search algorithm that decomposes the investment and scheduling operating facilitates identification of locally optimal solutions of the resulting optimization model. We applied the model to explore the techno-economics of such a system and calculate the LOCH for various operating scenarios including current and future prices of electricity and stack capital costs.
Dynamic operation overall results in binding constraints on both temperature and safety limits. Heat management and safety of PEM systems must continue to be an area of focus for the operators of PEM systems if dynamic operation is to be successfully implemented. The monitoring and manipulation of stack temperature and anode gas composition will require detailed process control.
Using the correlation outlined in this work, accounting for degradation resulted in a $\approx$2 year stack lifetime in comparison to the 7 year assumed lifetime in the 2022 scenario. The increase in stack replacement frequency could potentially increase the materials burden on critical minerals like Ir, unless recycling becomes commonplace. Previous works have highlighted the scarcity of Ir especially when considering aggressive decarbonization scenarios \cite{Minke2021-io,Riedmayer2023-nw}. The increased replacement frequency associated with dynamic operation could exacerbate this issue even with replacement rates projected to increase to $\approx 4$ years by 2030. While some experimental work has begun to address this issue, significant advancements in PEM electrolyzer efficiencies, a decrease in Ir loadings in PEM, and a focus on PEM catalyst recycling are critical going forward \cite{Bernt2018-rj}.
The LCOH is higher in the 2022 case with degradation when compared to running without degradation. This indicates an underestimation of LCOH when operating dynamically in models that do not account for use-dependent degradation. We also show that optimizing design without accounting for use-dependent degradation and subsequently incurring these degradation costs leads to 4.6\% higher LCOH than co-optimized design and operations case. Additionally, not accounting for stack degradation leads to an over-sizing in hydrogen storage.
Dynamic operation does indeed make more cost effective hydrogen under present cost scenarios, and we show that dynamic operation makes even more sense under future electricity and capital cost scenarios by significantly minimizing the variable operating cost of electricity (see Figure \ref{LCOH_final}).
While the results presented in this paper highlight issues with cost optimal design and operation of PEM electrolyzers, there are some limitations to this analysis. Our work sought to capture degradation behavior of PEM stacks, however, the degradation correlation was found using the limited data available in literature. Standard procedures for accelerated degradation of cells are needed in order to gather reliable data that then can be used in models. More in depth degradation studies that not only look at the effect of operating current density but also temperature and catalyst loading effects are needed. One other weakness of our approach is the fact that we solve a non-convex NLP with no guarantee of finding a global minimum. Indeed for a problem of this size, it could be expected that there would be multiple minima in the design and operation space.
Most importantly, the framework presented here to evaluate PEM electrolysis and hydrogen production can be applied to the design of other electricity-intensive processes. With growing interest in electrochemical/electricity driven processes, model set up and system design can be readily adapted to study other processes and systems. This framework can help quickly and effectively evaluate different processes on the basis of their economics while modeling the underlying physical phenomenon and dynamics.
\section*{Acknowledgements}
The authors would like to acknowledge Edward Graham for his insightful contributions to this work. Funding for this work was provided by Analog Devices Inc.
\section*{Data Availability Statement}
The model and data used to produce all figures and results will be provided upon reasonable request.
\beginsupplement
\subsection{Mass and Energy Balance and Electrochemical Relations}
The following is a detailed description of the mass and energy balances as well as the electrochemical relations used in this work. Tables \ref{setnomenclature} - \ref{variablenomenclature} outline the nomenclature used in this work.
\input{table_of_variables}
\subsubsection{PEM Electrolyzer Potentials}
The current and voltage relationship used in this work borrows from the work done by \cite{Gorgun2006-ej,Liso2018-iy,Espinosa-Lopez2018-hb} and follows a standard procedure of estimating open circuit voltage and related overpotentials for the system as seen in Equation \ref{totvolt}. The total voltage is then the sum of these potentials.
\begin{align}
V_{total} = V_{oc} + V_{act} + V_{ohm} + V_{deg} \label{totvolt}
\end{align}
Degradation of the cell appears as an increase in operating voltage of the cell. Discussion of the calculation of degradation for a cell as a function of usage is presented in the Optimization section.
\paragraph{Thermodynamic Potential}
Thermodynamic voltage is described by the free energy of reaction as shown in Equation \ref{eqvolt}. In general, this free energy of reaction as well as the entropy and enthalpy are functions of pressure and temperature.
\begin{center}
\begin{align}
V_{rev} = \frac{\Delta G}{2F} \label{eqvolt}\\
\Delta G = \Delta H - T\Delta S \label{gibbs}
\end{align}
\end{center}
For this work, the enthalpy and entropy are assumed to be functions of temperature only. Correlations used to estimate enthalpies and entropies for \ch{H2O}, \ch{H2}, and \ch{O2} were taken from the NIST webbook \cite{BurgessUnknown-hi}.
The Nernst Equation corrects this thermodynamic voltage for conditions other than the standard condition.
\begin{center}
\begin{align}
V_{oc} &= V_{rev}^{\circ} - \frac{RT}{2F}\ln\left( \frac{a_{H_{2}O}}{a_{H_{2}}a_{O_{2}}^{1/2}}\right) \label{Nernst Equation}\\
V_{oc} &= V_{rev}^{\circ} + \frac{RT}{2F}\ln\left( \frac{P_{H_{2}}}{P^{\circ}}\sqrt{\frac{P_{O_{2}}}{P^{\circ}}}\right) \label{NernstSimple}
\end{align}
\end{center}
As is consistent with other PEM electrolyzers models, the activity of water is assumed to be unity in order to reduce Equation \ref{Nernst Equation} to Equation \ref{NernstSimple} \cite{Gorgun2006-ej,Liso2018-iy,Espinosa-Lopez2018-hb}. The reference pressure is assumed to be 1 bar with $P_{\ch{H2}}$ and $P_{\ch{O2}}$ being 30 bar and 1 bar, respectively.
\paragraph{Activation Overpotential}
Activation overpotential is determined using the Butler-Volmer equation applied to the anode and cathode. The sum of the resulting overpotentials is the total activation overpotential.
\begin{center}
\begin{align}
V_{act} = \frac{RT}{\alpha_{an} F}\text{arcsinh}\left(\frac{i}{2i_{0,an}}\right) + \frac{RT}{\alpha_{cat} F}\text{arcsinh}\left(\frac{i}{2i_{0,cat}}\right)
\end{align}
\end{center}
According to the kinetic theory used to derive the Butler-Volmer equation above, $\alpha_{an}$ and $\alpha_{cat}$ should sum to the number of electrons transferred in the reaction \cite{Thomas_Fuller_undated-ig}. As such, for a PEM, system they should sum to 2. However, in literature, $\alpha_{an}$, $\alpha_{cat}, i_{0,an},$ and $i_{0,cat}$ are used as fitting parameters for models when data for a polarization curve is available. This had led to a wide variety of charge transfer coefficients and exchange current densities. For this model charge transfer coefficients were left as fitting parameters for the system with the constraint that their sum could not exceed $n = 2$. Parameters were found as minimization of the sum of squared errors. Reference exchange current densities were chosen from typical values used in literature and corrected for temperature and roughness of the catalyst by Equations \ref{correctT} - \ref{correctrough} as was done by Görgün et al \cite{Falcao2020-vj, Gorgun2006-ej}.
\begin{center}
\begin{align}
i_{0,k}^{*} &= i_{0,ref,k}e^{\left[\frac{-E_{a}}{R}\left(\frac{1}{T}-\frac{1}{T_{ref}}\right)\right]} \label{correctT}\\
\gamma_{M,k} &= \varphi_{l}m_{M,k}\frac{6}{\rho_{M,k}d_{M,k}} \label{roughness}\\
i_{0,k} &= \gamma_{M,k}i_{0,k}^{*} \label{correctrough}
\end{align}
\end{center}
\paragraph{Ohmic Overpotential}
The ohmic resistance is a function of the thickness of the membrane, membrane conductivity, and operating current density. Membrane conductivity is a function of membrane hydration and operating temperature of the electrolyzer as seen in Equations \ref{ohm} - \ref{conduct}. Görgün et al used an empirical relation for the water hydration factor, $\lambda_{m}$, and $\sigma_{mem}$ where $\frac{1}{\sigma_{mem}} = R_{ohm}$ \cite{Gorgun2006-ej}. Under normal operating conditions the activity of water is such that $\lambda_m \approx 21$ , which was the value used for this study \cite{Maier_undated-zb}.
\begin{align}
V_{ohm} &= \frac{\delta_{mem}}{\sigma_{mem}}i \label{ohm}
\end{align}
\begin{align}
\lambda_{m} &= 0.43 + 17.81a_{H_{2}O} -39.85a_{H_{2}O}^{2} + 36a_{H_{2}O}^{3} \label{hyd_fact}\\
\sigma_{mem} &= (0.00514\lambda_{m}-0.00326)e^{\left( 1268\left(\frac{1}{303}-\frac{1}{T}\right)\right)} \label{conduct}
\end{align}
\subsubsection{PEM Electrolyzer Mass Balances}
\paragraph{Anode}\mbox{}\\
A transient water balance around the electrolyzer is shown in Equation \ref{wateranmol1}.
\begin{center}
\begin{align}
& \frac{dN_{H_{2}O,an}}{dt} = \dot{n}_{H_{2}O,2} - \dot{n}_{H_{2}O,3} - \dot{n}_{H_{2}O,3}^{vap} - \dot{n}_{H_{2}O,cross} - \dot{n}_{H_{2}O,consum} + \dot{n}_{H_{2}O,gen}\label{wateranmol1}\\
& 0 = \dot{n}_{H_{2}O,2} - \dot{n}_{H_{2}O,3} - \dot{n}_{H_{2}O,3}^{vap} - \dot{n}_{H_{2}O,cross} - \dot{n}_{H_{2}O,consum} +\dot{n}_{H_{2},cross}X_{H_{2}}\label{wateranmol2}
\end{align}
\end{center}
As is done on the industrial scale, the system is fed with excess water. As such, flows of water in and out of the anode are much greater than any liquid and vapor holdup in the anode. Therefore to simplify the system, the derivative term is assumed to be approximately zero in Equation \ref{wateranmol2}. It is also assumed there is a recombination catalyst embedded in the anode side of the membrane which has a 90\% conversion of \ch{H2} and \ch{O2} to \ch{H2O}. Conversion of hydrogen is represented by $X_{\ch{H2}}$. This recombination catalyst is a standard part of industrially operating PEM electrolyzers and are used to reduce the chance of a safety event occurring at the anode \cite{Christopher_Bryce_Capuano_Morgan_Elizabeth_Pertosos_Nemanja_Danilovic2018-lw,Nicolas_Guillet2014-tq}.
$\dot{n}_{H_{2}O,2}$ represents the liquid flowrate of water into the anode, and $\dot{n}_{H_{2}O,3}$ represents the liquid flowrate of water out of the anode. Other terms in the steady state water balance in Equation \ref{wateranmol2}, are defined below:
\begin{center}
\begin{align}
& \dot{n}_{H_{2}O,3}^{vap} = \frac{P_{H_{2}O}^{sat}}{P_{an}}\dot{n}_{gas,3}\label{watervap3}\\
& \dot{n}_{H_{2}O,cross} = N_c\frac{n_g i A}{2F}\label{watercross}\\
& n_g = 2.27 - 0.70i - 0.02P + 0.02Pi + 0.003T + 0.005iT - 0.0002PT\label{waterng} \\
& \dot{n}_{H_{2}O,gen} = N_c \frac{iA}{2F} \label{waterconsum}
\end{align}
\end{center}
In Equation \ref{watervap3}, water vapor is assumed to be in equilibrium with liquid water at the temperature of the cell in accordance with Raoult's Law. Equations \ref{watercross} and \ref{waterng} combined describe the electro-osmotic drift of water from anode to cathode at a given temperature, cathode pressure, and operating current density as described by Medina et al. \cite{Medina2010-lh}. Equation \ref{waterconsum} describes the consumption of water by Faraday's law assuming 100\% faradaic efficiency.
Balances for oxygen and hydrogen at the anode are shown below in Equations \ref{oxymolbal1}-\ref{hydmolbal2}. Each species is based on a mole balance with Faraday's law of electrolysis modeling the electrochemical reaction term.
\begin{center}
\begin{align}
\frac{dN_{O_{2},an}}{dt} & = \dot{n}_{O_{2},gen} - \dot{n}_{O_{2},3} - \frac{1}{2}\dot{n}_{H_{2},cross}X_{H_{2}} \label{oxymolbal1}\\
& = N_{c}\frac{iA_{cell}}{4F} -y_{O_{2},an}\dot{n}_{gas,3} - \frac{1}{2}\dot{n}_{H_{2},cross}X_{H_{2}} \label{oxymolbal2}\\
\frac{dN_{H_{2},an}}{dt} &= \dot{n}_{H_{2},cross}(1-X_{H_{2}}) - \dot{n}_{H_{2},out} \label{hydmolbal1}\\
& = 0.31 \frac{i A_{cell}}{2F}(P_{\ch{H2},cat}-P_{\ch{H2},an})(1-X_{H_{2}}) - y_{H_{2},an}\dot{n}_{gas,3} \label{hydmolbal2}
\end{align}
\end{center}
Equation \ref{oxymolbal1} represents the transient mole balance on oxygen. It is assumed the oxygen concentration of the water entering the anode is near zero, hence there is no oxygen flow into the anode. Equation \ref{oxymolbal2} is the same transient mole balance rearranged assuming oxygen is an ideal gas.
Equation \ref{hydmolbal1} represents the transient mole balance on hydrogen. The only way hydrogen can enter the anode is through crossover from the cathode, and the only way hydrogen can exit is through the anode exit gas stream. Equation \ref{hydmolbal2} is the same transient mole balance rearranged assuming hydrogen as an ideal gas and using the hydrogen crossover correlation investigated by Bernt et. al \cite{Bernt2020-af}.
\paragraph{Cathode}\mbox{}\\
A transient water balance on the cathode with the same holdup assumptions as the anode is shown in Equations \ref{watercatmol1} - \ref{watercatmol2}
\begin{center}
\begin{align}
& \frac{dN_{H_{2}O,cat}}{dt} = \dot{n}_{H_{2}O,cross} - \dot{n}_{H_{2}O,4} - \dot{n}_{H_{2}O,4}^{vap} \label{watercatmol1}\\
& 0 = \dot{n}_{H_{2}O,cross} - \dot{n}_{H_{2}O,4} - \dot{n}_{H_{2}O,4}^{vap} \label{watercatmol2}
\end{align}
\end{center}
The material balance for hydrogen and at the cathode are shown below in Equations \ref{hydcat1} - \ref{hydcat2}. Each species is based on a mole balance with Faraday's law of electrolysis modeling the electrochemical reaction term and gases are assumed ideal.
\begin{center}
\begin{align}
& \frac{dN_{H_{2}}}{dt} = \dot{n}_{H_{2},gen} - \dot{n}_{H_{2},4} - \dot{n}_{H_{2},cross} \label{hydcat1} \\
& \frac{dN_{H_{2}}}{dt} = N_c\frac{iA}{2F} - y_{H_{2}}\dot{n}_{gas,4} - 0.31 \frac{i A_{cell}}{2F}(P_{\ch{H2},cat}-P_{\ch{H2},an})(1-X_{H_{2}}) \label{hydcat2}
\end{align}
\end{center}
Oxygen crossover is assumed negligible as the oxygen is much less permeable to the Nafion membrane, and the cathode is pressurized meaning oxygen transport will be effectively zero \cite{Bernt2020-af}.
\subsubsection{PEM Electrolyzer Energy Balance}
The energy balance is based on a lumped capacitance model described in Equations \ref{energybalance}-\ref{energyloss}. Water enters the anode at ambient conditions so that $T_{in} = T_{ref}$. Espinosa-Lopez et. al presented a thermal lumped capacitance model for a 60 cell PEM electrolyzer stack in which they reported a lumped thermal capacitance $C_{p,th}$ and thermal resistance $R_{th}$ normalized by the area of the cell \cite{Espinosa-Lopez2018-hb}. These values for the lumped thermal capacitance and heat loss resistance were used in the model as a reference and scaled up or down based on the area of the modeled system. $V_{th}$ is the thermoneutral voltage of the cell, assumed to be 1.48 \si{\volt}.
\begin{align}
\frac{dT}{dt} &= \frac{1}{C_{p,th}}\left(\dot{H}_{in}-\dot{H}_{out} +\dot{Q}_{gen} - \dot{Q}_{loss}\right) \label{energybalance}\\
\dot{W}_{in} &= \sum_{i} n_{i,in}H_i(T_{in}) \label{energyin}\\
\dot{W}_{out} &= \sum_{i} n_{i,out}H_i(T_{cell}) \label{energyout}\\
\dot{Q}_{gen} &= N_{c}\left(V_{total} - V_{tn}\right)iA \label{energygenerated}\\
\dot{Q}_{loss} &= \frac{1}{R_{th}}\left(T-T_{ambient}\right) \label{energyloss}\\
\end{align}
The standard enthalpies of formation are included in the thermoneutral voltage, and $T_{ref} = T_{in} = T_{ambient}$ for the thermodynamic calculations. This is consistent with the energy balance presented in Espinosa et al and allows the heat generation term to be written as the potential applied over the thermoneutral voltage \cite{Espinosa-Lopez2018-hb}.
Equation \ref{energybalance} represents the overall energy balance for the system including all possible sources and sinks for heat. Equations \ref{energyin} - \ref{energyloss} define those sinks and sources. For the energy in, species include water and nitrogen (if used), and for the energy out species include, water, water vapor, nitrogen (if used), hydrogen, and oxygen.
\subsection{K-Means Clustering for Representative Days} \label{clustering description}
K-means clustering was performed on the input price data series of hourly electricity prices. The k-means clustering algorithm works to find $k$ centroids to which each of the 365 days in the year most closely correspond.
Once each day is assigned a cluster, the day closest to the cluster centroid is chosen as the representative day for that cluster. Each of these representative days then have an associated weight which signifies the number of real days that belong to the cluster. This way, the optimization can be performed in a significantly reduced time domain while still capturing much of the variability in electricity price seen throughout the year.
\FloatBarrier
\subsection{Storage Formulation} \label{storageformulation}
\begin{figure}[!h]
\centering
\includegraphics[width=\textwidth]{Figures/Storage_Formulation.eps} % calls your figure and set the width dimension
\caption{Storage formulation and connection between representative days and real days. Each real day has an hour to hour changing amount of storage $\Lambda^{H_2,stor}_d$. Each real day is mapped to a representative day by the k-means clustering mapping $f: d\rightarrow r$. Thus each representative day has an hour to hour changing amount of storage $\bar{\Lambda}^{H_2,stor}_r$. $\delta_r$ allows for carry over of storage between real days. To relate the representative days to real days, the beginning of a real day of storage must match the end of the corresponding representative day minus the carry over associated with the representative day which is summarized in the equation in this figure.}% caption to the figure
\label{storage_form}
\end{figure}
\FloatBarrier
\subsection{Convergence Time}
% Please add the following required packages to your document preamble:
% \usepackage{booktabs}
\begin{table}[!h]
\caption{Convergence time for scenarios presented in the paper. Each case took 15 iterations of the GSS in order to converge within 0.1\% tolerance.}
\centering
\begin{tabular}{lc}
\toprule
Case & Time (hours) \\ \midrule
Degradation 2022 Base Case & 44.8 \\
No Degradation & 25.2 \\
Degradation, West Texas & 74.3 \\
Degradation, Fixed CAPEX & -- \\
High Temperature & 57.6 \\
No Safety Constraint, Fixed CAPEX & -- \\
Degradation 2030 Base Case & 52.1 \\
2030 High CAPEX & 48.4 \\
2030 Low CAPEX & 43.7 \\
2030 High Decarbonization & 45.7 \\
2030 Low Decarbonization & 49.4 \\ \bottomrule
\end{tabular}
\label{convergence_time}
\end{table}
\FloatBarrier
\subsection{Algorithm Scaling}
Convergence time increases with an increasing number of representative days. The vast majority of constraints for each problem type are equality constraints. The LCOH seems to be converging to a singular value, albeit non monotonically. Theoretically, as the number of representative days approaches the number of real days, the LCOH would approach the real LCOH which is the LCOH of using the full price series data. 7 representative days was chosen for this study for the base case's ability to converge $< 48$ hours. For this particular price series it happens to be the number of representative days that has the lowest LCOH, though it was not chosen for that reason.
% Please add the following required packages to your document preamble:
% \usepackage{booktabs}
\begin{table}[!h]
\caption{Convergence time for different numbers of representative days. Each case was run with the base case South LZ price series with degradation.}
\centering
\begin{tabular}{lccccccc}
\toprule
\thead[l]{Representative \\ Days} & \thead[c]{LCOH \\ (\$/kg)} & \thead[c]{Convergence \\Time \\ (hours)} &\thead[c]{Number of \\ Variables} & \thead[c]{Number of \\Equality \\Constraints} & \thead[c]{Number of \\ Inequality\\ Constraints} &\thead[c]{Number of GSS \\ Iterations} &\thead[c]{Avg. GSS \\ Iteration \\ Time \\(min)}\\
\midrule
2 & \$ 6.96 & 23.3 & 188,469 & 186,728 & 776 & 15 & 21.8 \\
4 & \$ 6.92 & 31.5 & 199,913 & 196,796 & 1,552 & 15 & 29.5 \\
7 & \$ 6.60 & 44.8 & 217,079 & 211,898 & 2,716 & 15 & 39.6 \\
9 &\$ 6.83 & 49.4 & 228,523 & 221,966 & 3,492 & 15 & 46.3 \\
10 & \$ 7.06 & 57.5 & 234,245 & 227,000 & 3,880 & 15 & 53.9 \\
14 & \$ 7.19 & 82.3 & 257,133 & 247,136 & 5,432 & 15 & 77.2\\ \bottomrule
\end{tabular}
\label{algo_scaling}
\end{table}
\FloatBarrier
\subsection{Results from Alternate Scenarios}
\subsubsection{Degradation Sensitivities}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/LCOH_sensitivity.eps} % calls your figure and set the width dimension
\caption{Sensitivity to the degradation correlation coefficient being reduced to 15 (middle bar) and the replacement threshold voltage being reduced to 0.5 V (right bar) are compared against the base degradation case (left bar). Somewhat unsurprisingly a less aggressive degradation rate results in a lower levelized cost. Reducing the maximum amount of degradation that can occur before stack replacement increases LCOH.}% caption to the figure
\label{deg_sensitivity}
\end{minipage}
\end{figure}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/diff_deg_scenarios.eps} % calls your figure and set the width dimension
\caption{Changing the underlying assumptions on degradation shifts the operating profile from the base case. With less degradation occuring when a = 15, the model is free to operate at higher current densities than the base case. Additionally when the replacement threshold voltage is lowered the model is incentivized to keep replacement rates as high as possible by degrading as little as possible.}% caption to the figure
\label{diff_deg_scenarios}
\end{minipage}
\end{figure}
\FloatBarrier
\subsubsection{Degradation with CAPEX Fixed at No Degradation Result}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/present_fixed_area_hist.eps} % calls your figure and set the width dimension
\caption{Current density, storage, and temperature distributions for the base degradation case and the degradation case run with the stack size and storage size of the no degradation case. Fixing the stack size and storage results in having to operate at much higher current densities and more frequent replacements.}% caption to the figure
\label{fixed_case}
\end{minipage}
\end{figure}
\FloatBarrier
\subsubsection{Impact of Higher Temperature}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/high_temp.eps} % calls your figure and set the width dimension
\caption{Current density profile and temperature profile of scenario where the upper bound of the temperature limit is changed from 80$^\circ$C to 90$^\circ$C. Current density operation is nearly the same for both cases with the case with a higher temperature upper bound operating at slightly higher current densities. This coincides with a larger amount of storage sized for this system.}% caption to the figure
\label{high_temp_operation}
\end{minipage}
\end{figure}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/high_temp_hist.eps} % calls your figure and set the width dimension
\caption{Higher temperature upper bound case current density, storage, and temperature distribution. The high temperature case operates at slightly higher current densities overall and as a result sizes and utilizes a greater amount of storage than the base degradation case.}% caption to the figure
\label{high_temp_current}
\end{minipage}
\end{figure}
\FloatBarrier
\subsubsection{Impact of Recombination Catalyst and Safety Constraint}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/no_safety.eps} % calls your figure and set the width dimension
\caption{Operational profiles for the case with no safety constraint. The case without a safety constraint operates at lower current densities since it does not have maintain O2 production at the anode to keep the H2 concentration below lower flammability limit.}% caption to the figure
\label{no_safety}
\end{minipage}
\end{figure}
\begin{figure}[!h]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/safety_hist.eps} % calls your figure and set the width dimension
\caption{Current density, hydrogen storage, and temperature distributions for the case run without a safety constraint and the base degradation case. With no incentive to keep hydrogen concentration at the anode low, the current density distribution shifts to lower current densities.}% caption to the figure
\label{no_safety_hist}
\end{minipage}
\end{figure}
\FloatBarrier
\subsubsection{Future Scenarios}
\begin{figure}
\centering
\includegraphics[width=\textwidth]{Figures/2030_diff_capex_only_temp.eps} % calls your figure and set the width dimension
\caption{Temperature profile of future CAPEX cases. Generally, the high CAPEX case with more cells operates at lower temperatures due to lower current density and therefore lower overpotential.}% caption to the figure
\label{Temp_Only}
\end{figure}
\begin{figure}[!ht]
\centering
\begin{minipage}[c]{\textwidth}
\includegraphics[width=\textwidth]{Figures/2030_diff_decarb.eps} % calls your figure and set the width dimension
\caption{Distribution of current density, hydrogen storage, and temperature for the future cases run with low, mid, and high decarbonized grids. Price scenario corresponding to high grid decarbonization leads to sizing greater storage and operating at lower current densities. Lower grid decarbonization based price scenarios allows for operation at higher current densities because of the lower price of electricity. However, these effects largely do not affect the LCOH (see Table \ref{results}).}% caption to the figure
\label{decarb}
\end{minipage}
\end{figure}
\begin{table}
\caption{Set Nomenclature}
\centering
\begin{tabular}{llccc}
\toprule
Name & Description & Units & Value & Notes\\ \midrule
Set Indices \\ \hline
$i \in I$ & Chemical species & - & \{\ch{H2},\ch{O2},\ch{H2O},\ch{N2}\} & - \\
$j \in J$ & Stream numbers & - &\{1,2,...,16\} & -\\
$k \in K$ & Anode/Cathode & - &\{\text{an},\text{cat}\} & -\\
$r \in D_r$ & Representative days & - &\{1,2,...,7\} & -\\
$d \in D$ & Actual days & - & \{1,2,...,365\} & -\\
\bottomrule
\end{tabular}
\label{setnomenclature}
\end{table}
\begin{table}
\caption{Parameter Nomenclature}
\resizebox{\columnwidth}{!}{%
\centering
\begin{tabular}{llccc}
\toprule
Name & Description & Units & Value & Notes\\ \midrule
$F$ & Faraday's constant & \si[per-mode=fraction]{\coulomb\per\mol} & 96,485 & -\\
$A_{cell}$ & Catalytic active area & \si{\square\centi\meter} & 450 & \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz} \\
$A_{mem}$ & Area of membrane/Size of one cell & \si{\square\centi\meter} & 450 & \cite{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz} \\
$\delta_{el}$ & Thickness of electrode & \si{\centi\meter} & $8.0 \times 10^{-3}$ & \cite{Abdin2015-nl} \\
$v_{an}$ & Anode volume & \si[per-mode=fraction]{\mL} & $\approx \delta_{el}A_{mem}$ & - \\
$v_{cat}$ & Cathode volume & \si[per-mode=fraction]{\mL} & $\approx \delta_{el}A_{mem}$ & - \\
$\delta_{mem}$ & Membrane thickness & \si[per-mode=fraction]{\centi\meter} & $1.75 \times 10^{-2}$ & \cite{Liso2018-iy} \\
$n_{g}$ & Electro-osmotic drift coefficient& unitless & - & \cite{Medina2010-lh}, Equation \ref{waterng} \\
$\Delta G^{\circ}$ & Standard free energy of reaction & \si[per-mode=fraction]{\kilo\joule\per\mol} & $237.2$ & \cite{Liso2018-iy} \\
$R_{gas}$ & Gas constant & \si[per-mode=fraction]{\joule\per\mol\per\K} & $8.314$ & - \\
$T_{ref}$ & Reference temperature for thermo & \si[per-mode=fraction]{\K} & $298$ & \cite{Liso2018-iy} \\
$H_{i}$ & Enthalpy of species $i$ & \si[per-mode=fraction]{\joule\per\mole} & - & \cite{BurgessUnknown-hi} \\
$C_{p,th}$ & Lumped thermal capacitance & \si[per-mode=fraction]{\joule\per\K} & - & Scaled from \cite{Espinosa-Lopez2018-hb} \\
$R_{th}$ & Thermal resistance of stack & \si[per-mode=fraction]{\second\per\joule\per\K} & - & Scaled from \cite{Espinosa-Lopez2018-hb} \\
$i_{0,an,ref}$ & Exchange current density at $T_{ref}$ & \si[per-mode=fraction]{\ampere\per\square\centi\meter} & $5 \times 10^{-12}$ & \cite{Liso2018-iy} \\
$i_{0,cat,ref}$ & Exchange current density at $T_{ref}$ & \si[per-mode=fraction]{\ampere\per\square\centi\meter} & $1 \times 10^{-3}$ & \cite{Liso2018-iy} \\
$\alpha_{an}$ & Bulter-Volmer equation parameter & - & - & Fitted parameter \\
$\alpha_{cat}$ & Bulter-Volmer equation parameter & - & - & Fitted parameter \\
$\gamma_{M,an}$ & Anode roughness factor & - & 1198 & Equation \ref{roughness}\\
$\gamma_{M,cat}$ & Cathode roughness factor & - & 286 & Equation \ref{roughness}\\
$\varphi_{I}$ & Fraction of metal catalyst in contact with ionomer & - & 0.75 & \cite{Liso2018-iy}\\
$m_{M,an}$ & Catalyst loading at anode& \si[per-mode=fraction]{\gram\per\square\centi\meter} & $0.9\times10^{-3}$ & \cite{Bernt2020-ne}\\
$m_{M,cat}$ & Catalyst loading at cathode & \si[per-mode=fraction]{\gram\per\square\centi\meter} & $0.3\times10^{-3}$ & \cite{Bernt2020-ne}\\
$\rho_{M,an}$ & Catalyst density anode & \si[per-mode=fraction]{\gram\per\centi\meter\cubed} & 11.66 & \cite{Bernt2020-ne}\\
$\rho_{M,cat}$ & Catalyst density cathode & \si[per-mode=fraction]{\gram\per\centi\meter\cubed} & 21.45 & \cite{Bernt2020-ne}\\
$d_{M,an}$ & Catalyst crystal diameter anode & \si[per-mode=fraction]{\centi\meter} & $2.9\times10^{-7}$ & \cite{Bernt2020-ne}\\
$d_{M,cat}$ & Catalyst crystal diameter cathode & \si[per-mode=fraction]{\centi\meter} & $2.2\times10^{-7}$ & \cite{Bernt2020-ne}\\
$\lambda_{m}$ & Hydration factor of PEM & $\frac{\text{mol}_{\ch{H2O}}}{\text{mol}_{\ch{SO3}}}$ & 21 & \cite{Gorgun2006-ej}, Equation \ref{hyd_fact}\\
$\sigma_{mem}$ & Conductance of membrane & - & - & \cite{Gorgun2006-ej}, Equation \ref{conduct}\\
$p_{elec}$ & Price of electricity & - & - & \cite{noauthor_undated-so,Gagnon_undated-so}\\
$p_{water}$ & Price of water & $\frac{\$}{1000 \text{ gal}}$ & 2.78 & \cite{Water_Innovations_Inc_undated-io}\\
\bottomrule
\end{tabular}
}
\label{parameter nomenclature}
\end{table}
\begin{table}
\caption{Variable Nomenclature}
\resizebox{\columnwidth}{!}{%
\centering
\begin{tabular}{llcccc}
\toprule
Name & Description & Units & Value & Primary Decision Variable &Notes\\ \midrule
$i(t)$ & Current density & \si{\ampere\per\square\centi\meter} & $0.1 - 4$& $\checkmark$ & - \\
$\dot{n}_{\ch{H2O},1}(t)$ & Inlet water flowrate & \si{\mol\per\second} & $0 - 50,000$& $\checkmark$ & base case\\
$\dot{n}_{3}(t)$ & Total flow out of anode & \si{\mol\per\second} & $0 - 10 N_c \frac{A_{cell}i}{4F}$& $\checkmark$ & base case\\
$\dot{n}_{4}(t)$ & Total flow out of cathode & \si{\mol\per\second} & $0 - 10 N_c \frac{A_{cell}i}{2F}$& $\checkmark$ & base case\\
$\dot{n}_{13}(t), \dot{n}_{14}$ & Split between satisfying demand and storage & \si{\mol\per\second} & $0 - 10^6$& $\checkmark$ & base case\\
$\dot{n}_{16}(t)$ & Flowrate out of storage & \si{\mol\per\second} & $0 - 10^6$& $\checkmark$ & base case\\
$\dot{n}_{11}(t)$ & Flowrate of \ch{N2} & \si{\mol\per\second} & $0 - 10^3$& $\checkmark$ & base case\\
$N_{i}(t)$ & Moles of species $i$ & \si{\mol} & - & & -\\
$\dot{n}_{i,j}(t)$ & Molar flow rate of species $j$ in stream $i$ & \si[per-mode=fraction]{\mol\per\second} & - & & -\\
$y_{j,k}(t)$ & Mole fraction of species $j$ in stream at electrode $k$ & - & - & & -\\
$T(t)$ & Temperature of anode and cathode & \si{\degreeCelsius} & $60-80$ & &base case\\
& & & $60-90$ & &sensitivity\\
$X_{\ch{H2}}$ & Conversion of hydrogen crossing from cathode to anode & - & 0.9 & &base case\\
$V^{total}(t)$ & Cell voltage & \si{\volt} & - & &base case\\
$V^{undeg}_{d}(t)$ & Cell voltage not including degradation on day $d$ & \si{\volt} & - & &base case\\
$\delta V^{deg}_{r}(t)$ & Intra-day change in degradation up to time $t$ & \si{\volt} & - & &base case\\
$V^{cuml,deg}_{d}(t)$ & Cumulative degradation up to day $d$ and time $t$ & \si{\volt} & - & &base case\\
$C^{elec}$ & Annual cost of electricity & \$ & - & &base case\\
$C^{BoP,elec}$ & Annual cost of electricity used by balance of plant & \$ & - & &base case\\
$C^{water}$ & Annual cost of de-ionized water & \$ & - & &base case\\
$C^{\ch{N2}}$ & Annual cost of purge nitrogen & \$ & - & &base case\\
$\delta_{r}$ & Change in storage over representative period $r$ & \si{\mole} & - & &\\
$\Lambda_{\ch{H2}}$ & Hydrogen storage & \si{\mole} & - & &\\
\bottomrule
\end{tabular}
}
\label{variablenomenclature}
\end{table}
% $ biblatex auxiliary file $
% $ biblatex bbl format version 3.2 $
% Do not modify the above lines!
%
% This is an auxiliary file used by the 'biblatex' package.
% This file may safely be deleted. It will be recreated by
% biber as required.
%
\begingroup
\makeatletter
\@ifundefined{ver@biblatex.sty}
{\@latex@error
{Missing 'biblatex' package}
{The bibliography requires the 'biblatex' package.}
\aftergroup\endinput}
{}
\endgroup
\refsection{1}
\datalist[entry]{none/global//global/global}
\entry{Yarmuth2021-wu}{misc}{}
\name{author}{1}{}{%
{{hash=d53d2d3a5a78a182441cabfb66dbf080}{%
family={Yarmuth},
familyi={Y\bibinitperiod},
given={John},
giveni={J\bibinitperiod}}}%
}
\strng{namehash}{d53d2d3a5a78a182441cabfb66dbf080}
\strng{fullhash}{d53d2d3a5a78a182441cabfb66dbf080}
\strng{bibnamehash}{d53d2d3a5a78a182441cabfb66dbf080}
\strng{authorbibnamehash}{d53d2d3a5a78a182441cabfb66dbf080}
\strng{authornamehash}{d53d2d3a5a78a182441cabfb66dbf080}
\strng{authorfullhash}{d53d2d3a5a78a182441cabfb66dbf080}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{9}
\field{number}{117-169}
\field{title}{Inflation Reduction Act}
\field{year}{2021}
\keyw{Dynamic PEM Paper}
\endentry
\entry{UnknownUnknown-ci}{misc}{}
\list{language}{1}{%
{en}%
}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labeltitlesource}{title}
\field{abstract}{Regional Clean Hydrogen Hubs Selections for Award Negotiations}
\field{booktitle}{Energy.gov}
\field{howpublished}{\url{https://www.energy.gov/oced/regional-clean-hydrogen-hubs-selections-award-negotiations}}
\field{note}{Accessed: 2023-10-16}
\field{title}{Regional Clean Hydrogen Hubs Selections for award negotiations}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Arjona2023-bc}{report}{}
\name{author}{1}{}{%
{{hash=b9726c7a62dd875cb8535793afe734f7}{%
family={Arjona},
familyi={A\bibinitperiod},
given={Vanessa},
giveni={V\bibinitperiod}}}%
}
\list{institution}{1}{%
{U.S. Department of Energy}%
}
\strng{namehash}{b9726c7a62dd875cb8535793afe734f7}
\strng{fullhash}{b9726c7a62dd875cb8535793afe734f7}
\strng{bibnamehash}{b9726c7a62dd875cb8535793afe734f7}
\strng{authorbibnamehash}{b9726c7a62dd875cb8535793afe734f7}
\strng{authornamehash}{b9726c7a62dd875cb8535793afe734f7}
\strng{authorfullhash}{b9726c7a62dd875cb8535793afe734f7}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{6}
\field{title}{{DOE} Hydrogen Program Record}
\field{type}{techreport}
\field{year}{2023}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Stansberry2017-eo}{article}{}
\name{author}{4}{}{%
{{hash=10f224d7be0eef1e3c6f592f58bad0fa}{%
family={Stansberry},
familyi={S\bibinitperiod},
given={John},
giveni={J\bibinitperiod}}}%
{{hash=ec8ab62cc11ef1bd128125d21e0de90b}{%
family={Hormaza\bibnamedelima Mejia},
familyi={H\bibinitperiod\bibinitdelim M\bibinitperiod},
given={Alejandra},
giveni={A\bibinitperiod}}}%
{{hash=1a85fb4271591180033cc6d2d2310539}{%
family={Zhao},
familyi={Z\bibinitperiod},
given={Li},
giveni={L\bibinitperiod}}}%
{{hash=9eca047f8a598e8b603310861e7b92f4}{%
family={Brouwer},
familyi={B\bibinitperiod},
given={Jack},
giveni={J\bibinitperiod}}}%
}
\strng{namehash}{cc05e14f7a65346b6f12b16fde9e85c6}
\strng{fullhash}{123b4197fcf93e4e1400008eb2051400}
\strng{bibnamehash}{123b4197fcf93e4e1400008eb2051400}
\strng{authorbibnamehash}{123b4197fcf93e4e1400008eb2051400}
\strng{authornamehash}{cc05e14f7a65346b6f12b16fde9e85c6}
\strng{authorfullhash}{123b4197fcf93e4e1400008eb2051400}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{A proton exchange membrane electrolyzer is integrated with a photovoltaic system using Maximum Power Point Tracking (MPPT) power electronics systems in several configurations to evaluate the extent to which the system can respond to the dynamics of photovoltaic power. Both direct current from the solar photovoltaics (DC) and alternating current rectified to DC from the grid (AC) configurations are evaluated and the overall efficiency and dynamic hydrogen production capabilities are assessed. Performance characteristics as a function of weather (winter vs. summer; cloudy vs. sunny) and as a function of operating conditions (e.g., ambient temperature, stack temperature) are evaluated. The direct DC configuration was found to provide the highest energy transfer efficiency from electrical power to hydrogen, with a relative increase of 7--8\% compared to the AC configuration. The grid-connected AC configuration produced a baseline case to compare the characteristic losses, efficiency, and dynamics of the PV integrated DC case. The highly dynamic PV production during cloudy weather was handled by the system, accommodating input power ramp rates as high as 1270 W/s, and part load conditions as low as 7.6\% of rated power. Transient operation exhibited small performance degradation compared to steady solar conditions. Hydrogen production was more efficient for high stack operating temperatures and high ambient temperatures.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{12}
\field{number}{52}
\field{title}{Experimental analysis of photovoltaic integration with a proton exchange membrane electrolysis system for power-to-gas}
\field{volume}{42}
\field{year}{2017}
\field{pages}{30569\bibrangedash 30583}
\range{pages}{15}
\keyw{Proton exchange membrane electrolyzer; Electrolysis; Solar photovoltaic; Power-to-gas; Renewable hydrogen;Dynamic PEM Paper}
\endentry
\entry{Laoun2016-mb}{article}{}
\name{author}{4}{}{%
{{hash=c2c82746419dcaaee4963aeb3d3acffa}{%
family={Laoun},
familyi={L\bibinitperiod},
given={Brahim},
giveni={B\bibinitperiod}}}%
{{hash=aa8f11f53d2d6113d5461ef4a7ac2410}{%
family={Khellaf},
familyi={K\bibinitperiod},
given={Abdallah},
giveni={A\bibinitperiod}}}%
{{hash=c78740b87d297a2e763f31c48b2ba005}{%
family={Naceur},
familyi={N\bibinitperiod},
given={Mohamed\bibnamedelima W},
giveni={M\bibinitperiod\bibinitdelim W\bibinitperiod}}}%
{{hash=7a02bde67e873e28d3829442bc625867}{%
family={Kannan},
familyi={K\bibinitperiod},
given={Arunachala\bibnamedelima M},
giveni={A\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
}
\strng{namehash}{9b8a88cee4ff7e100803036574109626}
\strng{fullhash}{2cd9980a1d8fcfc66b2df64222071fca}
\strng{bibnamehash}{2cd9980a1d8fcfc66b2df64222071fca}
\strng{authorbibnamehash}{2cd9980a1d8fcfc66b2df64222071fca}
\strng{authornamehash}{9b8a88cee4ff7e100803036574109626}
\strng{authorfullhash}{2cd9980a1d8fcfc66b2df64222071fca}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Hydrogen production by proton exchange membrane water electrolysis (PEMWE) directly powered by photovoltaics (PV) is analyzed by a detailed modeling approach which consists in three parts. The first part is devoted to the quantification of the solar irradiation received on the PV module with the integration of the geographic characteristic of the desired location, weather data, and an estimation of ozone concentration. In the second part, an elaborated mathematical model is presented to estimate the power delivered by PV module, the model include temperature and the amount of solar irradiation. In the third part, an elaborated model of proton exchange membrane water electrolysis was presented. Based on a thorough analysis of the thermodynamic and the electrochemistry of PEMWE, for the first time the efficiency of the PEMWE was well defined. The relevance and the effectiveness have been tested against experimental data with accuracy for each developed models. Then, the complete model, that combine the three parts, has been used to understand the effect of solar irradiation, temperature, pressure, PV electrical efficiency and PEMWE efficiency on the voltage required for electrolysis and hydrogen production rate. Additionally, a multi-objective optimization technic was applied in order to minimize the energy conversion loss, between the PV-PEMWE, and to increase hydrogen production rate. The model presented in this study offers accuracy and flexibility to the analysis of Solar PV- PEMWE system for hydrogen production and helps to analyze the influence of each component on the overall structure, and to develop a decision tool to determine the performance of the Solar PV--PEMWE, to assess the economics of the system and to assist for the deployment of PV-PEMWE technology.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{6}
\field{number}{24}
\field{title}{Modeling of solar photovoltaic-polymer electrolyte membrane electrolyzer direct coupling for hydrogen generation}
\field{volume}{41}
\field{year}{2016}
\field{pages}{10120\bibrangedash 10135}
\range{pages}{16}
\keyw{Solar-hydrogen; PV electrolysis; PEM water electrolyzer; Direct and optimal coupling;Dynamic PEM Paper}
\endentry
\entry{Sarrias-Mena2015-vy}{article}{}
\name{author}{4}{}{%
{{hash=5f8f3ff534a41ea1d328266fc5ad0b58}{%
family={Sarrias-Mena},
familyi={S\bibinithyphendelim M\bibinitperiod},
given={Raúl},
giveni={R\bibinitperiod}}}%
{{hash=477fcabbd29673d14640186d6c2ae242}{%
family={Fernández-Ramírez},
familyi={F\bibinithyphendelim R\bibinitperiod},
given={Luis\bibnamedelima M},
giveni={L\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=99430919459b934831305b900824733e}{%
family={García-Vázquez},
familyi={G\bibinithyphendelim V\bibinitperiod},
given={Carlos\bibnamedelima Andrés},
giveni={C\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=2ccaf4bce748f1200c906d643c06bfc6}{%
family={Jurado},
familyi={J\bibinitperiod},
given={Francisco},
giveni={F\bibinitperiod}}}%
}
\strng{namehash}{d314abba4732144b4cf1a487a0bb4c8b}
\strng{fullhash}{6564cb39aed361a2cdeb08c4e8417528}
\strng{bibnamehash}{6564cb39aed361a2cdeb08c4e8417528}
\strng{authorbibnamehash}{6564cb39aed361a2cdeb08c4e8417528}
\strng{authornamehash}{d314abba4732144b4cf1a487a0bb4c8b}
\strng{authorfullhash}{6564cb39aed361a2cdeb08c4e8417528}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{The continuous progress on the expansion of renewable energies leads to the development of hybrid power systems, where several power sources contribute to provide a clean and reliable alternative to traditional fossil fuels. The hydrogen technology is viewed with particular interest in this regard. Hydrogen is an outstanding energy carrier that can be exploited for various applications, including electricity generation. Hence, production of hydrogen from renewable sources has received the attention of many researchers lately. With this purpose, this paper deals with the coupled operation of electrolyzer (EZ) and wind turbine. Four different EZ models are presented and evaluated in this work. These models are aggregated to a variable speed wind turbine model using MATLAB/Simulink. The four configurations are evaluated, and their responses compared, under variable wind speed and grid demand.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{2}
\field{number}{7}
\field{title}{Electrolyzer models for hydrogen production from wind energy systems}
\field{volume}{40}
\field{year}{2015}
\field{pages}{2927\bibrangedash 2938}
\range{pages}{12}
\keyw{Electrolyzer; Hydrogen; Modelling; Wind turbine;Dynamic PEM Paper}
\endentry
\entry{Saadi2016-xg}{article}{}
\name{author}{3}{}{%
{{hash=4a29c7863b8353c2478710963c308181}{%
family={Saadi},
familyi={S\bibinitperiod},
given={A},
giveni={A\bibinitperiod}}}%
{{hash=4cc21f3f9d5b008a22f3ec2ec1165c1c}{%
family={Becherif},
familyi={B\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=badd905164de4ceedb8014a2ac857cab}{%
family={Ramadan},
familyi={R\bibinitperiod},
given={H\bibnamedelima S},
giveni={H\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
}
\strng{namehash}{8f02df9c1622ae875019170970937dba}
\strng{fullhash}{8f02df9c1622ae875019170970937dba}
\strng{bibnamehash}{8f02df9c1622ae875019170970937dba}
\strng{authorbibnamehash}{8f02df9c1622ae875019170970937dba}
\strng{authornamehash}{8f02df9c1622ae875019170970937dba}
\strng{authorfullhash}{8f02df9c1622ae875019170970937dba}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{This paper aims at studying the hydrogen production from solar energy available in Biskra region in the east of Algeria by the Proton Exchange Membrane (PEM) electrolyzer. The electrolyzer is supplied by solar photovoltaic (PV) system. Three kinds of PV generator with rated power up to 6 kW are used for producing hydrogen. The PV panel characteristics are experimentally validated. The instantaneous effects of temperature and solar radiation on PV characteristics are shown. Moreover, a comprehensive study is devoted to validate two different models for the areas solar radiation. The simulation results are compared to the experimental measurements using statistical analysis. The solar radiation model on a horizontal and on a tilted and oriented PV panel give sufficient results. The adequate agreement between the simulation results and the experimental data verifies the effectiveness of the proposed models.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{12}
\field{number}{47}
\field{title}{Hydrogen production horizon using solar energy in Biskra, Algeria}
\field{volume}{41}
\field{year}{2016}
\field{pages}{21899\bibrangedash 21912}
\range{pages}{14}
\keyw{(PEM) electrolyzer; Hydrogen; Photovoltaic panel; Solar radiation model;Dynamic PEM Paper}
\endentry
\entry{Zhao2015-bk}{article}{}
\name{author}{2}{}{%
{{hash=1a85fb4271591180033cc6d2d2310539}{%
family={Zhao},
familyi={Z\bibinitperiod},
given={Li},
giveni={L\bibinitperiod}}}%
{{hash=bac6f9a64636de0932b198c223cfc98f}{%
family={Brouwer},
familyi={B\bibinitperiod},
given={Jacob},
giveni={J\bibinitperiod}}}%
}
\strng{namehash}{930516b7c4bdc75f2557e2ffa3f3a44a}
\strng{fullhash}{930516b7c4bdc75f2557e2ffa3f3a44a}
\strng{bibnamehash}{930516b7c4bdc75f2557e2ffa3f3a44a}
\strng{authorbibnamehash}{930516b7c4bdc75f2557e2ffa3f3a44a}
\strng{authornamehash}{930516b7c4bdc75f2557e2ffa3f3a44a}
\strng{authorfullhash}{930516b7c4bdc75f2557e2ffa3f3a44a}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{To evaluate the dynamic operation and feasibility of designing and operating a self-sustainable hydrogen fueling station using renewable energy sources, dynamic system models have been developed for a hydrogen fueling station utilizing a proton exchange membrane (PEM) electrolyzer and fuel cell. Using fueling and power demand data from an existing public hydrogen station in Irvine, California, dynamic analyses of the self-sustainable station have been carried out. Various control strategies are developed and evaluated to determine the impacts of control strategies and renewable capacity factors on the efficiency and other performance characteristics of the station. The simulation results and analysis suggest that with careful sizing and system design, a self-sustainable hydrogen fueling station that relies completely upon renewable sources for hydrogen production, storage and dispensing is feasible. Moreover, a cost and sensitivity analysis is carried out to evaluate the levelized hydrogen cost for different station designs. The cost of the hydrogen is determined to be as low as $6.71 per kg or $9.14 per kg when the station is powered by 200 kW of wind turbines or 360 kW of PV arrays, respectively.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{3}
\field{number}{10}
\field{title}{Dynamic operation and feasibility study of a self-sustainable hydrogen fueling station using renewable energy sources}
\field{volume}{40}
\field{year}{2015}
\field{pages}{3822\bibrangedash 3837}
\range{pages}{16}
\keyw{Renewable hydrogen; Hydrogen fueling station; Electrolyzer; Fuel cell; Dynamics;Dynamic PEM Paper}
\endentry
\entry{Kalinci2015-re}{article}{}
\name{author}{3}{}{%
{{hash=90d9c49aca6b211b5a3b430f74bc9914}{%
family={Kalinci},
familyi={K\bibinitperiod},
given={Yildiz},
giveni={Y\bibinitperiod}}}%
{{hash=e546cc0f422166440ff7a3aac15b2b8e}{%
family={Hepbasli},
familyi={H\bibinitperiod},
given={Arif},
giveni={A\bibinitperiod}}}%
{{hash=1793ed9ab072c368761490dc2d6bca38}{%
family={Dincer},
familyi={D\bibinitperiod},
given={Ibrahim},
giveni={I\bibinitperiod}}}%
}
\strng{namehash}{d1568bae2191795884abed813f16a184}
\strng{fullhash}{d1568bae2191795884abed813f16a184}
\strng{bibnamehash}{d1568bae2191795884abed813f16a184}
\strng{authorbibnamehash}{d1568bae2191795884abed813f16a184}
\strng{authornamehash}{d1568bae2191795884abed813f16a184}
\strng{authorfullhash}{d1568bae2191795884abed813f16a184}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{In the present study, a hybrid renewable energy system using hydrogen energy as energy storage option is conceptually modeled for the Bozcaada Island in Turkey. The system is investigated from the techno-economic point of view. The Hybrid Optimization Model for Electric Renewable (HOMER) tool is used to define the optimum size of the equipment based on the geographical and meteorological data of the island. The HOMER uses the net present cost (NPC) method while ranking the system suitability. Also, the cost of energy (COE) is calculated with the total annual cost (ACT). The study considers two scenarios, which are only wind turbine and wind turbine/PV hybrid systems. Using the wind turbine/PV array system, instead of wind turbine only, decreases the NPC from \$14,624,343 to \$11,960,698. Also, it decreases the hydrogen tank capacity to 400 kg and hence affects other equipment size and NPC. So, the COE varies between \$1.016/kWh and \$0.83/kWh. According to this, the optimum system components are defined as PV array-300 kW, wind turbine (E33x2), fuel cell-100 kW, converter-150 kW, electrolyzer-200 kW and hydrogen tank-400 kg to supply a 1875 kWh/d primary load. Furthermore, the effects of some parameters and the equipment on NPC are examined. Increasing potential of the renewable energy sources, such as annual average wind speed or solar radiation, decreases both COE and NPC. While the annual average solar radiation increases to 5 kWh/m2/d, the NPC and COE decrease to \$11,673,704 and \$0.81/kWh. Likewise, an increase in the annual average wind speed will decrease the costs to \$11,452,712 and \$0.795/kWh, respectively. Also, increasing the real interest i, to 5\% increases the COE to \$1.043/kWh, as expected.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{6}
\field{number}{24}
\field{title}{Techno-economic analysis of a stand-alone hybrid renewable energy system with hydrogen production and storage options}
\field{volume}{40}
\field{year}{2015}
\field{pages}{7652\bibrangedash 7664}
\range{pages}{13}
\keyw{Renewable energy; Cost assessment; Electricity; Hydrogen production;Dynamic PEM Paper}
\endentry
\entry{Shibata2015-qt}{article}{}
\name{author}{1}{}{%
{{hash=7a9354541d5450781789e355e105a3e9}{%
family={Shibata},
familyi={S\bibinitperiod},
given={Yoshiaki},
giveni={Y\bibinitperiod}}}%
}
\strng{namehash}{7a9354541d5450781789e355e105a3e9}
\strng{fullhash}{7a9354541d5450781789e355e105a3e9}
\strng{bibnamehash}{7a9354541d5450781789e355e105a3e9}
\strng{authorbibnamehash}{7a9354541d5450781789e355e105a3e9}
\strng{authornamehash}{7a9354541d5450781789e355e105a3e9}
\strng{authorfullhash}{7a9354541d5450781789e355e105a3e9}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{IEEJ Energy J}
\field{number}{2}
\field{title}{Economic analysis of hydrogen production from variable renewables}
\field{volume}{10}
\field{year}{2015}
\field{pages}{26\bibrangedash 46}
\range{pages}{21}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Hurtubia2021-jo}{article}{}
\name{author}{2}{}{%
{{hash=f4e34d4f61995ebf0bea2f4b777a69f0}{%
family={Hurtubia},
familyi={H\bibinitperiod},
given={Byron},
giveni={B\bibinitperiod}}}%
{{hash=4b4673d881082d0a001a14be46ea68c2}{%
family={Sauma},
familyi={S\bibinitperiod},
given={Enzo},
giveni={E\bibinitperiod}}}%
}
\strng{namehash}{ae8bb7da1c9873ff25b47040f8408838}
\strng{fullhash}{ae8bb7da1c9873ff25b47040f8408838}
\strng{bibnamehash}{ae8bb7da1c9873ff25b47040f8408838}
\strng{authorbibnamehash}{ae8bb7da1c9873ff25b47040f8408838}
\strng{authornamehash}{ae8bb7da1c9873ff25b47040f8408838}
\strng{authorfullhash}{ae8bb7da1c9873ff25b47040f8408838}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Hydrogen (H2) can be a critical factor in achieving a more sustainable society reducing carbon emissions worldwide in the long run. However, there are some challenges to be solved yet. One of these challenges is that, right now, the economical competitiveness of electrolysis production strongly depends on the Renewable Energy Sources (RES) availability. This paper analyzes the economic and environmental implications of complementing the power supply of a H2 production plant that usually operates using only RES with grid electricity during the time RES are not available. For performing this analysis, we proposed an assessment model that incorporates Alkaline (ALK) and PEM electrolyzer technology features (such as the low dynamic load range of ALK electrolyzers, which is usually ignored when modeling H2 production). Moreover, we introduce a new metric, named the LCOH\&E, which captures the existing trade-off between reducing the Levelized Cost Of Hydrogen (LCOH) with more H2 production using grid electricity and increasing CO2 emissions by producing more H2 with grid electricity. We illustrate the proposed assessment model with a case study based on real data from Chile. Our results show that the optimal fraction of grid electricity to be used in producing H2 may be different when minimizing the LCOH and when minimizing the LCOH\&E because of the LCOH-CO2 emissions' trade off. Moreover, it is remarkable that using only 10\% of grid electricity in an ALK electrolyzer, it is possible to increase the H2 production in 25.7\%, without significantly increasing the CO2 emission level.}
\field{journaltitle}{Appl. Energy}
\field{month}{12}
\field{title}{Economic and environmental analysis of hydrogen production when complementing renewable energy generation with grid electricity}
\field{volume}{304}
\field{year}{2021}
\field{pages}{117739}
\range{pages}{1}
\keyw{Hydrogen production; Electrolysis; CO emissions; Clean energy; Levelized cost of hydrogen;Dynamic PEM Paper}
\endentry
\entry{International_Energy_Agency2023-wq}{report}{}
\name{author}{1}{}{%
{{hash=b33b47b3008c8c47f90a77e0c2dd3060}{%
family={{International Energy Agency}},
familyi={I\bibinitperiod}}}%
}
\list{institution}{1}{%
{IEA}%
}
\strng{namehash}{b33b47b3008c8c47f90a77e0c2dd3060}
\strng{fullhash}{b33b47b3008c8c47f90a77e0c2dd3060}
\strng{bibnamehash}{b33b47b3008c8c47f90a77e0c2dd3060}
\strng{authorbibnamehash}{b33b47b3008c8c47f90a77e0c2dd3060}
\strng{authornamehash}{b33b47b3008c8c47f90a77e0c2dd3060}
\strng{authorfullhash}{b33b47b3008c8c47f90a77e0c2dd3060}
\field{sortinit}{4}
\field{sortinithash}{9381316451d1b9788675a07e972a12a7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Global Hydrogen Review}
\field{type}{techreport}
\field{year}{2023}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Carmo2013-bq}{article}{}
\name{author}{4}{}{%
{{hash=5cb00422350f2f49fb84bfa63612d043}{%
family={Carmo},
familyi={C\bibinitperiod},
given={Marcelo},
giveni={M\bibinitperiod}}}%
{{hash=a58e4d85b82d1c1cbf3227df8f015e7f}{%
family={Fritz},
familyi={F\bibinitperiod},
given={David\bibnamedelima L},
giveni={D\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
{{hash=7b4ea7f032a55437297a56d45ef38593}{%
family={Mergel},
familyi={M\bibinitperiod},
given={Jürgen},
giveni={J\bibinitperiod}}}%
{{hash=fcb017aa00b2d8a14aedfe47f774dda4}{%
family={Stolten},
familyi={S\bibinitperiod},
given={Detlef},
giveni={D\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Elsevier BV}%
}
\strng{namehash}{6219a4401b5fe8181b25e667deebe8b5}
\strng{fullhash}{c90eb1bb8a808f1cb40fb8940cd53313}
\strng{bibnamehash}{c90eb1bb8a808f1cb40fb8940cd53313}
\strng{authorbibnamehash}{c90eb1bb8a808f1cb40fb8940cd53313}
\strng{authornamehash}{6219a4401b5fe8181b25e667deebe8b5}
\strng{authorfullhash}{c90eb1bb8a808f1cb40fb8940cd53313}
\field{sortinit}{6}
\field{sortinithash}{b33bc299efb3c36abec520a4c896a66d}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Hydrogen is often considered the best means by which to store energy coming from renewable and intermittent power sources. With the growing capacity of localized renewable energy sources surpassing the gigawatt range, a storage system of equal magnitude is required. PEM electrolysis provides a sustainable solution for the production of hydrogen, and is well suited to couple with energy sources such as wind and solar. However, due to low demand in electrolytic hydrogen in the last century, little research has been done on PEM electrolysis with many challenges still unexplored. The ever increasing desire for green energy has rekindled the interest on PEM electrolysis, thus the compilation and recovery of past research and developments is important and necessary. In this review, PEM water electrolysis is comprehensively highlighted and discussed. The challenges new and old related to electrocatalysts, solid electrolyte, current collectors, separator plates and modeling efforts will also be addressed. The main message is to clearly set the state-of-the-art for the PEM electrolysis technology, be insightful of the research that is already done and the challenges that still exist. This information will provide several future research directions and a road map in order to aid scientists in establishing PEM electrolysis as a commercially viable hydrogen production solution.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{4}
\field{number}{12}
\field{title}{A comprehensive review on {PEM} water electrolysis}
\field{volume}{38}
\field{year}{2013}
\field{pages}{4901\bibrangedash 4934}
\range{pages}{34}
\keyw{Read;Thesis Proposal;Intro to Hydrogen Electrolyzer Project;Dynamic PEM Paper}
\endentry
\entry{Guerra2019-ua}{article}{}
\name{author}{4}{}{%
{{hash=36d24df85f8985ffa64fc638a94ee801}{%
family={Guerra},
familyi={G\bibinitperiod},
given={Omar\bibnamedelima J},
giveni={O\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=ad93758606da968ff0d63bd89049db37}{%
family={Eichman},
familyi={E\bibinitperiod},
given={Joshua},
giveni={J\bibinitperiod}}}%
{{hash=a6c775dcb6e745c643682f119f0034a5}{%
family={Kurtz},
familyi={K\bibinitperiod},
given={Jennifer},
giveni={J\bibinitperiod}}}%
{{hash=b8dbc5062fd92ca92030e69159bd3a6d}{%
family={Hodge},
familyi={H\bibinitperiod},
given={Bri-Mathias},
giveni={B\bibinithyphendelim M\bibinitperiod}}}%
}
\strng{namehash}{329fe7bf08e4cf8dc2b162329ef2a1f6}
\strng{fullhash}{3ad59d2be0f211cdcdf30da217e52298}
\strng{bibnamehash}{3ad59d2be0f211cdcdf30da217e52298}
\strng{authorbibnamehash}{3ad59d2be0f211cdcdf30da217e52298}
\strng{authornamehash}{329fe7bf08e4cf8dc2b162329ef2a1f6}
\strng{authorfullhash}{3ad59d2be0f211cdcdf30da217e52298}
\field{sortinit}{7}
\field{sortinithash}{108d0be1b1bee9773a1173443802c0a3}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Summary Hydrogen is often considered an energy carrier of the future in part because of the costs to implement a hydrogen-based system. However, using 7,182 industrial and commercial US retail electric utility rates, this study dynamically simulates electrolyzer operations under different utility rate structures and shows that electrolysis units can already provide cost-competitive fuel in 20 states operating under 81 retail electricity rates. Our analysis shows that current hydrogen production costs range from US\$2.6 to US\$12.3 kg−1 and can be significantly reduced through flexible operation if dynamic tariffs are used. Hydrogen production costs for the 81 utility rates in 20 states are already less than US\$4.0 kg−1 and competitive with the fuel cost of gasoline vehicles. Thus, under the right conditions, hydrogen fuel is already cost competitive in the US energy sector and has several interesting possible roles to play in future energy and transportation systems.}
\field{journaltitle}{Joule}
\field{month}{10}
\field{number}{10}
\field{title}{Cost Competitiveness of Electrolytic Hydrogen}
\field{volume}{3}
\field{year}{2019}
\field{pages}{2425\bibrangedash 2443}
\range{pages}{19}
\keyw{hydrogen; electrolysis; retail electric utility rates; energy carrier; production cost; fuel cost assessment;Dynamic PEM Paper}
\endentry
\entry{Corengia2022-ha}{article}{}
\name{author}{2}{}{%
{{hash=787b9ebd0819f558956631935d220bea}{%
family={Corengia},
familyi={C\bibinitperiod},
given={Mariana},
giveni={M\bibinitperiod}}}%
{{hash=c4c94d98bc909f1819372cc189d09c47}{%
family={Torres},
familyi={T\bibinitperiod},
given={Ana\bibnamedelima I},
giveni={A\bibinitperiod\bibinitdelim I\bibinitperiod}}}%
}
\strng{namehash}{34b5d95168d7d1cdd2016fdb33a3ecf2}
\strng{fullhash}{34b5d95168d7d1cdd2016fdb33a3ecf2}
\strng{bibnamehash}{34b5d95168d7d1cdd2016fdb33a3ecf2}
\strng{authorbibnamehash}{34b5d95168d7d1cdd2016fdb33a3ecf2}
\strng{authornamehash}{34b5d95168d7d1cdd2016fdb33a3ecf2}
\strng{authorfullhash}{34b5d95168d7d1cdd2016fdb33a3ecf2}
\field{sortinit}{7}
\field{sortinithash}{108d0be1b1bee9773a1173443802c0a3}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Green hydrogen is a promising alternative for decarbonization of the economy. Its production is based on electrolysis powered by energy from renewable sources. Modern renewable sources like wind or solar are of an intermittent nature, a fact that poses a challenge to the operation of the electrolyzers. In this work, we present a superstructure-based optimization framework to design hydrogen production processes. The design involves selection of the power sources (wind/sun generators, grid), selection of the type and size of electrolyzer, and selection of energy storage devices (battery, vessels). The case studies include hydrogen production at different scales to allow a discussion on patterns of technology selection strategies and the synergies between them. Overall, our simulations show that solid oxide electrolyzers are a promising option, and that from the currently market available alternatives, alkaline electrolysis is preferred over proton exchange membrane electrolysis. Our results support the idea that complementary of the pattern of the energy sources should be sought, but not the idea of producing hydrogen to take advantage of energy surplus and avoid curtailment.}
\field{journaltitle}{Chem. Eng. Res. Des.}
\field{month}{7}
\field{title}{Coupling time varying power sources to production of green-hydrogen: A superstructure based approach for technology selection and optimal design}
\field{volume}{183}
\field{year}{2022}
\field{pages}{235\bibrangedash 249}
\range{pages}{15}
\keyw{Green hydrogen; Renewable energy; P2X; P2G; Decarbonization; Electrolysis; Optimal process design;Thesis Proposal;Dynamic PEM Paper}
\endentry
\entry{Chung2024-vr}{article}{}
\name{author}{6}{}{%
{{hash=5a93e2edcb54ecc0c4bb51725cf07068}{%
family={Chung},
familyi={C\bibinitperiod},
given={Doo\bibnamedelima Hyun},
giveni={D\bibinitperiod\bibinitdelim H\bibinitperiod}}}%
{{hash=384e460008a5042bafda2416cf0a5fe6}{%
family={Graham},
familyi={G\bibinitperiod},
given={Edward\bibnamedelima J},
giveni={E\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=86d4a97c4708e8a40d4845cae0cc55e1}{%
family={Paren},
familyi={P\bibinitperiod},
given={Benjamin\bibnamedelima A},
giveni={B\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=65f8f12d9eea501664f91732feacb34b}{%
family={Schofield},
familyi={S\bibinitperiod},
given={Landon},
giveni={L\bibinitperiod}}}%
{{hash=6fae6915ee4792c138348147a18329d1}{%
family={Shao-Horn},
familyi={S\bibinithyphendelim H\bibinitperiod},
given={Yang},
giveni={Y\bibinitperiod}}}%
{{hash=91cbc3b09461b985e4f5269a468e0831}{%
family={Mallapragada},
familyi={M\bibinitperiod},
given={Dharik\bibnamedelima S},
giveni={D\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
}
\list{publisher}{1}{%
{American Chemical Society}%
}
\strng{namehash}{5bba1b0a66de23909591bf9117acaf40}
\strng{fullhash}{78c7cd586b2305e41bb69565627496d2}
\strng{bibnamehash}{78c7cd586b2305e41bb69565627496d2}
\strng{authorbibnamehash}{78c7cd586b2305e41bb69565627496d2}
\strng{authornamehash}{5bba1b0a66de23909591bf9117acaf40}
\strng{authorfullhash}{78c7cd586b2305e41bb69565627496d2}
\field{sortinit}{7}
\field{sortinithash}{108d0be1b1bee9773a1173443802c0a3}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Proton Exchange Membrane (PEM) electrolysis is a promising pathway for producing low-carbon hydrogen via electrolysis coupled to variable renewable energy (VRE). This study introduces a physics-based PEM electrolyzer model into an integrated design and scheduling optimization routine, allowing for a comprehensive evaluation of the impact of reactor level metrics (e.g., cathode pressure and current density) on the levelized cost of hydrogen (LCOH) across various cost, technology, and electricity supply scenarios. Benefits of the static versus dynamic operation of PEM systems are outlined explicitly. The economic viability of a grid-based PEM electrolyzer producing 50,000 kg of hydrogen per day is assessed for both 2021 and 2035 technology and grid scenarios. Results show that dynamic operation reduces the LCOH by 8\% under the 2021 scenario (4.98--4.57 \$/kg-H2 at maximum current density 2 A/cm2). Under 2035 price, cost, and technology assumptions (maximum current density 4 A/cm2), the LCOH ranges between 2.18 and 3.93 \$/kg-H2 under static operation, and between 1.42 and 2.84 \$/kg-H2 under dynamic operation, resulting in LCOH reductions of 20--50\% depending on the electricity price profile. In addition, partial differential pressure mode with a cathode pressure of 5 bar was found to be the most cost-effective way to compress hydrogen to 30 bar in the 2021 scenario, while full differential pressure mode is preferred in the 2035 scenario. Finally, the study revealed that grid-based hydrogen production in 2021 does not meet the carbon intensity (CI) criteria for clean hydrogen in recent U.S. legislation, highlighting the need for additional measures to be considered for grid-connected electrolysis to qualify as ``clean'' hydrogen. These results suggest that capital cost reduction alone will not achieve low-cost electricity-based hydrogen production, emphasizing the need for further reductions in the cost of low-CI electricity to attain affordable and lower-carbon hydrogen production.}
\field{journaltitle}{Ind. Eng. Chem. Res.}
\field{month}{4}
\field{number}{16}
\field{title}{Design Space for {PEM} Electrolysis for {Cost-Effective} {H2} Production Using Grid Electricity}
\field{volume}{63}
\field{year}{2024}
\field{pages}{7258\bibrangedash 7270}
\range{pages}{13}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Tsay2023-se}{article}{}
\name{author}{2}{}{%
{{hash=8c58f45ed8a4bf6b1c804325b86a41b3}{%
family={Tsay},
familyi={T\bibinitperiod},
given={Calvin},
giveni={C\bibinitperiod}}}%
{{hash=8b8385c95a723c415bdf210f89cb7ac3}{%
family={Qvist},
familyi={Q\bibinitperiod},
given={Staffan},
giveni={S\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Wiley}%
}
\strng{namehash}{cc8f54299276f71b933255a3d788344d}
\strng{fullhash}{cc8f54299276f71b933255a3d788344d}
\strng{bibnamehash}{cc8f54299276f71b933255a3d788344d}
\strng{authorbibnamehash}{cc8f54299276f71b933255a3d788344d}
\strng{authornamehash}{cc8f54299276f71b933255a3d788344d}
\strng{authorfullhash}{cc8f54299276f71b933255a3d788344d}
\field{sortinit}{7}
\field{sortinithash}{108d0be1b1bee9773a1173443802c0a3}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{AbstractElectrolysis‐based hydrogen production can play a significant role in industrial decarbonization, and its economic competitiveness can be promoted by designing demand response operating schemes. Nevertheless, the scale of industrial supply plants may be significantly large (on the order of gigawatts), meaning that electricity prices cannot be treated as an input for scheduling problems, that is, the ``price taker'' approach. This article presents a framework for the optimization of a large‐scale, electricity‐powered hydrogen production facility considering its integration with the power grid. Using a computational case study, we present an iterative scheme for integrating the process model with a model for power grid optimization and capacity expansion, taking the popular GenX model as an example.}
\field{journaltitle}{AIChE J.}
\field{month}{12}
\field{number}{12}
\field{title}{Integrating process and power grid models for optimal design and demand response operation of giga‐scale green hydrogen}
\field{volume}{69}
\field{year}{2023}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Mallapragada_undated-iq}{misc}{}
\name{author}{6}{}{%
{{hash=91cbc3b09461b985e4f5269a468e0831}{%
family={Mallapragada},
familyi={M\bibinitperiod},
given={Dharik\bibnamedelima S},
giveni={D\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=12e68872cd7b211a09b3216bf7ca906d}{%
family={Junge},
familyi={J\bibinitperiod},
given={Cristian},
giveni={C\bibinitperiod}}}%
{{hash=a627d0806b9086dffc5cf460185f0e1c}{%
family={Wang},
familyi={W\bibinitperiod},
given={Cathy},
giveni={C\bibinitperiod}}}%
{{hash=b6815778b9d2c71f3bb314f247fe54fb}{%
family={Pfeifenberger},
familyi={P\bibinitperiod},
given={Hannes},
giveni={H\bibinitperiod}}}%
{{hash=9ae0e2136ad02535c951a1b8a14b1553}{%
family={Joskow},
familyi={J\bibinitperiod},
given={Paul\bibnamedelima L},
giveni={P\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
{{hash=4a248c42806779a485fb89901cb0d821}{%
family={Schmalensee},
familyi={S\bibinitperiod},
given={Richard},
giveni={R\bibinitperiod}}}%
}
\strng{namehash}{1efea8002aebe8f40c51d7d08197eb97}
\strng{fullhash}{e9f5abecf4743a5599c0cb05541f69c3}
\strng{bibnamehash}{e9f5abecf4743a5599c0cb05541f69c3}
\strng{authorbibnamehash}{e9f5abecf4743a5599c0cb05541f69c3}
\strng{authornamehash}{1efea8002aebe8f40c51d7d08197eb97}
\strng{authorfullhash}{e9f5abecf4743a5599c0cb05541f69c3}
\field{sortinit}{8}
\field{sortinithash}{a231b008ebf0ecbe0b4d96dcc159445f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{howpublished}{CEEPR Working Papers}
\field{title}{Electricity Pricing Problems in Future {Renewables-Dominant} Power Systems}
\keyw{MIT Energy Fellows Application;Dynamic PEM Paper}
\endentry
\entry{Otashu2019-ge}{article}{}
\name{author}{2}{}{%
{{hash=e69863ac8998f957062543f97c8a4a27}{%
family={Otashu},
familyi={O\bibinitperiod},
given={Joannah\bibnamedelima I},
giveni={J\bibinitperiod\bibinitdelim I\bibinitperiod}}}%
{{hash=04ffda06f7625dd331d89eb7fa707666}{%
family={Baldea},
familyi={B\bibinitperiod},
given={Michael},
giveni={M\bibinitperiod}}}%
}
\strng{namehash}{05cc9112e93ff7ea25a2fee4594ae677}
\strng{fullhash}{05cc9112e93ff7ea25a2fee4594ae677}
\strng{bibnamehash}{05cc9112e93ff7ea25a2fee4594ae677}
\strng{authorbibnamehash}{05cc9112e93ff7ea25a2fee4594ae677}
\strng{authornamehash}{05cc9112e93ff7ea25a2fee4594ae677}
\strng{authorfullhash}{05cc9112e93ff7ea25a2fee4594ae677}
\field{sortinit}{9}
\field{sortinithash}{0a5ebc79d83c96b6579069544c73c7d4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Power-intensive processes can potentially provide significant demand response (DR) services. Modeling such processes for demand response is not trivial as models must depict plant transient properties under highly dynamic operation while remaining computationally efficient. We develop a demand response-oriented model for an important power-intensive process i.e., chlor-alkali production using membrane cells, and demonstrate the provision of fast demand response by an industrial-size plant. Through an extensive simulation and optimization case study, we show that the fast modulation of the cell power demand is possible without adverse impact on cell concentration and temperature. Additionally, the cell temperature dynamics are found to restrict the demand response capacity of the plant and must to be explicitly accounted for to support dynamic cell operation in DR scenarios. Substantial load curtailment during peak electricity price periods can be achieved and the energy cost to the electrolysis plant can be reduced.}
\field{journaltitle}{Comput. Chem. Eng.}
\field{month}{2}
\field{title}{Demand response-oriented dynamic modeling and operational optimization of membrane-based chlor-alkali plants}
\field{volume}{121}
\field{year}{2019}
\field{pages}{396\bibrangedash 408}
\range{pages}{13}
\keyw{Dynamic modeling; Chlor-alkali process; Demand response; Energy storage; Electrolysis;Read;Important;Thesis Proposal;Chlor-Alkali Papers;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Hofmann2022-ti}{article}{}
\name{author}{7}{}{%
{{hash=6f021360fb35a1f46437542b56c4f701}{%
family={Hofmann},
familyi={H\bibinitperiod},
given={Mathias},
giveni={M\bibinitperiod}}}%
{{hash=02ce213245e657ef91390481c2e6a84c}{%
family={Müller},
familyi={M\bibinitperiod},
given={Robert},
giveni={R\bibinitperiod}}}%
{{hash=e596bc2322b839acca40d7158a0c7e88}{%
family={Christidis},
familyi={C\bibinitperiod},
given={Andreas},
giveni={A\bibinitperiod}}}%
{{hash=a55d7c9320ccb32e0eea577c8552f4e9}{%
family={Fischer},
familyi={F\bibinitperiod},
given={Peter},
giveni={P\bibinitperiod}}}%
{{hash=a6897bc2173f096415c54b1a0fa57621}{%
family={Klaucke},
familyi={K\bibinitperiod},
given={Franziska},
giveni={F\bibinitperiod}}}%
{{hash=584605420f865ffec4e3e955604e079d}{%
family={Vomberg},
familyi={V\bibinitperiod},
given={Sebastian},
giveni={S\bibinitperiod}}}%
{{hash=d2895b8b8c7252ddbd95ece56477ffab}{%
family={Tsatsaronis},
familyi={T\bibinitperiod},
given={George},
giveni={G\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Wiley}%
}
\strng{namehash}{038c4feec462f6eb5b5127d72ebc4eec}
\strng{fullhash}{90ba56501697b4689d9914676e1ac66b}
\strng{bibnamehash}{90ba56501697b4689d9914676e1ac66b}
\strng{authorbibnamehash}{90ba56501697b4689d9914676e1ac66b}
\strng{authornamehash}{038c4feec462f6eb5b5127d72ebc4eec}
\strng{authorfullhash}{90ba56501697b4689d9914676e1ac66b}
\field{sortinit}{9}
\field{sortinithash}{0a5ebc79d83c96b6579069544c73c7d4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Abstract Demand response (DR) can compensate for imbalances in variable renewable energy supplies. This possibility is particularly interesting for electrochemical processes, due to their high energy intensity. To determine the technical feasibility and economic viability of DR, we chose the chlor-alkali process with subsequent polyvinyl chloride production, including intermediate storage for ethylene dichloride. We estimate the maximum possible cost savings of implementing load flexibility measures. A process model is set up to determine the system characteristic. Subsequent optimizations result in the facility's best possible dispatch depending on additional and minimum power load, storage volume, and cost of a load change. Real plant data are used to specify model parameters and validate the system characteristic and the plant dispatch. An economic evaluation reveals the economic advantages of efficiency and flexibility. The approach can be used to analyze the DR potential of other chlorine value chains or facilities with high electricity demand in general.}
\field{journaltitle}{AIChE J.}
\field{month}{1}
\field{number}{1}
\field{title}{Flexible and economical operation of chlor‐alkali process with subsequent polyvinyl chloride production}
\field{volume}{68}
\field{year}{2022}
\keyw{Read;Thesis Proposal;Chlor-Alkali Papers;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Weigert2021-ik}{article}{}
\name{author}{5}{}{%
{{hash=943bb15264e60e6ea6c57f993c359910}{%
family={Weigert},
familyi={W\bibinitperiod},
given={Joris},
giveni={J\bibinitperiod}}}%
{{hash=8cadbf1a5e21084a38e5f25460d3ae55}{%
family={Hoffmann},
familyi={H\bibinitperiod},
given={Christian},
giveni={C\bibinitperiod}}}%
{{hash=a0c95c3982bc404dcb93dcdd63f6a10a}{%
family={Esche},
familyi={E\bibinitperiod},
given={Erik},
giveni={E\bibinitperiod}}}%
{{hash=a55d7c9320ccb32e0eea577c8552f4e9}{%
family={Fischer},
familyi={F\bibinitperiod},
given={Peter},
giveni={P\bibinitperiod}}}%
{{hash=1879bc34a91a1302ce331305d3e8848e}{%
family={Repke},
familyi={R\bibinitperiod},
given={Jens-Uwe},
giveni={J\bibinithyphendelim U\bibinitperiod}}}%
}
\strng{namehash}{b28feebbd22962c547aec1f5f9bb326f}
\strng{fullhash}{50b38c0c14c34924fe4e39679df443d6}
\strng{bibnamehash}{50b38c0c14c34924fe4e39679df443d6}
\strng{authorbibnamehash}{50b38c0c14c34924fe4e39679df443d6}
\strng{authornamehash}{b28feebbd22962c547aec1f5f9bb326f}
\strng{authorfullhash}{50b38c0c14c34924fe4e39679df443d6}
\field{sortinit}{9}
\field{sortinithash}{0a5ebc79d83c96b6579069544c73c7d4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{The chlor-alkali electrolysis promises advantageous application of demand-side management and several research groups have proposed dynamic models to obtain optimal trajectories for such applications. However, no dynamic model of those proposed so far has yet been validated with real, industrial plant data. This is highly important to determine trajectories, which are indeed feasible. This contribution addresses this issue by proposing a dynamic model of the chlor-alkali electrolysis that contains those variables relevant for dynamic operation, especially temperature and composition of the electrolytes. This model is validated with real plant data from an electrolysis located in Marl, Germany, for two scenarios: (1) several small load changes of around 4 MW and (2) a large load change of more than 50 MW.}
\field{journaltitle}{Comput. Chem. Eng.}
\field{month}{6}
\field{title}{Towards demand-side management of the chlor-alkali electrolysis: Dynamic modeling and model validation}
\field{volume}{149}
\field{year}{2021}
\field{pages}{107287}
\range{pages}{1}
\keyw{Chlor-alkali electrolysis; Demand response; Demand-side management; Flexible operation; Dynamic modeling;Thesis Proposal;Dynamic PEM Paper}
\endentry
\entry{Espinosa-Lopez2018-hb}{article}{}
\name{author}{8}{}{%
{{hash=7774e7f30d364d58e5bd8ff0139ee393}{%
family={Espinosa-López},
familyi={E\bibinithyphendelim L\bibinitperiod},
given={Manuel},
giveni={M\bibinitperiod}}}%
{{hash=7582beba9fec7b1976502aab24df91a0}{%
family={Darras},
familyi={D\bibinitperiod},
given={Christophe},
giveni={C\bibinitperiod}}}%
{{hash=e91eb0e1e762a15d7eb2e69408a190a0}{%
family={Poggi},
familyi={P\bibinitperiod},
given={Philippe},
giveni={P\bibinitperiod}}}%
{{hash=d38d274dcf4c5ca286c7f968fc7a8895}{%
family={Glises},
familyi={G\bibinitperiod},
given={Raynal},
giveni={R\bibinitperiod}}}%
{{hash=15636484f76af0d43b3a04828f276c87}{%
family={Baucour},
familyi={B\bibinitperiod},
given={Philippe},
giveni={P\bibinitperiod}}}%
{{hash=64807a335a3f61004b23a970fc71539b}{%
family={Rakotondrainibe},
familyi={R\bibinitperiod},
given={André},
giveni={A\bibinitperiod}}}%
{{hash=3f3572481ce4d2081043efe58f0fcefc}{%
family={Besse},
familyi={B\bibinitperiod},
given={Serge},
giveni={S\bibinitperiod}}}%
{{hash=cc5c3ac331353b522ce9b4a860924008}{%
family={Serre-Combe},
familyi={S\bibinithyphendelim C\bibinitperiod},
given={Pierre},
giveni={P\bibinitperiod}}}%
}
\strng{namehash}{9452a6d08702278c625147ecf198792b}
\strng{fullhash}{7437c987b4759cbcbfa8756528718cd3}
\strng{bibnamehash}{7437c987b4759cbcbfa8756528718cd3}
\strng{authorbibnamehash}{7437c987b4759cbcbfa8756528718cd3}
\strng{authornamehash}{9452a6d08702278c625147ecf198792b}
\strng{authorfullhash}{7437c987b4759cbcbfa8756528718cd3}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{The objective of this paper is to present the modelling and experimental validation of the 46 kW PEM high pressure water electrolyzer installed on the MYRTE platform, which is a real-scale demonstrator that aims to study the deployment of hydrogen to store the energy associated to intermittent renewable energy source systems. An electrochemical steady-state and semi-empirical submodel coupled with a lumped thermal capacitance dynamic submodel is developed to predict the stack voltage and the stack temperature evolution from instantaneous operating conditions such as the applied current, the gas storage pressure tanks (H2 and O2) and the ambient temperature. The Particle Swarm Optimization algorithm is used to find the electrochemical submodel parameters and a multivariable Matlab-Simulink®{} linked modular mathematical model is developed for validation. Results indicate that within a temperature range of 20--60 °C, and a pressure range of 15--35 bar, the stack voltage and the temperature evolution can be predicted even in transitory operating phases. The strategy used for the parameters identification is explained in detail and can be applied to any PEM water electrolyzer.}
\field{journaltitle}{Renewable Energy}
\field{month}{4}
\field{title}{Modelling and experimental validation of a 46 kW {PEM} high pressure water electrolyzer}
\field{volume}{119}
\field{year}{2018}
\field{pages}{160\bibrangedash 173}
\range{pages}{14}
\keyw{Electrolyzer; Hydrogen; Renewable energy; Photovoltaic hybrid system;Important;Reference;Thesis Proposal;First Principles Model;PEM Model Documentation;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Stansberry2020-zm}{article}{}
\name{author}{2}{}{%
{{hash=04e789b2253e9c2742e2ef1d7398a294}{%
family={Stansberry},
familyi={S\bibinitperiod},
given={John\bibnamedelima M},
giveni={J\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=bac6f9a64636de0932b198c223cfc98f}{%
family={Brouwer},
familyi={B\bibinitperiod},
given={Jacob},
giveni={J\bibinitperiod}}}%
}
\strng{namehash}{2fcc41c09873d6ca3e7c8a2fdaa80fba}
\strng{fullhash}{2fcc41c09873d6ca3e7c8a2fdaa80fba}
\strng{bibnamehash}{2fcc41c09873d6ca3e7c8a2fdaa80fba}
\strng{authorbibnamehash}{2fcc41c09873d6ca3e7c8a2fdaa80fba}
\strng{authornamehash}{2fcc41c09873d6ca3e7c8a2fdaa80fba}
\strng{authorfullhash}{2fcc41c09873d6ca3e7c8a2fdaa80fba}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{A 60 kW PEM electrolyzer was modified to have dynamic dispatch capabilities through the use of an external mass flow controller and was subsequently operated and studied in detail as a part of the UC Irvine power-to-gas (P2G) demonstration project. The system operated in load following for both rooftop solar PV output and aggregated wind farm power. The electrolyzer system was able to handle 100\% up and down ramps of the electrolyzer stack in sub-second intervals during variable renewable energy load following experiments. Overall system efficiency was improved by imposing a minimum load condition to avoid high parasitic losses at low power conditions, and cycling the system off/on as required and enabled by quick cold start-up capability. Performance was characterized for varying levels of part load and dynamic operation. The gas drying process was found to be a significant source of loss at low power conditions.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{3}
\field{number}{16}
\field{title}{Experimental dynamic dispatch of a 60 kW proton exchange membrane electrolyzer in power-to-gas application}
\field{volume}{45}
\field{year}{2020}
\field{pages}{9305\bibrangedash 9316}
\range{pages}{12}
\keyw{Proton exchange membrane electrolyzer; Solar photovoltaic; Power-to-gas; Wind power; Electrolysis;Thesis Proposal;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Corengia2020-on}{article}{}
\name{author}{3}{}{%
{{hash=787b9ebd0819f558956631935d220bea}{%
family={Corengia},
familyi={C\bibinitperiod},
given={Mariana},
giveni={M\bibinitperiod}}}%
{{hash=08f5c0ab7d29152c509d4c185a8fe1cd}{%
family={Estefan},
familyi={E\bibinitperiod},
given={Nicolás},
giveni={N\bibinitperiod}}}%
{{hash=c4c94d98bc909f1819372cc189d09c47}{%
family={Torres},
familyi={T\bibinitperiod},
given={Ana\bibnamedelima I},
giveni={A\bibinitperiod\bibinitdelim I\bibinitperiod}}}%
}
\name{editor}{4}{}{%
{{hash=9b8473c304986d996ec97e7cccfaf371}{%
family={Pierucci},
familyi={P\bibinitperiod},
given={Sauro},
giveni={S\bibinitperiod}}}%
{{hash=0e47a9a3accbe5b3428f8a6814e050c2}{%
family={Manenti},
familyi={M\bibinitperiod},
given={Flavio},
giveni={F\bibinitperiod}}}%
{{hash=d8c4026feaaec4addc0e1e8a5db2b5c6}{%
family={Bozzano},
familyi={B\bibinitperiod},
given={Giulia\bibnamedelima Luisa},
giveni={G\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
{{hash=fded9f1b0bc983ca22d3297cc2bbec24}{%
family={Manca},
familyi={M\bibinitperiod},
given={Davide},
giveni={D\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Elsevier}%
}
\strng{namehash}{a161dfcc0934107f6defbd01afddd8b0}
\strng{fullhash}{a161dfcc0934107f6defbd01afddd8b0}
\strng{bibnamehash}{a161dfcc0934107f6defbd01afddd8b0}
\strng{authorbibnamehash}{a161dfcc0934107f6defbd01afddd8b0}
\strng{authornamehash}{a161dfcc0934107f6defbd01afddd8b0}
\strng{authorfullhash}{a161dfcc0934107f6defbd01afddd8b0}
\strng{editorbibnamehash}{eaf7d925b87ddbca7ea33adb6cd39a14}
\strng{editornamehash}{81156f045c65ae2e84eca7e2975727d1}
\strng{editorfullhash}{eaf7d925b87ddbca7ea33adb6cd39a14}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Modern renewable energy sources such as wind or sunlight cannot be programmed to be available when needed. A robust design of electrical grids with high penetration of these usually imply the existence of energy surplus. This surplus can be taken advantage of by seeking integration with fuel hydrogen production systems. However, the surplus is neither constant nor periodic, hence there is a trade-off between usage resource and economic goals. For a given energy surplus vs time curve, this paper presents three different decision-making problems to find the optimal hydrogen production capacity that needs to be installed, and the associated hydrogen prices.}
\field{month}{1}
\field{title}{Analyzing Hydrogen Production Capacities to Seize Renewable Energy Surplus}
\field{volume}{48}
\field{year}{2020}
\field{pages}{1549\bibrangedash 1554}
\range{pages}{6}
\keyw{renewable energy surplus; power to hydrogen; green fuels; optimization;Dynamic PEM Paper}
\endentry
\entry{Holst2021-nd}{report}{}
\name{author}{5}{}{%
{{hash=dfe124ceee483eee1893334d44a17691}{%
family={Holst},
familyi={H\bibinitperiod},
given={Marius},
giveni={M\bibinitperiod}}}%
{{hash=65765dbb2d27cdd996d2987d1378dcc5}{%
family={Aschbrenner},
familyi={A\bibinitperiod},
given={Stefan},
giveni={S\bibinitperiod}}}%
{{hash=d28ec3d4e7c8e912c5d1b0c4a65084f3}{%
family={Smolinka},
familyi={S\bibinitperiod},
given={Tom},
giveni={T\bibinitperiod}}}%
{{hash=3a53d73301a834ea97c92a782fb14360}{%
family={Voglstatter},
familyi={V\bibinitperiod},
given={Christopher},
giveni={C\bibinitperiod}}}%
{{hash=04fc40cf75b63a4a720ebeafc14f17b8}{%
family={Grimm},
familyi={G\bibinitperiod},
given={Gunter},
giveni={G\bibinitperiod}}}%
}
\list{institution}{1}{%
{Fraunhofer Institute for Solar Energy Systems}%
}
\strng{namehash}{3e25340c89ce9a0e8469cca2b106604d}
\strng{fullhash}{25cc4bf9d5c4e50e829ca7374c7238d0}
\strng{bibnamehash}{25cc4bf9d5c4e50e829ca7374c7238d0}
\strng{authorbibnamehash}{25cc4bf9d5c4e50e829ca7374c7238d0}
\strng{authornamehash}{3e25340c89ce9a0e8469cca2b106604d}
\strng{authorfullhash}{25cc4bf9d5c4e50e829ca7374c7238d0}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{10}
\field{title}{Cost Forecast for Low Temperature Electrolysis - Technology Driven {Bottom-Up} Prognosis for {PEM} and Alkaline Water Electrolysis Systems}
\field{type}{techreport}
\field{year}{2021}
\keyw{Thesis Proposal;TEA;Dynamic PEM Paper}
\endentry
\entry{Bernt2020-af}{article}{}
\name{author}{4}{}{%
{{hash=42f5e61f9bfea36a87738c04d6c7e8fb}{%
family={Bernt},
familyi={B\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=dedf87e9e46ba2b772e2a6e5fceaafac}{%
family={Schröter},
familyi={S\bibinitperiod},
given={J},
giveni={J\bibinitperiod}}}%
{{hash=0f78499390c0ba16de1c2142fd4e7792}{%
family={Möckl},
familyi={M\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=05008cfe2fe0e12a383f5e7b100b9f5a}{%
family={Gasteiger},
familyi={G\bibinitperiod},
given={H\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{35c8fc4e55b15a2d80412860b5a05bd6}
\strng{fullhash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\strng{bibnamehash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\strng{authorbibnamehash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\strng{authornamehash}{35c8fc4e55b15a2d80412860b5a05bd6}
\strng{authorfullhash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{In this study, on-line mass spectrometry is used to determine hydrogen permeation during proton exchange membrane water electrolyzer (PEM-WE) operation for a wide range of current densities (0--6 A cm−2) and operating pressures (1--30 bar, differential pressure). H2 permeation measurements with a permeation cell setup, i.e., without applying a current, show a linear correlation between permeation rate and H2 partial pressure, indicating diffusion as the main crossover mechanism. Measurements with full membrane electrode assemblies (MEAs) during PEM-WE operation reveal a significant increase of the gas permeation rate at high current densities, by up to $\approx$20-fold at 1 bar H2 and up to $\approx$1.2-fold at 30 bar H2 (Nafion®{} 212 or Nafion®{} 117 membrane; Ir-black (anode) and Pt/C (cathode)). Recently, H2 super-saturation of the ionomer phase in the cathode catalyst layer was shown to be the reason for this increase, and we discuss the impact of this effect for different electrode compositions and operating conditions. Finally, the determined H2 permeation rates and electrolyzer performance are used to discuss the overall PEM-WE efficiency for different membrane thicknesses and it is shown that the formation of an explosive gas mixture in the anode at low current densities requires additional mitigation strategies.}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{8}
\field{number}{12}
\field{title}{Analysis of Gas Permeation Phenomena in a {PEM} Water Electrolyzer Operated at High Pressure and High Current Density}
\field{volume}{167}
\field{year}{2020}
\field{pages}{124502}
\range{pages}{1}
\keyw{Read;Important;Reference;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Christopher_Bryce_Capuano_Morgan_Elizabeth_Pertosos_Nemanja_Danilovic2018-lw}{patent}{}
\name{author}{1}{}{%
{{hash=da7e29db7fd550d698ea303079dd7619}{%
family={{Christopher Bryce Capuano, Morgan Elizabeth Pertosos, Nemanja Danilovic}},
familyi={C\bibinitperiod}}}%
}
\strng{namehash}{da7e29db7fd550d698ea303079dd7619}
\strng{fullhash}{da7e29db7fd550d698ea303079dd7619}
\strng{bibnamehash}{da7e29db7fd550d698ea303079dd7619}
\strng{authorbibnamehash}{da7e29db7fd550d698ea303079dd7619}
\strng{authornamehash}{da7e29db7fd550d698ea303079dd7619}
\strng{authorfullhash}{da7e29db7fd550d698ea303079dd7619}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{3}
\field{number}{2018/0062192 A1}
\field{title}{Membrane Electrode Assembly and Method of Making the Same}
\field{year}{2018}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Nicolas_Guillet2014-tq}{patent}{}
\name{author}{1}{}{%
{{hash=fb9b9d452af2afb9c79844102a44845d}{%
family={Nicolas\bibnamedelima Guillet},
familyi={N\bibinitperiod\bibinitdelim G\bibinitperiod},
given={Eric\bibnamedelima Mayousse},
giveni={E\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
}
\strng{namehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{fullhash}{fb9b9d452af2afb9c79844102a44845d}
\strng{bibnamehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{authorbibnamehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{authornamehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{authorfullhash}{fb9b9d452af2afb9c79844102a44845d}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{5}
\field{number}{20140116877 A1}
\field{title}{Membrane Electrode Assembly for an Electrolysis Device}
\field{year}{2014}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Hartig-Weis2021-mi}{article}{}
\name{author}{4}{}{%
{{hash=c177d833d21af17e52300051608f23f1}{%
family={Hartig-Wei{ß}},
familyi={H\bibinithyphendelim W\bibinitperiod},
given={Alexandra},
giveni={A\bibinitperiod}}}%
{{hash=4f47c6343f444c3e9e61b6d1cfaec14b}{%
family={Bernt},
familyi={B\bibinitperiod},
given={Maximilian},
giveni={M\bibinitperiod}}}%
{{hash=985d96e55c19d7e462d2ade4bdf4c78c}{%
family={Siebel},
familyi={S\bibinitperiod},
given={Armin},
giveni={A\bibinitperiod}}}%
{{hash=b43fb849dae1a0b0667cba868dd9999d}{%
family={Gasteiger},
familyi={G\bibinitperiod},
given={Hubert\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{ef2a43289bbd29a7c60b4f2f079adfd4}
\strng{fullhash}{39bf14cc1135683c34b080190201a407}
\strng{bibnamehash}{39bf14cc1135683c34b080190201a407}
\strng{authorbibnamehash}{39bf14cc1135683c34b080190201a407}
\strng{authornamehash}{ef2a43289bbd29a7c60b4f2f079adfd4}
\strng{authorfullhash}{39bf14cc1135683c34b080190201a407}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{We present a platinum wire micro-reference electrode (Pt-WRE) suitable for recording individual electrochemical impedance spectra of both the anode and the cathode in a proton exchange membrane water electrolyzer (PEM-WE). For this purpose, a thin, insulated Pt-wire reference electrode (Pt-WRE) was laminated centrally between two 50 $\mu$m Nafion®{} membranes, whereby the potential of the Pt-WRE is determined by the ratio of the local H2 and O2 permeation fluxes at the tip of the Pt-WRE. Impedance analysis with the Pt-WRE allows determination of the proton sheet resistance of the anode, the anode catalyst layer capacitance, and the high-frequency resistance (HFR) of both electrodes individually, using a simple transmission-line model. This new diagnostic tool was used to analyze performance degradation during an accelerated stress test (AST), where low and high current densities were alternated with idle periods without current (i.e., at open circuit voltage (OCV)), mimicking the fluctuating operation of a PEM-WE with renewable energy. Our analysis revealed that the increasing HFR that was observed over the course of the OCV-AST, which is the main cause for the observed performance decrease, can unequivocally be assigned to an increasing contact resistance between the anode electrode and the porous transport layer.}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{11}
\field{number}{11}
\field{title}{A Platinum {Micro-Reference} Electrode for Impedance Measurements in a {PEM} Water Electrolysis Cell}
\field{volume}{168}
\field{year}{2021}
\field{pages}{114511}
\range{pages}{1}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Feng2017-kz}{article}{}
\name{author}{7}{}{%
{{hash=60d32e5e5c0aeb219b221bc422c03542}{%
family={Feng},
familyi={F\bibinitperiod},
given={Qi},
giveni={Q\bibinitperiod}}}%
{{hash=734c984ed06279636354721fd98498cb}{%
family={Yuan},
familyi={Y\bibinitperiod},
given={Xiao−zi},
giveni={X\bibinithyphendelim z\bibinitperiod}}}%
{{hash=2ac43f5187d5174dde2165be9a863303}{%
family={Liu},
familyi={L\bibinitperiod},
given={Gaoyang},
giveni={G\bibinitperiod}}}%
{{hash=c36051632ea6fabe08cf0989078c32bf}{%
family={Wei},
familyi={W\bibinitperiod},
given={Bing},
giveni={B\bibinitperiod}}}%
{{hash=4be62463c6bbd4876f8fceaf19e1edcf}{%
family={Zhang},
familyi={Z\bibinitperiod},
given={Zhen},
giveni={Z\bibinitperiod}}}%
{{hash=9b1528d0b6c9bc88bfd8c569f290bcec}{%
family={Li},
familyi={L\bibinitperiod},
given={Hui},
giveni={H\bibinitperiod}}}%
{{hash=89cf2f4b34604ebf324e9ace01926e21}{%
family={Wang},
familyi={W\bibinitperiod},
given={Haijiang},
giveni={H\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Elsevier BV}%
}
\strng{namehash}{4ceda0929793eac7e077b4d3d02f9505}
\strng{fullhash}{e25500897637db5a6bb7f219cb8aa9a6}
\strng{bibnamehash}{e25500897637db5a6bb7f219cb8aa9a6}
\strng{authorbibnamehash}{e25500897637db5a6bb7f219cb8aa9a6}
\strng{authornamehash}{4ceda0929793eac7e077b4d3d02f9505}
\strng{authorfullhash}{e25500897637db5a6bb7f219cb8aa9a6}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Proton exchange membrane water electrolysis (PEMWE) is an advanced and effective solution to the primary energy storage technologies. A better understanding of performance and durability of PEMWE is critical for the engineers and researchers to further advance this technology for its market penetration, and for the manufacturers of PEM water electrolyzers to implement quality control procedures for the production line or on-site process monitoring/diagnosis. This paper reviews the published works on performance degradations and mitigation strategies for PEMWE. Sources of degradation for individual components are introduced. With degradation causes discussed and degradation mechanisms examined, the review emphasizes on feasible strategies to mitigate the components degradation. To avoid lengthy real lifetime degradation tests and their high costs, the importance of accelerated stress tests and protocols is highlighted for various components. In the end, R\&D directions are proposed to move the PEMWE technology forward to become a key element in future energy scenarios.}
\field{journaltitle}{J. Power Sources}
\field{month}{10}
\field{title}{A review of proton exchange membrane water electrolysis on degradation mechanisms and mitigation strategies}
\field{volume}{366}
\field{year}{2017}
\field{pages}{33\bibrangedash 55}
\range{pages}{23}
\keyw{Read;Thesis Proposal;Intro to Hydrogen Electrolyzer Project;Dynamic PEM Paper}
\endentry
\entry{Weis2019-vj}{article}{}
\name{author}{6}{}{%
{{hash=ec417d20b0905cde8a81229c7b7808bd}{%
family={Wei{ß}},
familyi={W\bibinitperiod},
given={A},
giveni={A\bibinitperiod}}}%
{{hash=6a21609a80939dafdf4a5973a62e7ce3}{%
family={Siebel},
familyi={S\bibinitperiod},
given={A},
giveni={A\bibinitperiod}}}%
{{hash=42f5e61f9bfea36a87738c04d6c7e8fb}{%
family={Bernt},
familyi={B\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=e3ca975a973f21ad8f90f4b795e9f6dd}{%
family={Shen},
familyi={S\bibinitperiod},
given={T-H},
giveni={T\bibinithyphendelim H\bibinitperiod}}}%
{{hash=08557a55942f2c1565ea14aee09eb794}{%
family={Tileli},
familyi={T\bibinitperiod},
given={V},
giveni={V\bibinitperiod}}}%
{{hash=05008cfe2fe0e12a383f5e7b100b9f5a}{%
family={Gasteiger},
familyi={G\bibinitperiod},
given={H\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{ef9ced13dfad4ac97eead3ef83d3a819}
\strng{fullhash}{5776de29dd445716a78bb5cd070d2178}
\strng{bibnamehash}{5776de29dd445716a78bb5cd070d2178}
\strng{authorbibnamehash}{5776de29dd445716a78bb5cd070d2178}
\strng{authornamehash}{ef9ced13dfad4ac97eead3ef83d3a819}
\strng{authorfullhash}{5776de29dd445716a78bb5cd070d2178}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{4}
\field{number}{8}
\field{title}{Impact of Intermittent Operation on Lifetime and Performance of a {PEM} Water Electrolyzer}
\field{volume}{166}
\field{year}{2019}
\field{pages}{F487}
\range{pages}{-1}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz}{report}{}
\name{author}{1}{}{%
{{hash=4125179660bd51242924b976b94e70ca}{%
family={{Brian D James, Daniel A DeSantis, Genevieve Saur}},
familyi={B\bibinitperiod}}}%
}
\list{institution}{1}{%
{Strategic Analysis Inc}%
}
\strng{namehash}{4125179660bd51242924b976b94e70ca}
\strng{fullhash}{4125179660bd51242924b976b94e70ca}
\strng{bibnamehash}{4125179660bd51242924b976b94e70ca}
\strng{authorbibnamehash}{4125179660bd51242924b976b94e70ca}
\strng{authornamehash}{4125179660bd51242924b976b94e70ca}
\strng{authorfullhash}{4125179660bd51242924b976b94e70ca}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{9}
\field{title}{Final Report: Hydrogen Production Pathways Cost Analysis (2013 -- 2016)}
\field{type}{techreport}
\field{year}{2016}
\keyw{Thesis Proposal References;AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Suermann2019-kz}{article}{}
\name{author}{3}{}{%
{{hash=e87a77f733524fd32c69afd9322d89a9}{%
family={Suermann},
familyi={S\bibinitperiod},
given={Michel},
giveni={M\bibinitperiod}}}%
{{hash=1fadf57f9ef9e20601eb2a2a5b9f6d83}{%
family={Bensmann},
familyi={B\bibinitperiod},
given={Boris},
giveni={B\bibinitperiod}}}%
{{hash=88f731b9d9db30338a9ff363ad03c699}{%
family={Hanke-Rauschenbach},
familyi={H\bibinithyphendelim R\bibinitperiod},
given={Richard},
giveni={R\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{The Electrochemical Society}%
}
\strng{namehash}{749743bcfd584a9ff528d7b501ae6201}
\strng{fullhash}{749743bcfd584a9ff528d7b501ae6201}
\strng{bibnamehash}{749743bcfd584a9ff528d7b501ae6201}
\strng{authorbibnamehash}{749743bcfd584a9ff528d7b501ae6201}
\strng{authornamehash}{749743bcfd584a9ff528d7b501ae6201}
\strng{authorfullhash}{749743bcfd584a9ff528d7b501ae6201}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{J. Electrochem. Soc.}
\field{number}{10}
\field{title}{Degradation of proton exchange membrane ({PEM}) water electrolysis cells: Looking beyond the cell voltage increase}
\field{volume}{166}
\field{year}{2019}
\field{pages}{F645\bibrangedash F652}
\range{pages}{-1}
\keyw{AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Rakousky2017-te}{article}{}
\name{author}{7}{}{%
{{hash=204c836155ae754d98dccb486489f502}{%
family={Rakousky},
familyi={R\bibinitperiod},
given={Christoph},
giveni={C\bibinitperiod}}}%
{{hash=1cb1ed02a9085847f9854800a915a943}{%
family={Reimer},
familyi={R\bibinitperiod},
given={Uwe},
giveni={U\bibinitperiod}}}%
{{hash=040c60f99446878bccc9c21c216a9863}{%
family={Wippermann},
familyi={W\bibinitperiod},
given={Klaus},
giveni={K\bibinitperiod}}}%
{{hash=d6b3c2a769db983279b064b7606de794}{%
family={Kuhri},
familyi={K\bibinitperiod},
given={Susanne},
giveni={S\bibinitperiod}}}%
{{hash=5cb00422350f2f49fb84bfa63612d043}{%
family={Carmo},
familyi={C\bibinitperiod},
given={Marcelo},
giveni={M\bibinitperiod}}}%
{{hash=4794f1fa11b1c13f32f1a91cf7967b4a}{%
family={Lueke},
familyi={L\bibinitperiod},
given={Wiebke},
giveni={W\bibinitperiod}}}%
{{hash=fcb017aa00b2d8a14aedfe47f774dda4}{%
family={Stolten},
familyi={S\bibinitperiod},
given={Detlef},
giveni={D\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Elsevier BV}%
}
\strng{namehash}{b0b07ea3fe5d689f06315ca96e2e4e64}
\strng{fullhash}{182ec4d02ab62fd8a5e3c75d163f37e8}
\strng{bibnamehash}{182ec4d02ab62fd8a5e3c75d163f37e8}
\strng{authorbibnamehash}{182ec4d02ab62fd8a5e3c75d163f37e8}
\strng{authornamehash}{b0b07ea3fe5d689f06315ca96e2e4e64}
\strng{authorfullhash}{182ec4d02ab62fd8a5e3c75d163f37e8}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Polymer electrolyte membrane (PEM) water electrolysis generates `green' hydrogen when conducted with electricity from renewable - but fluctuating - sources like wind or solar photovoltaic. Unfortunately, the long-term stability of the electrolyzer performance is still not fully understood under these input power profiles. In this study, we contrast the degradation behavior of our PEM water electrolysis single cells that occurs under operation with constant and intermittent power and derive preferable operating states. For this purpose, five different current density profiles are used, of which two were constant and three dynamic. Cells operated at 1 A cm−2 show no degradation. However, degradation was observed for the remaining four profiles, all of which underwent periods of high current density (2 A cm−2). Hereby, constant operation at 2 A cm−2 led to the highest degradation rate (194 $\mu$V h−1). Degradation can be greatly reduced when the cells are operated with an intermittent profile. Current density switching has a positive effect on durability, as it causes reversible parts of degradation to recover and results in a substantially reduced degradation per mole of hydrogen produced. Two general degradation phenomena were identified, a decreased anode exchange current density and an increased contact resistance at the titanium porous transport layer (Ti-PTL).}
\field{journaltitle}{J. Power Sources}
\field{month}{2}
\field{title}{Polymer electrolyte membrane water electrolysis: Restraining degradation in the presence of fluctuating power}
\field{volume}{342}
\field{year}{2017}
\field{pages}{38\bibrangedash 47}
\range{pages}{10}
\keyw{AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Papakonstantinou2020-hj}{article}{}
\name{author}{6}{}{%
{{hash=0bda7d286be838df184ce9233e01eb2f}{%
family={Papakonstantinou},
familyi={P\bibinitperiod},
given={Georgios},
giveni={G\bibinitperiod}}}%
{{hash=23f6dbc309a62b5d0c3757956e878be1}{%
family={Algara-Siller},
familyi={A\bibinithyphendelim S\bibinitperiod},
given={Gerardo},
giveni={G\bibinitperiod}}}%
{{hash=a0c50ce538c8a27fd24747ae3fe57546}{%
family={Teschner},
familyi={T\bibinitperiod},
given={Detre},
giveni={D\bibinitperiod}}}%
{{hash=2896723d460de85d78d5c3da3cd4d47b}{%
family={Vidaković-Koch},
familyi={V\bibinithyphendelim K\bibinitperiod},
given={Tanja},
giveni={T\bibinitperiod}}}%
{{hash=7d74d011fb0161695e222884a82e03f2}{%
family={Schlögl},
familyi={S\bibinitperiod},
given={Robert},
giveni={R\bibinitperiod}}}%
{{hash=a5148fca6c6bde4ba775cee31842e0b4}{%
family={Sundmacher},
familyi={S\bibinitperiod},
given={Kai},
giveni={K\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Elsevier BV}%
}
\strng{namehash}{07c1ffd0c61492a5710c0cbf660efbd9}
\strng{fullhash}{e3d584d95d88996f3034665349f7744e}
\strng{bibnamehash}{e3d584d95d88996f3034665349f7744e}
\strng{authorbibnamehash}{e3d584d95d88996f3034665349f7744e}
\strng{authornamehash}{07c1ffd0c61492a5710c0cbf660efbd9}
\strng{authorfullhash}{e3d584d95d88996f3034665349f7744e}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Appl. Energy}
\field{month}{12}
\field{number}{115911}
\field{title}{Degradation study of a proton exchange membrane water electrolyzer under dynamic operation conditions}
\field{volume}{280}
\field{year}{2020}
\field{pages}{115911}
\range{pages}{1}
\keyw{AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Alia2019-lw}{article}{}
\name{author}{3}{}{%
{{hash=bfe7ab67336247d32cc88c3e6c9d5c96}{%
family={Alia},
familyi={A\bibinitperiod},
given={Shaun\bibnamedelima M},
giveni={S\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=4db86ea50eeeb6cd9265b02b509a3afe}{%
family={Stariha},
familyi={S\bibinitperiod},
given={Sarah},
giveni={S\bibinitperiod}}}%
{{hash=49e5e1746b37ccc2f4ad328f0282374e}{%
family={Borup},
familyi={B\bibinitperiod},
given={Rod\bibnamedelima L},
giveni={R\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{c103b47e32d79bd9d697c1aa3f85910d}
\strng{fullhash}{c103b47e32d79bd9d697c1aa3f85910d}
\strng{bibnamehash}{c103b47e32d79bd9d697c1aa3f85910d}
\strng{authorbibnamehash}{c103b47e32d79bd9d697c1aa3f85910d}
\strng{authornamehash}{c103b47e32d79bd9d697c1aa3f85910d}
\strng{authorfullhash}{c103b47e32d79bd9d697c1aa3f85910d}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Membrane electrode assembly durability is explored for polymer electrolyte membrane electrolyzers, focusing on catalyst (iridium, Ir) degradation at low loading and dynamic operation. Low catalyst loading and high cell potential are critical to observing durability losses over reasonably short experiments, regardless of test profile. While small losses are seen during steady operation, cycling greatly accelerates performance decreases. Ir dissolution mechanistically drives performance loss, thinning the anode catalyst layer and resulting in increasing kinetic losses during extended operation. While morphological changes to the catalyst layer are found, increasing polarization resistance suggests that degradation at the catalyst/ionomer/membrane interface may also contribute. Electrolyzer operation with model wind and solar profiles results in less severe performance losses compared to triangle- and square-wave potential cycling due to the lower cycling frequency of the renewable profiles. However, in both cases kinetics dominated the loss, indicating that higher cycling rates accelerate loss and can be used to project the impact of intermittency on device lifetime. These results suggest that performance losses impact electrolyzers' abilities to operate with low catalyst loading and intermittent inputs, and that a combination of component development and system controls are needed to limit potential and performance loss.}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{10}
\field{number}{15}
\field{title}{Electrolyzer Durability at Low Catalyst Loading and with Dynamic Operation}
\field{volume}{166}
\field{year}{2019}
\field{pages}{F1164}
\range{pages}{-1}
\keyw{Important;Unread;AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Li2021-nj}{article}{}
\name{author}{3}{}{%
{{hash=0c3bb2625d92599a7df7d96021e7219d}{%
family={Li},
familyi={L\bibinitperiod},
given={Na},
giveni={N\bibinitperiod}}}%
{{hash=d7bdd613e3085de0f89098327c7efd1d}{%
family={Araya},
familyi={A\bibinitperiod},
given={Samuel\bibnamedelima Simon},
giveni={S\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=5004d188295853e5aa42c4aef2b284c3}{%
family={K{æ}r},
familyi={K\bibinitperiod},
given={S{ø}ren\bibnamedelima Knudsen},
giveni={S\bibinitperiod\bibinitdelim K\bibinitperiod}}}%
}
\strng{namehash}{eb0aed20b0c0cb0712bcd295e4f57cba}
\strng{fullhash}{eb0aed20b0c0cb0712bcd295e4f57cba}
\strng{bibnamehash}{eb0aed20b0c0cb0712bcd295e4f57cba}
\strng{authorbibnamehash}{eb0aed20b0c0cb0712bcd295e4f57cba}
\strng{authornamehash}{eb0aed20b0c0cb0712bcd295e4f57cba}
\strng{authorfullhash}{eb0aed20b0c0cb0712bcd295e4f57cba}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{In this paper, novel accelerated stress protocols have been proposed based on polarization and electrochemical impedance spectroscopy characterization conducted on three single cell assemblies: one for reference at constant load of 1 A cm −2 and two dynamic loads consisting of a high load cycling between 1.2 A cm −2 and 2 A cm −2 and a low load cycling between 0 A cm −2 (open circuit voltage) and 0.5 A cm −2. Results showed that the degradation rate is 29.8 µV h−1 for constant load test, 51.4 µV h−1 for low load cycling test and -55.8 µV h−1 for high load cycling test. Compared to the constant load operation, low load cycling operation could accelerate the cell performance decrease mainly due to cathode catalyst degradation. The high load cycling could lead to increased cell performance by reducing the ohmic resistance but degrades the anode catalyst significantly, which could compromise the cell's long-term stability. Besides, it is believed that this dynamic condition could accelerate the cell aging by accelerating the membrane thinning.}
\field{journaltitle}{Electrochim. Acta}
\field{month}{2}
\field{title}{Investigating low and high load cycling tests as accelerated stress tests for proton exchange membrane water electrolysis}
\field{volume}{370}
\field{year}{2021}
\field{pages}{137748}
\range{pages}{1}
\keyw{Accelerated stress test; PEM water electrolyser; Load cycling; Hydrogen;Important;Unread;AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Sun2014-iz}{article}{}
\name{author}{5}{}{%
{{hash=fc08d9965771a09cbe409e014a313234}{%
family={Sun},
familyi={S\bibinitperiod},
given={Shucheng},
giveni={S\bibinitperiod}}}%
{{hash=ac8135a5bcc66f4cc89df8b9ef34d94c}{%
family={Shao},
familyi={S\bibinitperiod},
given={Zhigang},
giveni={Z\bibinitperiod}}}%
{{hash=bd51e6fd7266e3e05ef27f544d18fcba}{%
family={Yu},
familyi={Y\bibinitperiod},
given={Hongmei},
giveni={H\bibinitperiod}}}%
{{hash=62818deee0b5b7655d121ce59d589feb}{%
family={Li},
familyi={L\bibinitperiod},
given={Guangfu},
giveni={G\bibinitperiod}}}%
{{hash=19d8340aa4a8208364573fc1dd71d334}{%
family={Yi},
familyi={Y\bibinitperiod},
given={Baolian},
giveni={B\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Elsevier BV}%
}
\strng{namehash}{db67997f390ef33c0b1c7f67e63674e3}
\strng{fullhash}{f62dc19df3c8a0f526f34631d5cc1bc1}
\strng{bibnamehash}{f62dc19df3c8a0f526f34631d5cc1bc1}
\strng{authorbibnamehash}{f62dc19df3c8a0f526f34631d5cc1bc1}
\strng{authornamehash}{db67997f390ef33c0b1c7f67e63674e3}
\strng{authorfullhash}{f62dc19df3c8a0f526f34631d5cc1bc1}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{A 9-cell proton exchange membrane (PEM) water electrolysis stack is developed and tested for 7800 h. The average degradation rate of 35.5 $\mu$V h−1 per cell is measured. The 4th MEA of the stack is offline investigated and characterized. The electrochemical impedance spectroscopy (EIS) shows that the charge transfer resistance and ionic resistance of the cell both increase. The linear sweep scan (LSV) shows the hydrogen crossover rate of the membrane has slight increase. The electron probe X-ray microanalyze (EPMA) illustrates further that Ca, Cu and Fe elements distribute in the membrane and catalyst layers of the catalyst-coated membranes (CCMs). The cations occupy the ion exchange sites of the Nafion polymer electrolyte in the catalyst layers and membrane, which results in the increase in the anode and the cathode overpotentials. The metallic impurities originate mainly from the feed water and the components of the electrolysis unit. Fortunately, the degradation was reversible and can be almost recovered to the initial performance by using 0.5 M H2SO4. This indicates the performance degradation of the stack running 7800 h is mainly caused by a recoverable contamination.}
\field{journaltitle}{J. Power Sources}
\field{month}{12}
\field{title}{Investigations on degradation of the long-term proton exchange membrane water electrolysis stack}
\field{volume}{267}
\field{year}{2014}
\field{pages}{515\bibrangedash 520}
\range{pages}{6}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Frensch2019-ff}{article}{}
\name{author}{6}{}{%
{{hash=a3b2ef301ac23f14f32c9a4fc00de427}{%
family={Frensch},
familyi={F\bibinitperiod},
given={Steffen\bibnamedelima Henrik},
giveni={S\bibinitperiod\bibinitdelim H\bibinitperiod}}}%
{{hash=97090d0b88f70badfc757083e38d9ea2}{%
family={Fouda-Onana},
familyi={F\bibinithyphendelim O\bibinitperiod},
given={Frédéric},
giveni={F\bibinitperiod}}}%
{{hash=d2271a645add41fe9eae68dd94d43bbe}{%
family={Serre},
familyi={S\bibinitperiod},
given={Guillaume},
giveni={G\bibinitperiod}}}%
{{hash=ba62beea3bd3e6efd31a9e0620815d1f}{%
family={Thoby},
familyi={T\bibinitperiod},
given={Dominique},
giveni={D\bibinitperiod}}}%
{{hash=d7bdd613e3085de0f89098327c7efd1d}{%
family={Araya},
familyi={A\bibinitperiod},
given={Samuel\bibnamedelima Simon},
giveni={S\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=5004d188295853e5aa42c4aef2b284c3}{%
family={K{æ}r},
familyi={K\bibinitperiod},
given={S{ø}ren\bibnamedelima Knudsen},
giveni={S\bibinitperiod\bibinitdelim K\bibinitperiod}}}%
}
\strng{namehash}{e78c1d238b95a712a844f579cf2f4ce0}
\strng{fullhash}{bf32761b9d67db93fd151bbed632d4a8}
\strng{bibnamehash}{bf32761b9d67db93fd151bbed632d4a8}
\strng{authorbibnamehash}{bf32761b9d67db93fd151bbed632d4a8}
\strng{authornamehash}{e78c1d238b95a712a844f579cf2f4ce0}
\strng{authorfullhash}{bf32761b9d67db93fd151bbed632d4a8}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{This paper compares the performance over time of seven different operation modes with the scope of investigating realistic degradation responses of a polymer electrolyte membrane water electrolysis (PEM WE) to different operation modes for grid-balancing services. Among these modes are constant current and constant voltage operations at different temperatures and current cycling operations, including a solar profile. It was found that faster current cycling improved the overall cell performance over the test period of 500 h, mainly due to a decrease in total ohmic resistance, which is also observed during the break-in phase preceding the experiments. Dynamic operation led to more severe fluoride emission from the catalyst binder, which improved the performance in terms of cell potential but could be a concern for long term degradation as membrane thinning promotes higher gas crossover. All other operation modes on the other hand suffered from an increase in total ohmic resistance, leading to an overall performance decrease, which is suspected to be due to the passivation of the Ti components. Higher operating temperatures were found to enhance cell performance, but are detrimental from a durability point of view since they exacerbate both membrane thinning and passivation processes.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{11}
\field{number}{57}
\field{title}{Influence of the operation mode on {PEM} water electrolysis degradation}
\field{volume}{44}
\field{year}{2019}
\field{pages}{29889\bibrangedash 29898}
\range{pages}{10}
\keyw{Polymer electrolyte membrane; Water electrolysis; Membrane degradation; Dynamic operation mode; Durability;AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Caparros_Mancera2020-rv}{article}{}
\name{author}{5}{}{%
{{hash=0787b479774efe16385e81f02400c623}{%
family={Caparrós\bibnamedelima Mancera},
familyi={C\bibinitperiod\bibinitdelim M\bibinitperiod},
given={Julio\bibnamedelima José},
giveni={J\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=f0d244bf9e939f8254b8f7934955270a}{%
family={Segura\bibnamedelima Manzano},
familyi={S\bibinitperiod\bibinitdelim M\bibinitperiod},
given={Francisca},
giveni={F\bibinitperiod}}}%
{{hash=2ce517136cdcafc60e6cf9e241605765}{%
family={Andújar},
familyi={A\bibinitperiod},
given={José\bibnamedelima Manuel},
giveni={J\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=aee62c9feade3611a199a62e9e27c315}{%
family={Vivas},
familyi={V\bibinitperiod},
given={Francisco\bibnamedelima José},
giveni={F\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=29bad09956003e146f47998284273673}{%
family={Calderón},
familyi={C\bibinitperiod},
given={Antonio\bibnamedelima José},
giveni={A\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Multidisciplinary Digital Publishing Institute}%
}
\strng{namehash}{4e127016a23851230c042060b97a295c}
\strng{fullhash}{8c31bc472336ed02a9c232dc23968d71}
\strng{bibnamehash}{8c31bc472336ed02a9c232dc23968d71}
\strng{authorbibnamehash}{8c31bc472336ed02a9c232dc23968d71}
\strng{authornamehash}{4e127016a23851230c042060b97a295c}
\strng{authorfullhash}{8c31bc472336ed02a9c232dc23968d71}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{The progressive increase in hydrogen technologies' role in transport, mobility, electrical microgrids, and even in residential applications, as well as in other sectors is expected. However, to achieve it, it is necessary to focus efforts on improving features of hydrogen-based systems, such as efficiency, start-up time, lifespan, and operating power range, among others. A key sector in the development of hydrogen technology is its production, renewable if possible, with the objective to obtain increasingly efficient, lightweight, and durable electrolyzers. For this, scientific works are currently being produced on stacks technology improvement (mainly based on two technologies: polymer electrolyte membrane (PEM) and alkaline) and on the balance of plant (BoP) or the industrial plant (its size depends on the power of the electrolyzer) that runs the stack for its best performance. PEM technology offers distinct advantages, apart from the high cost of its components, its durability that is not yet guaranteed and the availability in the MW range. Therefore, there is an open field of research for achievements in this technology. The two elements to improve are the stacks and BoP, also bearing in mind that improving BoP will positively affect the stack operation. This paper develops the design, implementation, and practical experimentation of a BoP for a medium-size PEM electrolyzer. It is based on the realization of the optimal design of the BoP, paying special attention to the subsystems that comprise it: the power supply subsystem, water management subsystem, hydrogen production subsystem, cooling subsystem, and control subsystem. Based on this, a control logic has been developed that guarantees efficient and safe operation. Experimental results validate the designed control logic in various operating cases, including warning and failure cases. Additionally, the experimental results show the correct operation in the different states of the plant, analyzing the evolution of the hydrogen flow pressure and temperature. The capacity of the developed PEM electrolysis plant is probed regarding its production rate, wide operating power range, reduced pressurization time, and high efficiency.}
\field{journaltitle}{Electronics}
\field{month}{5}
\field{number}{5}
\field{title}{An Optimized Balance of Plant for a {Medium-Size} {PEM} Electrolyzer: Design, Control and Physical Implementation}
\field{volume}{9}
\field{year}{2020}
\field{pages}{871}
\range{pages}{1}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Roh2019-vj}{article}{}
\name{author}{5}{}{%
{{hash=414da7cc64781854b5522b3357eade3e}{%
family={Roh},
familyi={R\bibinitperiod},
given={Kosan},
giveni={K\bibinitperiod}}}%
{{hash=a167b1a74ff0f2fcd622b94b7438539b}{%
family={Brée},
familyi={B\bibinitperiod},
given={Luisa\bibnamedelima C},
giveni={L\bibinitperiod\bibinitdelim C\bibinitperiod}}}%
{{hash=262734fe9754a8419a8fa52868fb99c4}{%
family={Perrey},
familyi={P\bibinitperiod},
given={Karen},
giveni={K\bibinitperiod}}}%
{{hash=6bb07166c1461d81e5fb81f396e7b58b}{%
family={Bulan},
familyi={B\bibinitperiod},
given={Andreas},
giveni={A\bibinitperiod}}}%
{{hash=e56363c80535b637a5107d1a3708edf7}{%
family={Mitsos},
familyi={M\bibinitperiod},
given={Alexander},
giveni={A\bibinitperiod}}}%
}
\strng{namehash}{e98d4a064879a38f424dead72fc670d2}
\strng{fullhash}{1f5f5207961f4541eb81958482d48b53}
\strng{bibnamehash}{1f5f5207961f4541eb81958482d48b53}
\strng{authorbibnamehash}{1f5f5207961f4541eb81958482d48b53}
\strng{authornamehash}{e98d4a064879a38f424dead72fc670d2}
\strng{authorfullhash}{1f5f5207961f4541eb81958482d48b53}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Flexible operation of chlor-alkali electrolysis is a novel strategy for minimizing electricity costs in electricity markets with volatile prices. We recently proposed an operation strategy to adjust the process to electricity price profiles by varying the load and by switching the operation between two different modes (Brée et al., 2019). This flexible operation leads to savings in operating costs but to increased capital investment due to retrofitting for overcapacity and/or switchable operation. Here, we solve an optimization problem which determines optimal operation and (if necessary) oversizing of the switchable chlor-alkali process. This optimization allows analyzing the electricity utilization as well as economic potential of the flexible operation. To arrive at the optimum in reasonable computation time, we use a piecewise linear model for the power demand calculation. Moreover, we use a decomposition algorithm based on the golden-section search of the complicating variable representing oversizing. We show that the flexible operation is capable of reducing electricity utilization as well as costs. Higher savings in the operating costs can be gained with a stronger electricity price fluctuation, a larger operational margin, a bigger chlorine storage tank, and lower capital investment for retrofitting. Lastly, electrolyzer lifetime reduction due to the flexible operation somewhat increases investment costs and thus slightly diminishes cost savings.}
\field{journaltitle}{Appl. Energy}
\field{month}{12}
\field{title}{Flexible operation of switchable chlor-alkali electrolysis for demand side management}
\field{volume}{255}
\field{year}{2019}
\field{pages}{113880}
\range{pages}{1}
\keyw{Demand side management; Chlor-alkali electrolysis; Bifunctional cathode; Decomposition; MILP;Dynamic PEM Paper}
\endentry
\entry{Kahn2021-jv}{report}{}
\name{author}{4}{}{%
{{hash=8dfd63d0ab1647187b4e735251e24011}{%
family={Kahn},
familyi={K\bibinitperiod},
given={Mohd\bibnamedelima Adnan},
giveni={M\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=244b51526d25f948174318be605f9614}{%
family={Young},
familyi={Y\bibinitperiod},
given={Cameron},
giveni={C\bibinitperiod}}}%
{{hash=0436fd625dba88e56a93d60031eff8ae}{%
family={Mackinnon},
familyi={M\bibinitperiod},
given={Catherine},
giveni={C\bibinitperiod}}}%
{{hash=a1f6d7360cc2ff7fcea409d8c1825cf9}{%
family={Layzell},
familyi={L\bibinitperiod},
given={David\bibnamedelima B},
giveni={D\bibinitperiod\bibinitdelim B\bibinitperiod}}}%
}
\list{institution}{1}{%
{The Transition Accelerator}%
}
\strng{namehash}{187d2061aa85ff99e4349af6e390efa2}
\strng{fullhash}{d30f62bc8f08bf5838e6582648175fa1}
\strng{bibnamehash}{d30f62bc8f08bf5838e6582648175fa1}
\strng{authorbibnamehash}{d30f62bc8f08bf5838e6582648175fa1}
\strng{authornamehash}{187d2061aa85ff99e4349af6e390efa2}
\strng{authorfullhash}{d30f62bc8f08bf5838e6582648175fa1}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{10}
\field{title}{The {Techno-Economics} of Hydrogen Compression}
\field{type}{techreport}
\field{year}{2021}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Narayanan2022-uj}{article}{}
\name{author}{5}{}{%
{{hash=a6470458f96683d3b39f830b2c1e8ab1}{%
family={Narayanan},
familyi={N\bibinitperiod},
given={Thaneer\bibnamedelima Malai},
giveni={T\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=30f0965ad95a72e4711c7228c813944e}{%
family={He},
familyi={H\bibinitperiod},
given={Guannan},
giveni={G\bibinitperiod}}}%
{{hash=93b3f84be0d52ed87f57ea48d3e58aac}{%
family={Gençer},
familyi={G\bibinitperiod},
given={Emre},
giveni={E\bibinitperiod}}}%
{{hash=6fae6915ee4792c138348147a18329d1}{%
family={Shao-Horn},
familyi={S\bibinithyphendelim H\bibinitperiod},
given={Yang},
giveni={Y\bibinitperiod}}}%
{{hash=91cbc3b09461b985e4f5269a468e0831}{%
family={Mallapragada},
familyi={M\bibinitperiod},
given={Dharik\bibnamedelima S},
giveni={D\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
}
\list{publisher}{1}{%
{American Chemical Society}%
}
\strng{namehash}{6b6c5f5a3d6151cad8ca4ed2119701ee}
\strng{fullhash}{d525d223c69ac20e2ec9a0e74ff10963}
\strng{bibnamehash}{d525d223c69ac20e2ec9a0e74ff10963}
\strng{authorbibnamehash}{d525d223c69ac20e2ec9a0e74ff10963}
\strng{authornamehash}{6b6c5f5a3d6151cad8ca4ed2119701ee}
\strng{authorfullhash}{d525d223c69ac20e2ec9a0e74ff10963}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Hydrogen can be valuable for deep decarbonization of electricity systems as well as energy end-uses where direct electricity is challenged. While most hydrogen supply chain analyses focus on storing hydrogen as compressed gas hydrogen (CGH2) storage, hydrogen storage systems with lower capital cost of storage capacity such as liquefied hydrogen (LH2) and liquid organic hydrogen carriers (LOHC) may provide a differentiated value for energy system decarbonization. However, the latter systems have disadvantages (e.g., higher capital cost of charge/discharge capacity, boil-off, high energy requirement for conversion/reconversion) that could impede their deployment. In this study, we expand upon a previously developed electricity-hydrogen infrastructure planning model, DOLPHyN, to explore the value of liquid hydrogen solutions for energy storage and transport in a deeply decarbonized energy system. First, we show that TOL-LOHC (i.e., toluene-based LOHC) and LH2 are beneficial as seasonal energy storage systems, while CGH2 is more suitable for shorter-duration storage (e.g., weekly) cycling. The addition of LH2 and LOHC enables more efficient utilization of deployed electric generation and power-to-hydrogen generation infrastructures. Second, we show that LH2 is more favorable than LOHC for providing long-term storage for the power sector because its primary capital costs and energy penalty for conversion are incurred during charging periods, which typically correspond to when renewable energy is more abundant. Cost effectiveness of LH2 is based on accounting for boil-off rates of large-scale LH2 systems. Third, we show that commercial viability of LOHC is strongly tied to the ability to lower the energetic consumption of the dehydrogenation process as well as its capital cost.}
\field{journaltitle}{ACS Sustainable Chem. Eng.}
\field{month}{8}
\field{number}{33}
\field{title}{Role of Liquid Hydrogen Carriers in Deeply Decarbonized Energy Systems}
\field{volume}{10}
\field{year}{2022}
\field{pages}{10768\bibrangedash 10780}
\range{pages}{13}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Kotzur2018-rx}{article}{}
\name{author}{4}{}{%
{{hash=a9cd4b048c05308c48f40dfb8c07dfb9}{%
family={Kotzur},
familyi={K\bibinitperiod},
given={Leander},
giveni={L\bibinitperiod}}}%
{{hash=6b273f6d552b49b637ac33a4a8703123}{%
family={Markewitz},
familyi={M\bibinitperiod},
given={Peter},
giveni={P\bibinitperiod}}}%
{{hash=98d585d0ab76779848762e34a74e1de5}{%
family={Robinius},
familyi={R\bibinitperiod},
given={Martin},
giveni={M\bibinitperiod}}}%
{{hash=fcb017aa00b2d8a14aedfe47f774dda4}{%
family={Stolten},
familyi={S\bibinitperiod},
given={Detlef},
giveni={D\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Elsevier BV}%
}
\strng{namehash}{6e1ae8238aebbe918417dbf0ad5e2337}
\strng{fullhash}{c4736217476739fc9b3ee5f681cb9a00}
\strng{bibnamehash}{c4736217476739fc9b3ee5f681cb9a00}
\strng{authorbibnamehash}{c4736217476739fc9b3ee5f681cb9a00}
\strng{authornamehash}{6e1ae8238aebbe918417dbf0ad5e2337}
\strng{authorfullhash}{c4736217476739fc9b3ee5f681cb9a00}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{The optimization-based design of renewable energy systems is a computationally demanding task because of the high temporal fluctuation of supply and demand time series. In order to reduce these time series, the aggregation of typical operation periods has become common. The problem with this method is that these aggregated typical periods are modeled independently and cannot exchange energy. Therefore, seasonal storage cannot be adequately taken into account, although this will be necessary for energy systems with a high share of renewable generation.To address this issue, this paper proposes a novel mathematical description for storage inventories based on the superposition of inter-period and intra-period states. Inter-period states connect the typical periods and are able to account their sequence. The approach has been adopted for different energy system configurations. The results show that a significant reduction in the computational load can be achieved also for long term storage-based energy system models in comparison to optimization models based on the full annual time series.}
\field{journaltitle}{Appl. Energy}
\field{month}{3}
\field{title}{Time series aggregation for energy system design: Modeling seasonal storage}
\field{volume}{213}
\field{year}{2018}
\field{pages}{123\bibrangedash 135}
\range{pages}{13}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Sandhya-Rani2019-pw}{article}{}
\name{author}{3}{}{%
{{hash=2b84bbdb5d454ac5f36297d1a351321c}{%
family={Sandhya\bibnamedelima Rani},
familyi={S\bibinitperiod\bibinitdelim R\bibinitperiod},
given={G},
giveni={G\bibinitperiod}}}%
{{hash=acc724040dbe1f5343ab3cd2b4f2f087}{%
family={Jayan},
familyi={J\bibinitperiod},
given={Sarada},
giveni={S\bibinitperiod}}}%
{{hash=86142db808b13a469e5ad99207efa3d4}{%
family={Nagaraja},
familyi={N\bibinitperiod},
given={K\bibnamedelima V},
giveni={K\bibinitperiod\bibinitdelim V\bibinitperiod}}}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{dac021b0a0928ccb62d0dedda79edd03}
\strng{fullhash}{dac021b0a0928ccb62d0dedda79edd03}
\strng{bibnamehash}{dac021b0a0928ccb62d0dedda79edd03}
\strng{authorbibnamehash}{dac021b0a0928ccb62d0dedda79edd03}
\strng{authornamehash}{dac021b0a0928ccb62d0dedda79edd03}
\strng{authorfullhash}{dac021b0a0928ccb62d0dedda79edd03}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Abstract Golden section search method is one of the fastest direct search algorithms to solve single variable optimization problems, in which the search space is reduced from [a, b] to [0,1]. This paper describes an extended golden section search method in order to find the minimum of an n-variable function by transforming its n-dimensional cubic search space to the zero-one n-dimensional cube. The paper also provides a MATLAB code for two-dimensional and three-dimensional golden section search algorithms for a zero-one n-dimensional cube. Numerical results for some benchmark functions up to five dimensions and a comparison of the proposed algorithm with the Neldor Mead Simplex Algorithm is also provided.}
\field{journaltitle}{IOP Conf. Ser. Mater. Sci. Eng.}
\field{month}{11}
\field{number}{1}
\field{title}{An extension of golden section algorithm for n-variable functions with {MATLAB} code}
\field{volume}{577}
\field{year}{2019}
\field{pages}{012175}
\range{pages}{1}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Nicholson2018-mw}{article}{}
\name{author}{5}{}{%
{{hash=0942ef0e93c6ca0619e446c97db5f1c5}{%
family={Nicholson},
familyi={N\bibinitperiod},
given={Bethany},
giveni={B\bibinitperiod}}}%
{{hash=56a744a01a7a808030086b81758a6a4c}{%
family={Siirola},
familyi={S\bibinitperiod},
given={John\bibnamedelima D},
giveni={J\bibinitperiod\bibinitdelim D\bibinitperiod}}}%
{{hash=674ebab2a61476821ad4b4bcd4d9c524}{%
family={Watson},
familyi={W\bibinitperiod},
given={Jean-Paul},
giveni={J\bibinithyphendelim P\bibinitperiod}}}%
{{hash=97069c5aa2ca7e73b5f887b09412a1b0}{%
family={Zavala},
familyi={Z\bibinitperiod},
given={Victor\bibnamedelima M},
giveni={V\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=b4cf951e494c45f4de462df04ec3b5df}{%
family={Biegler},
familyi={B\bibinitperiod},
given={Lorenz\bibnamedelima T},
giveni={L\bibinitperiod\bibinitdelim T\bibinitperiod}}}%
}
\strng{namehash}{60ccb2d67386e8673b21f2ff88a3a898}
\strng{fullhash}{ca66a9300d950f38cba4180a8119dfac}
\strng{bibnamehash}{ca66a9300d950f38cba4180a8119dfac}
\strng{authorbibnamehash}{ca66a9300d950f38cba4180a8119dfac}
\strng{authornamehash}{60ccb2d67386e8673b21f2ff88a3a898}
\strng{authorfullhash}{ca66a9300d950f38cba4180a8119dfac}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Math. Program. Comput.}
\field{title}{pyomo.dae: a modeling and automatic discretization framework for optimization with differential and algebraic equations}
\field{year}{2018}
\keyw{Thesis Proposal References;MIT Energy Fellows Application;AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Wachter2006-la}{article}{}
\name{author}{2}{}{%
{{hash=dc724f071bd0fcc1e9eabf5f96209481}{%
family={Wächter},
familyi={W\bibinitperiod},
given={Andreas},
giveni={A\bibinitperiod}}}%
{{hash=b4cf951e494c45f4de462df04ec3b5df}{%
family={Biegler},
familyi={B\bibinitperiod},
given={Lorenz\bibnamedelima T},
giveni={L\bibinitperiod\bibinitdelim T\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{2}{%
{Springer Science}%
{Business Media LLC}%
}
\strng{namehash}{842d46b1a24346d113a0e8274fa09dd0}
\strng{fullhash}{842d46b1a24346d113a0e8274fa09dd0}
\strng{bibnamehash}{842d46b1a24346d113a0e8274fa09dd0}
\strng{authorbibnamehash}{842d46b1a24346d113a0e8274fa09dd0}
\strng{authornamehash}{842d46b1a24346d113a0e8274fa09dd0}
\strng{authorfullhash}{842d46b1a24346d113a0e8274fa09dd0}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{We present a primal-dual interior-point algorithm with a filter line-search method for nonlinear programming. Local and global convergence properties of this method were analyzed in previous work. Here we provide a comprehensive description of the algorithm, including the feasibility restoration phase for the filter method, second-order corrections, and inertia correction of the KKT matrix. Heuristics are also considered that allow faster performance. This method has been implemented in the IPOPT code, which we demonstrate in a detailed numerical study based on 954 problems from the CUTEr test set. An evaluation is made of several line-search options, and a comparison is provided with two state-of-the-art interior-point codes for nonlinear programming.}
\field{journaltitle}{Math. Program.}
\field{month}{3}
\field{number}{1}
\field{title}{On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming}
\field{volume}{106}
\field{year}{2006}
\field{pages}{25\bibrangedash 57}
\range{pages}{33}
\keyw{Dynamic PEM Paper}
\endentry
\entry{UnknownUnknown-hg}{misc}{}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labeltitlesource}{title}
\field{title}{{HSL}, a collection of Fortran codes for large-scale scientific computation. See https://www.hsl.rl.ac.uk}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Reuther2018-xn}{article}{}
\name{author}{18}{}{%
{{hash=9a096b7ae5126f7a35ef175fdf5c6564}{%
family={Reuther},
familyi={R\bibinitperiod},
given={A},
giveni={A\bibinitperiod}}}%
{{hash=1fbd8361442b9fa84916709f76bfa614}{%
family={Kepner},
familyi={K\bibinitperiod},
given={J},
giveni={J\bibinitperiod}}}%
{{hash=48e50b2ed7cc7811b29e5dbfc0ff3d9f}{%
family={Byun},
familyi={B\bibinitperiod},
given={C},
giveni={C\bibinitperiod}}}%
{{hash=99e831b39244c28fbad13d6a82a1858c}{%
family={Samsi},
familyi={S\bibinitperiod},
given={S},
giveni={S\bibinitperiod}}}%
{{hash=ac869e1b05c9201b4382e50e1ee2f835}{%
family={Arcand},
familyi={A\bibinitperiod},
given={W},
giveni={W\bibinitperiod}}}%
{{hash=2a83b34284ab89018a0c7145dbb11ba1}{%
family={Bestor},
familyi={B\bibinitperiod},
given={David},
giveni={D\bibinitperiod}}}%
{{hash=4a5174dc4328ceb6e7c10bf93b1687a5}{%
family={Bergeron},
familyi={B\bibinitperiod},
given={Bill},
giveni={B\bibinitperiod}}}%
{{hash=1242a6263ab7fd94390d1d89faed0f37}{%
family={Gadepally},
familyi={G\bibinitperiod},
given={V},
giveni={V\bibinitperiod}}}%
{{hash=38cb20e2eee5400e3d2fe045bcbd7efa}{%
family={Houle},
familyi={H\bibinitperiod},
given={Michael},
giveni={M\bibinitperiod}}}%
{{hash=28ce505281c2dee48e3a6013fc4f9707}{%
family={Hubbell},
familyi={H\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=5a25bc91f524ca6dfc2ecf9f4a13903c}{%
family={Jones},
familyi={J\bibinitperiod},
given={Michael},
giveni={M\bibinitperiod}}}%
{{hash=8273bd19b76ac8667287839b10d87af0}{%
family={Klein},
familyi={K\bibinitperiod},
given={Anna},
giveni={A\bibinitperiod}}}%
{{hash=9d340d6142e93c65341cfacdd1e3dded}{%
family={Milechin},
familyi={M\bibinitperiod},
given={Lauren},
giveni={L\bibinitperiod}}}%
{{hash=ee6934af00a35f2698ca5f628c268996}{%
family={Mullen},
familyi={M\bibinitperiod},
given={J},
giveni={J\bibinitperiod}}}%
{{hash=0a4efc26dfdafbb32e9190b61b735adf}{%
family={Prout},
familyi={P\bibinitperiod},
given={Andrew},
giveni={A\bibinitperiod}}}%
{{hash=45c1cc077bbaff8696e017b76ca74744}{%
family={Rosa},
familyi={R\bibinitperiod},
given={Antonio},
giveni={A\bibinitperiod}}}%
{{hash=010a43fef13ded9e5680a2fac6f95740}{%
family={Yee},
familyi={Y\bibinitperiod},
given={Charles},
giveni={C\bibinitperiod}}}%
{{hash=29eb02303a9d97f0996689631bb4201b}{%
family={Michaleas},
familyi={M\bibinitperiod},
given={P},
giveni={P\bibinitperiod}}}%
}
\strng{namehash}{15608c1011d0411ae08c329ff9b9eaa4}
\strng{fullhash}{c5eabb9d995f57554db76b990da0461b}
\strng{bibnamehash}{c5eabb9d995f57554db76b990da0461b}
\strng{authorbibnamehash}{c5eabb9d995f57554db76b990da0461b}
\strng{authornamehash}{15608c1011d0411ae08c329ff9b9eaa4}
\strng{authorfullhash}{c5eabb9d995f57554db76b990da0461b}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Interactive massively parallel computations are critical for machine learning and data analysis. These computations are a staple of the MIT Lincoln Laboratory Supercomputing Center (LLSC) and has required the LLSC to develop unique interactive supercomputing capabilities. Scaling interactive machine learning frameworks, such as TensorFlow, and data analysis environments, such as MATLAB/Octave, to tens of thousands of cores presents many technical challenges -- in particular, rapidly dispatching many tasks through a scheduler, such as Slurm, and starting many instances of applications with thousands of dependencies. Careful tuning of launches and prepositioning of applications overcome these challenges and allow the launching of thousands of tasks in seconds on a 40,000-core supercomputer. Specifically, this work demonstrates launching 32,000 TensorFlow processes in 4 seconds and launching 262,000 Octave processes in 40 seconds. These capabilities allow researchers to rapidly explore novel machine learning architecture and data analysis algorithms.}
\field{journaltitle}{HPEC}
\field{month}{7}
\field{title}{Interactive supercomputing on 40,000 cores for machine learning and data analysis}
\field{year}{2018}
\field{pages}{1\bibrangedash 6}
\range{pages}{6}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Lee2020-db}{article}{}
\name{author}{9}{}{%
{{hash=8c4c2d867f443f8c3c23b36457432cf6}{%
family={Lee},
familyi={L\bibinitperiod},
given={Jason\bibnamedelima K},
giveni={J\bibinitperiod\bibinitdelim K\bibinitperiod}}}%
{{hash=a7ab33694446511cbdad2613ccd417f5}{%
family={Lee},
familyi={L\bibinitperiod},
given={Chunghyuk},
giveni={C\bibinitperiod}}}%
{{hash=4d48d49d7d6bddea93815ebf58708dcc}{%
family={Fahy},
familyi={F\bibinitperiod},
given={Kieran\bibnamedelima F},
giveni={K\bibinitperiod\bibinitdelim F\bibinitperiod}}}%
{{hash=c5bb9d2e9ebaf2f377830456502b29a8}{%
family={Zhao},
familyi={Z\bibinitperiod},
given={Benzhong},
giveni={B\bibinitperiod}}}%
{{hash=7997b15bf47d723b2f6b08ece0e32451}{%
family={LaManna},
familyi={L\bibinitperiod},
given={Jacob\bibnamedelima M},
giveni={J\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
{{hash=a13a38d54593e4198d19e45e4fa20eae}{%
family={Baltic},
familyi={B\bibinitperiod},
given={Elias},
giveni={E\bibinitperiod}}}%
{{hash=d122fc60a4a05ee4f710e6ccc169bce2}{%
family={Jacobson},
familyi={J\bibinitperiod},
given={David\bibnamedelima L},
giveni={D\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
{{hash=4a05832ba4e69e4e03bac02ffe099d5c}{%
family={Hussey},
familyi={H\bibinitperiod},
given={Daniel\bibnamedelima S},
giveni={D\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=047e66405bb176b70b3dc22a9a61606c}{%
family={Bazylak},
familyi={B\bibinitperiod},
given={Aimy},
giveni={A\bibinitperiod}}}%
}
\strng{namehash}{ca5a94a77167598679140cfc1fa7dc7f}
\strng{fullhash}{21ca84be2c3d89dd1af7ee39c1c26393}
\strng{bibnamehash}{21ca84be2c3d89dd1af7ee39c1c26393}
\strng{authorbibnamehash}{21ca84be2c3d89dd1af7ee39c1c26393}
\strng{authornamehash}{ca5a94a77167598679140cfc1fa7dc7f}
\strng{authorfullhash}{21ca84be2c3d89dd1af7ee39c1c26393}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Summary Reaching high current densities is absolutely imperative for electrochemical energy conversion, from fuel cells to CO2 reduction. Here, we identify the existence of a performance indicator for gas-evolving electrochemical energy conversion devices: the critical current density. The critical current density pinpoints a performance inflection point whereby both the gas saturations and mass transport overpotentials suddenly dominate cell performance and exacerbate failure. We elucidate the mass transport behavior of a polymer electrolyte membrane (PEM) water electrolyzer using in operando neutron imaging at operating current densities as high as 9 A $\cdot$ cm−2. Product gases become heterogeneously distributed in the porous transport layer adjacent to the catalyst layer and promote disastrous local hotspots. Optimizing new materials and cell architectures with this performance indicator may unlock higher than previously reported performances for electrochemical energy conversion.}
\field{journaltitle}{Cell Reports Physical Science}
\field{month}{8}
\field{number}{8}
\field{title}{Critical Current Density as a Performance Indicator for {Gas-Evolving} Electrochemical Devices}
\field{volume}{1}
\field{year}{2020}
\field{pages}{100147}
\range{pages}{1}
\keyw{polymer electrolyte membrane water electrolyzer; critical current density; neutron imaging; gas saturation; reactant flow regime; porous transport layer;Good Data;Thesis Proposal;Dynamic PEM Paper}
\endentry
\entry{Marangio2011-jc}{article}{}
\name{author}{4}{}{%
{{hash=d3417d4ddbcf8af80c20f7286a118a40}{%
family={Marangio},
familyi={M\bibinitperiod},
given={F},
giveni={F\bibinitperiod}}}%
{{hash=4179d6edffd0f515dd0e3222556506b4}{%
family={Pagani},
familyi={P\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=19fd88f0186972403ba5bc1400b8ec74}{%
family={Santarelli},
familyi={S\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=64083406350723dcb509e001fb893535}{%
family={Calı̀},
familyi={C\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
}
\strng{namehash}{fb4186a8836d5595dc1c0411438c44fd}
\strng{fullhash}{bf9f60e7b429f911913a37ec7f2b19fc}
\strng{bibnamehash}{bf9f60e7b429f911913a37ec7f2b19fc}
\strng{authorbibnamehash}{bf9f60e7b429f911913a37ec7f2b19fc}
\strng{authornamehash}{fb4186a8836d5595dc1c0411438c44fd}
\strng{authorfullhash}{bf9f60e7b429f911913a37ec7f2b19fc}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{This paper describes the research activity regarding High Pressure PEM Water Electrolysis conducted at the Energetics Department at the Politecnico di Torino (Turin, Italy). In particular, the first phase of the design and assembly of an in-house made prototype is discussed. This activity was started after a detailed theoretical and experimental work carried on a prototype manufactured by Giner Electrochemical Systems. The experience learned suggested that the new activity should aim at a pressure of about 30--45 bar and higher temperatures, because this pressure range is the one which minimises the electrolysis + compression overall power request, while higher temperature allows to work at lower voltages, thus increasing the electrolyser and the overall process efficiency. On this basis, the design of the electrolyser has been performed, and the elements of the electrolyser have been tested as described in the paper. The tests have been performed at different pressure values, and at different operating temperatures: 40 °C and 55 °C; a wide range of single cell voltage has been observed: at a current density of 1 A/cm2, it ranges from 2.1 V (T = 55 °C, p = 10 bar) to 2.4 V (T = 40 °C, p = 70 bar).}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{7}
\field{number}{13}
\field{title}{Concept of a high pressure {PEM} electrolyser prototype}
\field{volume}{36}
\field{year}{2011}
\field{pages}{7807\bibrangedash 7815}
\range{pages}{9}
\keyw{Hydrogen; High pressure; PEM; Prototype;Thesis Proposal;Good Data;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Martin2022-ek}{article}{}
\name{author}{4}{}{%
{{hash=76dad5a7c8b72964c7cb9de160bbeea4}{%
family={Martin},
familyi={M\bibinitperiod},
given={Agate},
giveni={A\bibinitperiod}}}%
{{hash=4a1ce7e41a33302515047ec4fd94f241}{%
family={Trinke},
familyi={T\bibinitperiod},
given={Patrick},
giveni={P\bibinitperiod}}}%
{{hash=1fadf57f9ef9e20601eb2a2a5b9f6d83}{%
family={Bensmann},
familyi={B\bibinitperiod},
given={Boris},
giveni={B\bibinitperiod}}}%
{{hash=88f731b9d9db30338a9ff363ad03c699}{%
family={Hanke-Rauschenbach},
familyi={H\bibinithyphendelim R\bibinitperiod},
given={Richard},
giveni={R\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{663cc9bd81b06e7fbf81869d903b0f69}
\strng{fullhash}{a8d8959a4c69df6c312247fb958e987a}
\strng{bibnamehash}{a8d8959a4c69df6c312247fb958e987a}
\strng{authorbibnamehash}{a8d8959a4c69df6c312247fb958e987a}
\strng{authornamehash}{663cc9bd81b06e7fbf81869d903b0f69}
\strng{authorfullhash}{a8d8959a4c69df6c312247fb958e987a}
\field{sortinit}{4}
\field{sortinithash}{9381316451d1b9788675a07e972a12a7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Hydrogen Crossover in PEM Water Electrolysis at Current Densities up to 10 A cm−2 , Agate Martin, Patrick Trinke, Boris Bensmann, Richard Hanke-Rauschenbach}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{9}
\field{number}{9}
\field{title}{Hydrogen Crossover in {PEM} Water Electrolysis at Current Densities up to 10 A cm−2}
\field{volume}{169}
\field{year}{2022}
\field{pages}{094507}
\range{pages}{1}
\keyw{Dynamic PEM Paper}
\endentry
\entry{noauthor_undated-so}{misc}{}
\list{language}{1}{%
{en}%
}
\field{sortinit}{4}
\field{sortinithash}{9381316451d1b9788675a07e972a12a7}
\field{labeltitlesource}{title}
\field{abstract}{Get AS market prices, including day-ahead market clearing prices, BES, commercially significant constraints, shadow prices and RPRS.}
\field{booktitle}{Market Prices}
\field{howpublished}{\url{https://www.ercot.com/mktinfo/prices}}
\field{note}{Accessed: 2023-11-16}
\field{title}{{ERCOT} Market Prices}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Gagnon_undated-so}{misc}{}
\name{author}{3}{}{%
{{hash=3b2a3e140fc173994f9c23761e43de14}{%
family={Gagnon},
familyi={G\bibinitperiod},
given={Pieter},
giveni={P\bibinitperiod}}}%
{{hash=373a29c40a28810fa89a241b7ec1c88a}{%
family={Cowiestoll},
familyi={C\bibinitperiod},
given={Brady},
giveni={B\bibinitperiod}}}%
{{hash=f186d10ac274113e3e2a57ed5f577d8d}{%
family={Schwarz},
familyi={S\bibinitperiod},
given={Marty},
giveni={M\bibinitperiod}}}%
}
\strng{namehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{fullhash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{bibnamehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{authorbibnamehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{authornamehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{authorfullhash}{291be4dda9dbef64eb2177af58d5f1cb}
\field{sortinit}{4}
\field{sortinithash}{9381316451d1b9788675a07e972a12a7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Cambium 2022 Scenario Descriptions and Documentation}
\keyw{Dynamic PEM Paper}
\endentry
\entry{He2021-kv}{article}{}
\name{author}{5}{}{%
{{hash=30f0965ad95a72e4711c7228c813944e}{%
family={He},
familyi={H\bibinitperiod},
given={Guannan},
giveni={G\bibinitperiod}}}%
{{hash=91cbc3b09461b985e4f5269a468e0831}{%
family={Mallapragada},
familyi={M\bibinitperiod},
given={Dharik\bibnamedelima S},
giveni={D\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=aa7fe1fb4a97a69c3f5c63ec41cbfdb5}{%
family={Bose},
familyi={B\bibinitperiod},
given={Abhishek},
giveni={A\bibinitperiod}}}%
{{hash=3a85b6053063aca0b10c7b8a76f77f43}{%
family={Heuberger-Austin},
familyi={H\bibinithyphendelim A\bibinitperiod},
given={Clara\bibnamedelima F},
giveni={C\bibinitperiod\bibinitdelim F\bibinitperiod}}}%
{{hash=93b3f84be0d52ed87f57ea48d3e58aac}{%
family={Gençer},
familyi={G\bibinitperiod},
given={Emre},
giveni={E\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{The Royal Society of Chemistry}%
}
\strng{namehash}{45aa6ef61983dba2b99a41e6139c6ec8}
\strng{fullhash}{9ba3c9e416578be9e4c535ade5092d4e}
\strng{bibnamehash}{9ba3c9e416578be9e4c535ade5092d4e}
\strng{authorbibnamehash}{9ba3c9e416578be9e4c535ade5092d4e}
\strng{authornamehash}{45aa6ef61983dba2b99a41e6139c6ec8}
\strng{authorfullhash}{9ba3c9e416578be9e4c535ade5092d4e}
\field{sortinit}{4}
\field{sortinithash}{9381316451d1b9788675a07e972a12a7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{There is growing interest in using hydrogen (H2) as a long-duration energy storage resource in a future electric grid dominated by variable renewable energy (VRE) generation. Modeling H2 use exclusively for grid-scale energy storage, often referred to as ``power-to-gas-to-power (P2G2P)'', overlooks the cost-sharing and CO2 emission benefits from using the deployed H2 assets to decarbonize other end-use sectors where direct electrification is challenging. Here, we develop a generalized framework for co-optimizing infrastructure investments across the electricity and H2 supply chains, accounting for the spatio-temporal variations in energy demand and supply. We apply this sector-coupling framework to the U.S. Northeast under a range of technology cost and carbon price scenarios and find greater value of power-to-H2 (P2G) vs. P2G2P routes. Specifically, P2G provides grid flexibility to support VRE integration without the round-trip efficiency penalty and additional cost incurred by P2G2P routes. This form of sector coupling leads to: (a) VRE generation increase by 13--56\%, and (b) total system cost (and levelized costs of energy) reduction by 7--16\% under deep decarbonization scenarios. Both effects increase as H2 demand for other end-uses increases, more than doubling for a 97\% decarbonization scenario as H2 demand quadruples. We also find that the grid flexibility enabled by sector coupling makes deployment of carbon capture and storage (CCS) for power generation less cost-effective than its use for low-carbon H2 production. These findings highlight the importance of using an integrated energy system framework with multiple energy vectors in planning cost-effective energy system decarbonization.}
\field{journaltitle}{Energy Environ. Sci.}
\field{month}{9}
\field{number}{9}
\field{title}{Sector coupling via hydrogen to lower the cost of energy system decarbonization}
\field{volume}{14}
\field{year}{2021}
\field{pages}{4635\bibrangedash 4646}
\range{pages}{12}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Minke2021-io}{article}{}
\name{author}{4}{}{%
{{hash=d1b50827cf0f8b4d7e56c47b0c37bba0}{%
family={Minke},
familyi={M\bibinitperiod},
given={Christine},
giveni={C\bibinitperiod}}}%
{{hash=e87a77f733524fd32c69afd9322d89a9}{%
family={Suermann},
familyi={S\bibinitperiod},
given={Michel},
giveni={M\bibinitperiod}}}%
{{hash=1fadf57f9ef9e20601eb2a2a5b9f6d83}{%
family={Bensmann},
familyi={B\bibinitperiod},
given={Boris},
giveni={B\bibinitperiod}}}%
{{hash=88f731b9d9db30338a9ff363ad03c699}{%
family={Hanke-Rauschenbach},
familyi={H\bibinithyphendelim R\bibinitperiod},
given={Richard},
giveni={R\bibinitperiod}}}%
}
\strng{namehash}{3374e023c1b8560bf9314b4c07f95342}
\strng{fullhash}{21e1d15d360e5a3c5d68718c18bdc9e9}
\strng{bibnamehash}{21e1d15d360e5a3c5d68718c18bdc9e9}
\strng{authorbibnamehash}{21e1d15d360e5a3c5d68718c18bdc9e9}
\strng{authornamehash}{3374e023c1b8560bf9314b4c07f95342}
\strng{authorfullhash}{21e1d15d360e5a3c5d68718c18bdc9e9}
\field{sortinit}{5}
\field{sortinithash}{20e9b4b0b173788c5dace24730f47d8c}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Proton exchange membrane water electrolysis (PEMWE) is a key technology for future sustainable energy systems. Proton exchange membrane (PEM) electrolysis cells use iridium, one of the scarcest elements on earth, as catalyst for the oxygen evolution reaction. In the present study, the expected iridium demand and potential bottlenecks in the realization of PEMWE for hydrogen production in the targeted GW a−1 scale are assessed in a model built on three pillars: (i) an in-depth analysis of iridium reserves and mine production, (ii) technical prospects for the optimization of PEM water electrolyzers, and (iii) PEMWE installation rates for a market ramp-up and maturation model covering 50 years. As a main result, two necessary preconditions have been identified to meet the immense future iridium demand: first, the dramatic reduction of iridium catalyst loading in PEM electrolysis cells and second, the development of a recycling infrastructure for iridium catalysts with technical end-of-life recycling rates of at least 90\%.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{7}
\field{number}{46}
\field{title}{Is iridium demand a potential bottleneck in the realization of large-scale {PEM} water electrolysis?}
\field{volume}{46}
\field{year}{2021}
\field{pages}{23581\bibrangedash 23590}
\range{pages}{10}
\keyw{Proton exchange membrane water electrolysis; PEM electrolyzer; Hydrogen economy; Iridium; Catalyst loading; Market model;Dynamic PEM Paper}
\endentry
\entry{Riedmayer2023-nw}{article}{}
\name{author}{5}{}{%
{{hash=c050453347dcc2c434015cd5dd7ec64a}{%
family={Riedmayer},
familyi={R\bibinitperiod},
given={Rebecca},
giveni={R\bibinitperiod}}}%
{{hash=86d4a97c4708e8a40d4845cae0cc55e1}{%
family={Paren},
familyi={P\bibinitperiod},
given={Benjamin\bibnamedelima A},
giveni={B\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=65f8f12d9eea501664f91732feacb34b}{%
family={Schofield},
familyi={S\bibinitperiod},
given={Landon},
giveni={L\bibinitperiod}}}%
{{hash=6fae6915ee4792c138348147a18329d1}{%
family={Shao-Horn},
familyi={S\bibinithyphendelim H\bibinitperiod},
given={Yang},
giveni={Y\bibinitperiod}}}%
{{hash=c7d97ae5abbe1aea4cbb8bd0ea8bce2d}{%
family={Mallapragada},
familyi={M\bibinitperiod},
given={Dharik},
giveni={D\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{American Chemical Society (ACS)}%
}
\strng{namehash}{1c8736a7306d8cd9cd43727744081334}
\strng{fullhash}{624a55b9f942a534865b094790b01d9f}
\strng{bibnamehash}{624a55b9f942a534865b094790b01d9f}
\strng{authorbibnamehash}{624a55b9f942a534865b094790b01d9f}
\strng{authornamehash}{1c8736a7306d8cd9cd43727744081334}
\strng{authorfullhash}{624a55b9f942a534865b094790b01d9f}
\field{sortinit}{5}
\field{sortinithash}{20e9b4b0b173788c5dace24730f47d8c}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Energy Fuels}
\field{month}{6}
\field{number}{12}
\field{title}{Proton exchange membrane electrolysis performance targets for achieving 2050 expansion goals constrained by iridium supply}
\field{volume}{37}
\field{year}{2023}
\field{pages}{8614\bibrangedash 8623}
\range{pages}{10}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Bernt2018-rj}{article}{}
\name{author}{3}{}{%
{{hash=4f47c6343f444c3e9e61b6d1cfaec14b}{%
family={Bernt},
familyi={B\bibinitperiod},
given={Maximilian},
giveni={M\bibinitperiod}}}%
{{hash=985d96e55c19d7e462d2ade4bdf4c78c}{%
family={Siebel},
familyi={S\bibinitperiod},
given={Armin},
giveni={A\bibinitperiod}}}%
{{hash=b43fb849dae1a0b0667cba868dd9999d}{%
family={Gasteiger},
familyi={G\bibinitperiod},
given={Hubert\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{8fc4f021f70b499d451c714fcf5854cb}
\strng{fullhash}{8fc4f021f70b499d451c714fcf5854cb}
\strng{bibnamehash}{8fc4f021f70b499d451c714fcf5854cb}
\strng{authorbibnamehash}{8fc4f021f70b499d451c714fcf5854cb}
\strng{authornamehash}{8fc4f021f70b499d451c714fcf5854cb}
\strng{authorfullhash}{8fc4f021f70b499d451c714fcf5854cb}
\field{sortinit}{5}
\field{sortinithash}{20e9b4b0b173788c5dace24730f47d8c}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{In this study, the influence of catalyst loading on the performance of a proton exchange membrane (PEM) water electrolyzer is investigated (Nafion 212 membrane; IrO2/TiO2 (anode) and Pt/C (cathode)). Due to the fast kinetics of the hydrogen evolution reaction (HER) on platinum (Pt), the Pt loading on the cathode can be reduced from 0.30 mgPt cm−2 to 0.025 mgPt cm−2 without any negative effect on performance. On the anode, the iridium (Ir) loading was varied between 0.20--5.41 mgIr cm−2 and an optimum in performance at operational current densities ($\geq$1 A cm−2) was found for 1--2 mgIr cm−2. At higher Ir loadings, the performance decreases at high current densities due to insufficient water transport through the catalyst layer whereas at Ir loadings <0.5 mgIr cm−2 the catalyst layer becomes inhomogeneous, which leads to a lower electrochemically active area and catalyst utilization, resulting in a significant decrease of performance. To investigate the potential for a large-scale application of PEM water electrolysis, the Ir-specific power density (gIr kW−1) for membrane electrode assemblies (MEAs) with different catalyst loadings is analyzed as a function of voltage efficiency, and the consequences regarding catalyst material requirements are discussed.}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{3}
\field{number}{5}
\field{title}{Analysis of Voltage Losses in {PEM} Water Electrolyzers with Low Platinum Group Metal Loadings}
\field{volume}{165}
\field{year}{2018}
\field{pages}{F305}
\range{pages}{-1}
\keyw{Read;Important;Reference;Dynamic PEM Paper}
\endentry
\enddatalist
\endrefsection
\refsection{2}
\datalist[entry]{none/global//global/global}
\entry{Brian_D_James_Daniel_A_DeSantis_Genevieve_Saur2016-wz}{report}{}
\name{author}{1}{}{%
{{hash=4125179660bd51242924b976b94e70ca}{%
family={{Brian D James, Daniel A DeSantis, Genevieve Saur}},
familyi={B\bibinitperiod}}}%
}
\list{institution}{1}{%
{Strategic Analysis Inc}%
}
\strng{namehash}{4125179660bd51242924b976b94e70ca}
\strng{fullhash}{4125179660bd51242924b976b94e70ca}
\strng{bibnamehash}{4125179660bd51242924b976b94e70ca}
\strng{authorbibnamehash}{4125179660bd51242924b976b94e70ca}
\strng{authornamehash}{4125179660bd51242924b976b94e70ca}
\strng{authorfullhash}{4125179660bd51242924b976b94e70ca}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{9}
\field{title}{Final Report: Hydrogen Production Pathways Cost Analysis (2013 -- 2016)}
\field{type}{techreport}
\field{year}{2016}
\keyw{Thesis Proposal References;AIChE 2023;Dynamic PEM Paper}
\endentry
\entry{Abdin2015-nl}{article}{}
\name{author}{3}{}{%
{{hash=ee44aa8e06a568106fc39c8b43be1540}{%
family={Abdin},
familyi={A\bibinitperiod},
given={Z},
giveni={Z\bibinitperiod}}}%
{{hash=141c0009aa6538429b49f435f49b64e6}{%
family={Webb},
familyi={W\bibinitperiod},
given={C\bibnamedelima J},
giveni={C\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=34f9605e63b03561f2c4430c1cdaea30}{%
family={Gray},
familyi={G\bibinitperiod},
given={E\bibnamedelima Maca},
giveni={E\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
}
\strng{namehash}{33a1edbabeb00a415835c5be1c069e02}
\strng{fullhash}{33a1edbabeb00a415835c5be1c069e02}
\strng{bibnamehash}{33a1edbabeb00a415835c5be1c069e02}
\strng{authorbibnamehash}{33a1edbabeb00a415835c5be1c069e02}
\strng{authornamehash}{33a1edbabeb00a415835c5be1c069e02}
\strng{authorfullhash}{33a1edbabeb00a415835c5be1c069e02}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{An enhanced model of a PEM electrolyser cell is presented, based on linked modular mathematical models of the anode, cathode, membrane and cell voltage. In addition to predicting the overall equilibrium electrolyser cell performance, based on parameters related to the materials of construction and the configuration of its components, the model is a powerful tool for exploring control strategies and understanding the contributions to cell voltage from the various internal components as a guide to researchers aiming for improved performance.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{10}
\field{number}{39}
\field{title}{Modelling and simulation of a proton exchange membrane ({PEM}) electrolyser cell}
\field{volume}{40}
\field{year}{2015}
\field{pages}{13243\bibrangedash 13257}
\range{pages}{15}
\keyw{Electrolyser; Anode; Cathode; PEM; Overpotential; Modelling;Important;Read;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Liso2018-iy}{article}{}
\name{author}{5}{}{%
{{hash=c6d4149f9b3aaaac588add8722b34ca4}{%
family={Liso},
familyi={L\bibinitperiod},
given={Vincenzo},
giveni={V\bibinitperiod}}}%
{{hash=f472ddc9cc41a658ef12f9b19f13a717}{%
family={Savoia},
familyi={S\bibinitperiod},
given={Giorgio},
giveni={G\bibinitperiod}}}%
{{hash=d7bdd613e3085de0f89098327c7efd1d}{%
family={Araya},
familyi={A\bibinitperiod},
given={Samuel\bibnamedelima Simon},
giveni={S\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=c2187921bdf0ce3055e56fb7d5769f9b}{%
family={Cinti},
familyi={C\bibinitperiod},
given={Giovanni},
giveni={G\bibinitperiod}}}%
{{hash=5004d188295853e5aa42c4aef2b284c3}{%
family={K{æ}r},
familyi={K\bibinitperiod},
given={S{ø}ren\bibnamedelima Knudsen},
giveni={S\bibinitperiod\bibinitdelim K\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Multidisciplinary Digital Publishing Institute}%
}
\strng{namehash}{5c214bd7507ad8c6bcfd28ad5fd32d31}
\strng{fullhash}{4659a821e7ae7423f0735f684e6be893}
\strng{bibnamehash}{4659a821e7ae7423f0735f684e6be893}
\strng{authorbibnamehash}{4659a821e7ae7423f0735f684e6be893}
\strng{authornamehash}{5c214bd7507ad8c6bcfd28ad5fd32d31}
\strng{authorfullhash}{4659a821e7ae7423f0735f684e6be893}
\field{sortinit}{4}
\field{sortinithash}{9381316451d1b9788675a07e972a12a7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{In this paper, a simplified model of a Polymer Electrolyte Membrane (PEM) water electrolysis cell is presented and compared with experimental data at 60 °C and 80 °C. The model utilizes the same modelling approach used in previous work where the electrolyzer cell is divided in four subsections: cathode, anode, membrane and voltage. The model of the electrodes includes key electrochemical reactions and gas transport mechanism (i.e., H2, O2 and H2O) whereas the model of the membrane includes physical mechanisms such as water diffusion, electro osmotic drag and hydraulic pressure. Voltage was modelled including main overpotentials (i.e., activation, ohmic, concentration). First and second law efficiencies were defined. Key empirical parameters depending on temperature were identified in the activation and ohmic overpotentials. The electrodes reference exchange current densities and change transfer coefficients were related to activation overpotentials whereas hydrogen ion diffusion to Ohmic overvoltages. These model parameters were empirically fitted so that polarization curve obtained by the model predicted well the voltage at different current found by the experimental results. Finally, from the efficiency calculation, it was shown that at low current densities the electrolyzer cell absorbs heat from the surroundings. The model is not able to describe the transients involved during the cell electrochemical reactions, however these processes are assumed relatively fast. For this reason, the model can be implemented in system dynamic modelling for hydrogen production and storage where components dynamic is generally slower compared to the cell electrochemical reactions dynamics.}
\field{journaltitle}{Energies}
\field{month}{11}
\field{number}{12}
\field{title}{Modelling and Experimental Analysis of a Polymer Electrolyte Membrane Water Electrolysis Cell at Different Operating Temperatures}
\field{volume}{11}
\field{year}{2018}
\field{pages}{3273}
\range{pages}{1}
\keyw{Important;Read;Thesis Proposal;First Principles Model;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Medina2010-lh}{article}{}
\name{author}{2}{}{%
{{hash=d8f5a26ea4c37eb035c621c697e35c42}{%
family={Medina},
familyi={M\bibinitperiod},
given={P},
giveni={P\bibinitperiod}}}%
{{hash=19fd88f0186972403ba5bc1400b8ec74}{%
family={Santarelli},
familyi={S\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
}
\strng{namehash}{7e27bffb63d646802901f7d29be00a48}
\strng{fullhash}{7e27bffb63d646802901f7d29be00a48}
\strng{bibnamehash}{7e27bffb63d646802901f7d29be00a48}
\strng{authorbibnamehash}{7e27bffb63d646802901f7d29be00a48}
\strng{authornamehash}{7e27bffb63d646802901f7d29be00a48}
\strng{authorfullhash}{7e27bffb63d646802901f7d29be00a48}
\field{sortinit}{5}
\field{sortinithash}{20e9b4b0b173788c5dace24730f47d8c}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{This paper analyzes, through experimental data and a transport model, the water transported through the membrane under different operating conditions in a on a Proton Exchange Membrane (PEM) electrolyzer operating with a high-pressure gradient across the membrane from the cathode (high-pressure) side to the anode (nearly ambient-pressure) side. The phenomena involved in this movement are described and analyzed, with a focus on the electro-osmotic drag coefficient, neo. We have observed that the behavior of the hydraulic percolation determines the results obtained for the electro-osmotic drag, while the contribution of the water diffusion is negligible. In general, the cathode pressure significantly reduces the water transport (a positive effect). Also, operation at lower current density reduces the net electro-osmotic drag coefficient, ng; therefore, the best operation strategy for obtaining dried hydrogen at the cathode is to impose high cathode pressure and low current density.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{6}
\field{number}{11}
\field{title}{Analysis of water transport in a high pressure {PEM} electrolyzer}
\field{volume}{35}
\field{year}{2010}
\field{pages}{5173\bibrangedash 5186}
\range{pages}{14}
\keyw{Electrolysis; High pressure; Water drag;Important;Thesis Proposal;Read;Dynamic PEM Paper}
\endentry
\entry{BurgessUnknown-hi}{incollection}{}
\name{author}{1}{}{%
{{hash=9f01a910becbffdd7291574efc031624}{%
family={Burgess},
familyi={B\bibinitperiod},
given={Donald\bibnamedelima R},
giveni={D\bibinitperiod\bibinitdelim R\bibinitperiod}}}%
}
\name{editor}{2}{}{%
{{hash=888d426f66ad626403eed0a8d35b5f32}{%
family={Linstrom},
familyi={L\bibinitperiod},
given={P\bibnamedelima J},
giveni={P\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=d389729f110111c312169e9a73f539fb}{%
family={Mallard},
familyi={M\bibinitperiod},
given={W\bibnamedelima G},
giveni={W\bibinitperiod\bibinitdelim G\bibinitperiod}}}%
}
\list{location}{1}{%
{Gaithersburg MD, 20899}%
}
\list{publisher}{2}{%
{National Institute of Standards}%
{Technology}%
}
\strng{namehash}{9f01a910becbffdd7291574efc031624}
\strng{fullhash}{9f01a910becbffdd7291574efc031624}
\strng{bibnamehash}{9f01a910becbffdd7291574efc031624}
\strng{authorbibnamehash}{9f01a910becbffdd7291574efc031624}
\strng{authornamehash}{9f01a910becbffdd7291574efc031624}
\strng{authorfullhash}{9f01a910becbffdd7291574efc031624}
\strng{editorbibnamehash}{fcfd2c56673f315d42b77094c6dac7c9}
\strng{editornamehash}{fcfd2c56673f315d42b77094c6dac7c9}
\strng{editorfullhash}{fcfd2c56673f315d42b77094c6dac7c9}
\field{sortinit}{8}
\field{sortinithash}{a231b008ebf0ecbe0b4d96dcc159445f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{{\textbf{NIST} Chemistry {WebBook}, {NIST} Standard Reference Database Number 69}}
\field{title}{Thermochemical Data}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Espinosa-Lopez2018-hb}{article}{}
\name{author}{8}{}{%
{{hash=7774e7f30d364d58e5bd8ff0139ee393}{%
family={Espinosa-López},
familyi={E\bibinithyphendelim L\bibinitperiod},
given={Manuel},
giveni={M\bibinitperiod}}}%
{{hash=7582beba9fec7b1976502aab24df91a0}{%
family={Darras},
familyi={D\bibinitperiod},
given={Christophe},
giveni={C\bibinitperiod}}}%
{{hash=e91eb0e1e762a15d7eb2e69408a190a0}{%
family={Poggi},
familyi={P\bibinitperiod},
given={Philippe},
giveni={P\bibinitperiod}}}%
{{hash=d38d274dcf4c5ca286c7f968fc7a8895}{%
family={Glises},
familyi={G\bibinitperiod},
given={Raynal},
giveni={R\bibinitperiod}}}%
{{hash=15636484f76af0d43b3a04828f276c87}{%
family={Baucour},
familyi={B\bibinitperiod},
given={Philippe},
giveni={P\bibinitperiod}}}%
{{hash=64807a335a3f61004b23a970fc71539b}{%
family={Rakotondrainibe},
familyi={R\bibinitperiod},
given={André},
giveni={A\bibinitperiod}}}%
{{hash=3f3572481ce4d2081043efe58f0fcefc}{%
family={Besse},
familyi={B\bibinitperiod},
given={Serge},
giveni={S\bibinitperiod}}}%
{{hash=cc5c3ac331353b522ce9b4a860924008}{%
family={Serre-Combe},
familyi={S\bibinithyphendelim C\bibinitperiod},
given={Pierre},
giveni={P\bibinitperiod}}}%
}
\strng{namehash}{9452a6d08702278c625147ecf198792b}
\strng{fullhash}{7437c987b4759cbcbfa8756528718cd3}
\strng{bibnamehash}{7437c987b4759cbcbfa8756528718cd3}
\strng{authorbibnamehash}{7437c987b4759cbcbfa8756528718cd3}
\strng{authornamehash}{9452a6d08702278c625147ecf198792b}
\strng{authorfullhash}{7437c987b4759cbcbfa8756528718cd3}
\field{sortinit}{9}
\field{sortinithash}{0a5ebc79d83c96b6579069544c73c7d4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{The objective of this paper is to present the modelling and experimental validation of the 46 kW PEM high pressure water electrolyzer installed on the MYRTE platform, which is a real-scale demonstrator that aims to study the deployment of hydrogen to store the energy associated to intermittent renewable energy source systems. An electrochemical steady-state and semi-empirical submodel coupled with a lumped thermal capacitance dynamic submodel is developed to predict the stack voltage and the stack temperature evolution from instantaneous operating conditions such as the applied current, the gas storage pressure tanks (H2 and O2) and the ambient temperature. The Particle Swarm Optimization algorithm is used to find the electrochemical submodel parameters and a multivariable Matlab-Simulink®{} linked modular mathematical model is developed for validation. Results indicate that within a temperature range of 20--60 °C, and a pressure range of 15--35 bar, the stack voltage and the temperature evolution can be predicted even in transitory operating phases. The strategy used for the parameters identification is explained in detail and can be applied to any PEM water electrolyzer.}
\field{journaltitle}{Renewable Energy}
\field{month}{4}
\field{title}{Modelling and experimental validation of a 46 kW {PEM} high pressure water electrolyzer}
\field{volume}{119}
\field{year}{2018}
\field{pages}{160\bibrangedash 173}
\range{pages}{14}
\keyw{Electrolyzer; Hydrogen; Renewable energy; Photovoltaic hybrid system;Important;Reference;Thesis Proposal;First Principles Model;PEM Model Documentation;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Bernt2020-ne}{article}{}
\name{author}{8}{}{%
{{hash=4f47c6343f444c3e9e61b6d1cfaec14b}{%
family={Bernt},
familyi={B\bibinitperiod},
given={Maximilian},
giveni={M\bibinitperiod}}}%
{{hash=c177d833d21af17e52300051608f23f1}{%
family={Hartig-Wei{ß}},
familyi={H\bibinithyphendelim W\bibinitperiod},
given={Alexandra},
giveni={A\bibinitperiod}}}%
{{hash=b6cba77e5a77c67573d476d0bc563479}{%
family={Tovini},
familyi={T\bibinitperiod},
given={Mohammad\bibnamedelima Fathi},
giveni={M\bibinitperiod\bibinitdelim F\bibinitperiod}}}%
{{hash=98004ae31ff7f9a5990fb3a3d722746c}{%
family={El-Sayed},
familyi={E\bibinithyphendelim S\bibinitperiod},
given={Hany\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=717d5c5793107b9b94edb9973efa30b6}{%
family={Schramm},
familyi={S\bibinitperiod},
given={Carina},
giveni={C\bibinitperiod}}}%
{{hash=144c56327f7fbd32a2b3d1b04d0b3fe7}{%
family={Schröter},
familyi={S\bibinitperiod},
given={Jonas},
giveni={J\bibinitperiod}}}%
{{hash=75dbf0094f16af878431c2031f86891a}{%
family={Gebauer},
familyi={G\bibinitperiod},
given={Christian},
giveni={C\bibinitperiod}}}%
{{hash=b43fb849dae1a0b0667cba868dd9999d}{%
family={Gasteiger},
familyi={G\bibinitperiod},
given={Hubert\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{Wiley}%
}
\strng{namehash}{340cdcb9ff439f3cb68c9dd7a0d3e447}
\strng{fullhash}{c2e014fba1a0245c8e0e05fd0817209d}
\strng{bibnamehash}{c2e014fba1a0245c8e0e05fd0817209d}
\strng{authorbibnamehash}{c2e014fba1a0245c8e0e05fd0817209d}
\strng{authornamehash}{340cdcb9ff439f3cb68c9dd7a0d3e447}
\strng{authorfullhash}{c2e014fba1a0245c8e0e05fd0817209d}
\field{extraname}{1}
\field{sortinit}{1}
\field{sortinithash}{4f6aaa89bab872aa0999fec09ff8e98a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{Abstract This work addresses current challenges in catalyst development for proton exchange membrane water electrolyzers (PEM-WEs). To reduce the amount of iridium at the oxygen anode to levels commensurate with large-scale application of PEM-WEs, high-structured catalysts with a low packing density are required. To allow an efficient development of such catalysts, activity and durability screening tests are essential. Rotating disk electrode measurements are suitable to determine catalyst activity, while accelerated stress tests on the MEA level are required to evaluate catalyst stability.}
\field{journaltitle}{Chem. Ing. Tech.}
\field{month}{1}
\field{number}{1-2}
\field{title}{Current challenges in catalyst development for {PEM} water electrolyzers}
\field{volume}{92}
\field{year}{2020}
\field{pages}{31\bibrangedash 39}
\range{pages}{9}
\keyw{Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Gorgun2006-ej}{article}{}
\name{author}{1}{}{%
{{hash=3d4456a8b8d14ab8ba8ae8c89e9c4446}{%
family={Görgün},
familyi={G\bibinitperiod},
given={Haluk},
giveni={H\bibinitperiod}}}%
}
\strng{namehash}{3d4456a8b8d14ab8ba8ae8c89e9c4446}
\strng{fullhash}{3d4456a8b8d14ab8ba8ae8c89e9c4446}
\strng{bibnamehash}{3d4456a8b8d14ab8ba8ae8c89e9c4446}
\strng{authorbibnamehash}{3d4456a8b8d14ab8ba8ae8c89e9c4446}
\strng{authornamehash}{3d4456a8b8d14ab8ba8ae8c89e9c4446}
\strng{authorfullhash}{3d4456a8b8d14ab8ba8ae8c89e9c4446}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{This paper describes a dynamic model for PEM electrolyzer based on conservation of mole balance at the anode and the cathode. A further feature of the model is it includes water phenomena, electro-osmotic drag and diffusion, through the membrane. The model considers PEM electrolyzer to be composed of four ancillaries: anode, cathode, membrane and voltage ancillary. Additionally, hydrogen storage dynamics is presented. The developed model is suitable for determining control strategy that will ensure efficient and reliable operation of the electrolyzer. Moreover, the dynamic model can be integrated with renewable energy systems models to design, analyze and optimize sustainable energy systems. The study illustrates the dynamic interactions within a PEM electrolyzer and shows the necessity of the proposed approach of separate ancillaries.}
\field{journaltitle}{Int. J. Hydrogen Energy}
\field{month}{1}
\field{number}{1}
\field{title}{Dynamic modelling of a proton exchange membrane ({PEM}) electrolyzer}
\field{volume}{31}
\field{year}{2006}
\field{pages}{29\bibrangedash 38}
\range{pages}{10}
\keyw{Important;Unread;Thesis Proposal;First Principles Model;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{noauthor_undated-so}{misc}{}
\list{language}{1}{%
{en}%
}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labeltitlesource}{title}
\field{abstract}{Get AS market prices, including day-ahead market clearing prices, BES, commercially significant constraints, shadow prices and RPRS.}
\field{booktitle}{Market Prices}
\field{howpublished}{\url{https://www.ercot.com/mktinfo/prices}}
\field{note}{Accessed: 2023-11-16}
\field{title}{{ERCOT} Market Prices}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Gagnon_undated-so}{misc}{}
\name{author}{3}{}{%
{{hash=3b2a3e140fc173994f9c23761e43de14}{%
family={Gagnon},
familyi={G\bibinitperiod},
given={Pieter},
giveni={P\bibinitperiod}}}%
{{hash=373a29c40a28810fa89a241b7ec1c88a}{%
family={Cowiestoll},
familyi={C\bibinitperiod},
given={Brady},
giveni={B\bibinitperiod}}}%
{{hash=f186d10ac274113e3e2a57ed5f577d8d}{%
family={Schwarz},
familyi={S\bibinitperiod},
given={Marty},
giveni={M\bibinitperiod}}}%
}
\strng{namehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{fullhash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{bibnamehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{authorbibnamehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{authornamehash}{291be4dda9dbef64eb2177af58d5f1cb}
\strng{authorfullhash}{291be4dda9dbef64eb2177af58d5f1cb}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Cambium 2022 Scenario Descriptions and Documentation}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Water_Innovations_Inc_undated-io}{report}{}
\name{author}{1}{}{%
{{hash=d8422c5b8c8dad99035e60e8edda5d4e}{%
family={{Water Innovations Inc.}},
familyi={W\bibinitperiod}}}%
}
\strng{namehash}{d8422c5b8c8dad99035e60e8edda5d4e}
\strng{fullhash}{d8422c5b8c8dad99035e60e8edda5d4e}
\strng{bibnamehash}{d8422c5b8c8dad99035e60e8edda5d4e}
\strng{authorbibnamehash}{d8422c5b8c8dad99035e60e8edda5d4e}
\strng{authornamehash}{d8422c5b8c8dad99035e60e8edda5d4e}
\strng{authorfullhash}{d8422c5b8c8dad99035e60e8edda5d4e}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Operating Cost for {WDI} Deionized Water System}
\field{type}{techreport}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Thomas_Fuller_undated-ig}{incollection}{}
\name{author}{1}{}{%
{{hash=28f786c68b7f0f2d9a04c41f10382123}{%
family={Thomas\bibnamedelima Fuller},
familyi={T\bibinitperiod\bibinitdelim F\bibinitperiod},
given={John\bibnamedelima Harb},
giveni={J\bibinitperiod\bibinitdelim H\bibinitperiod}}}%
}
\strng{namehash}{28f786c68b7f0f2d9a04c41f10382123}
\strng{fullhash}{28f786c68b7f0f2d9a04c41f10382123}
\strng{bibnamehash}{28f786c68b7f0f2d9a04c41f10382123}
\strng{authorbibnamehash}{28f786c68b7f0f2d9a04c41f10382123}
\strng{authornamehash}{28f786c68b7f0f2d9a04c41f10382123}
\strng{authorfullhash}{28f786c68b7f0f2d9a04c41f10382123}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Electrochemical Engineering}
\field{title}{Chapter 3 - Electrochemical Kinetics}
\keyw{Read;Fuller and Harb;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Falcao2020-vj}{article}{}
\name{author}{2}{}{%
{{hash=690fda48f26ce57fff6964f46aac3731}{%
family={Falcão},
familyi={F\bibinitperiod},
given={D\bibnamedelima S},
giveni={D\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
{{hash=ef79267fdbcec0949d0fc1d40cdf8bc8}{%
family={Pinto},
familyi={P\bibinitperiod},
given={A\bibnamedelima M\bibnamedelimb F\bibnamedelima R},
giveni={A\bibinitperiod\bibinitdelim M\bibinitperiod\bibinitdelim F\bibinitperiod\bibinitdelim R\bibinitperiod}}}%
}
\strng{namehash}{5b3e77578da9424160d6dac2399b45fe}
\strng{fullhash}{5b3e77578da9424160d6dac2399b45fe}
\strng{bibnamehash}{5b3e77578da9424160d6dac2399b45fe}
\strng{authorbibnamehash}{5b3e77578da9424160d6dac2399b45fe}
\strng{authornamehash}{5b3e77578da9424160d6dac2399b45fe}
\strng{authorfullhash}{5b3e77578da9424160d6dac2399b45fe}
\field{sortinit}{2}
\field{sortinithash}{8b555b3791beccb63322c22f3320aa9a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{From a sustainability perspective, a synergy between hydrogen, electricity and Renewable Energy Sources (RES) is particularly attractive. A combination of Proton Exchange Membrane (PEM) fuel cells and PEM electrolyzers provides a back-up system for RES avoiding the intermittency: electrolyzers convert the excess of energy from RES into hydrogen and PEM fuel cells use this hydrogen to convert it back into electricity when it is needed. Models are an essential tool in electrolyzer development because it allows to understand the influence of different parameters on the electrolyzer performance enabling an efficient simulation, design and optimization of electrolyzer systems. Modelling studies have been developed with different degrees of complexity. This review presents a compilation of published models focusing on the main equations used to predict cell voltage, including reversible voltage, activation losses, ohmic losses and mass transport losses. Special sections are devoted to dynamic behaviour, two-phase flow effects and inclusion of thermal effects in model development. Also, a brief description of empirical/semi-empirical models is provided. This review aims to provide the main guidelines for the beginners on PEM electrolyzer modelling and, for this purpose, includes a section with the technology basic principles.}
\field{journaltitle}{J. Clean. Prod.}
\field{month}{7}
\field{title}{A review on {PEM} electrolyzer modelling: Guidelines for beginners}
\field{volume}{261}
\field{year}{2020}
\field{pages}{121184}
\range{pages}{1}
\keyw{PEM Electrolyzer; Modelling; Review;Thesis Proposal;Reference;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\entry{Maier_undated-zb}{misc}{}
\name{author}{6}{}{%
{{hash=1a0f81c3ae8c37af8cf117c7250f6909}{%
family={Maier},
familyi={M\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=9232ccab869a1e28b236a776037240ea}{%
family={Smith},
familyi={S\bibinitperiod},
given={K},
giveni={K\bibinitperiod}}}%
{{hash=92fd1839aee7399e5cfc61c3950b87db}{%
family={Dodwell},
familyi={D\bibinitperiod},
given={J},
giveni={J\bibinitperiod}}}%
{{hash=3d345afc3af12887825d2836cb7095b6}{%
family={Hinds},
familyi={H\bibinitperiod},
given={G},
giveni={G\bibinitperiod}}}%
{{hash=23fda2cdaf43f9517d551fcfe4356b6a}{%
family={Shearing},
familyi={S\bibinitperiod},
given={P\bibnamedelima R},
giveni={P\bibinitperiod\bibinitdelim R\bibinitperiod}}}%
{{hash=f040fb59f3fac6de4e5ac71bf221ab8b}{%
family={Bretta},
familyi={B\bibinitperiod},
given={D\bibnamedelima J\bibnamedelima L},
giveni={D\bibinitperiod\bibinitdelim J\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
}
\strng{namehash}{5fafe71287b1afad46a488fb964610cb}
\strng{fullhash}{3631483d0b5a1dd1ed36da27e34207b1}
\strng{bibnamehash}{3631483d0b5a1dd1ed36da27e34207b1}
\strng{authorbibnamehash}{3631483d0b5a1dd1ed36da27e34207b1}
\strng{authornamehash}{5fafe71287b1afad46a488fb964610cb}
\strng{authorfullhash}{3631483d0b5a1dd1ed36da27e34207b1}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Mass Transport in {PEM} Water Electrolysers: A Review}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Christopher_Bryce_Capuano_Morgan_Elizabeth_Pertosos_Nemanja_Danilovic2018-lw}{patent}{}
\name{author}{1}{}{%
{{hash=da7e29db7fd550d698ea303079dd7619}{%
family={{Christopher Bryce Capuano, Morgan Elizabeth Pertosos, Nemanja Danilovic}},
familyi={C\bibinitperiod}}}%
}
\strng{namehash}{da7e29db7fd550d698ea303079dd7619}
\strng{fullhash}{da7e29db7fd550d698ea303079dd7619}
\strng{bibnamehash}{da7e29db7fd550d698ea303079dd7619}
\strng{authorbibnamehash}{da7e29db7fd550d698ea303079dd7619}
\strng{authornamehash}{da7e29db7fd550d698ea303079dd7619}
\strng{authorfullhash}{da7e29db7fd550d698ea303079dd7619}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{3}
\field{number}{2018/0062192 A1}
\field{title}{Membrane Electrode Assembly and Method of Making the Same}
\field{year}{2018}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Nicolas_Guillet2014-tq}{patent}{}
\name{author}{1}{}{%
{{hash=fb9b9d452af2afb9c79844102a44845d}{%
family={Nicolas\bibnamedelima Guillet},
familyi={N\bibinitperiod\bibinitdelim G\bibinitperiod},
given={Eric\bibnamedelima Mayousse},
giveni={E\bibinitperiod\bibinitdelim M\bibinitperiod}}}%
}
\strng{namehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{fullhash}{fb9b9d452af2afb9c79844102a44845d}
\strng{bibnamehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{authorbibnamehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{authornamehash}{fb9b9d452af2afb9c79844102a44845d}
\strng{authorfullhash}{fb9b9d452af2afb9c79844102a44845d}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{month}{5}
\field{number}{20140116877 A1}
\field{title}{Membrane Electrode Assembly for an Electrolysis Device}
\field{year}{2014}
\keyw{Dynamic PEM Paper}
\endentry
\entry{Bernt2020-af}{article}{}
\name{author}{4}{}{%
{{hash=42f5e61f9bfea36a87738c04d6c7e8fb}{%
family={Bernt},
familyi={B\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=dedf87e9e46ba2b772e2a6e5fceaafac}{%
family={Schröter},
familyi={S\bibinitperiod},
given={J},
giveni={J\bibinitperiod}}}%
{{hash=0f78499390c0ba16de1c2142fd4e7792}{%
family={Möckl},
familyi={M\bibinitperiod},
given={M},
giveni={M\bibinitperiod}}}%
{{hash=05008cfe2fe0e12a383f5e7b100b9f5a}{%
family={Gasteiger},
familyi={G\bibinitperiod},
given={H\bibnamedelima A},
giveni={H\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
}
\list{language}{1}{%
{en}%
}
\list{publisher}{1}{%
{IOP Publishing}%
}
\strng{namehash}{35c8fc4e55b15a2d80412860b5a05bd6}
\strng{fullhash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\strng{bibnamehash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\strng{authorbibnamehash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\strng{authornamehash}{35c8fc4e55b15a2d80412860b5a05bd6}
\strng{authorfullhash}{1a2ccffdfcf0ccd3f4efb9d9f18f5525}
\field{extraname}{2}
\field{sortinit}{3}
\field{sortinithash}{ad6fe7482ffbd7b9f99c9e8b5dccd3d7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{abstract}{In this study, on-line mass spectrometry is used to determine hydrogen permeation during proton exchange membrane water electrolyzer (PEM-WE) operation for a wide range of current densities (0--6 A cm−2) and operating pressures (1--30 bar, differential pressure). H2 permeation measurements with a permeation cell setup, i.e., without applying a current, show a linear correlation between permeation rate and H2 partial pressure, indicating diffusion as the main crossover mechanism. Measurements with full membrane electrode assemblies (MEAs) during PEM-WE operation reveal a significant increase of the gas permeation rate at high current densities, by up to $\approx$20-fold at 1 bar H2 and up to $\approx$1.2-fold at 30 bar H2 (Nafion®{} 212 or Nafion®{} 117 membrane; Ir-black (anode) and Pt/C (cathode)). Recently, H2 super-saturation of the ionomer phase in the cathode catalyst layer was shown to be the reason for this increase, and we discuss the impact of this effect for different electrode compositions and operating conditions. Finally, the determined H2 permeation rates and electrolyzer performance are used to discuss the overall PEM-WE efficiency for different membrane thicknesses and it is shown that the formation of an explosive gas mixture in the anode at low current densities requires additional mitigation strategies.}
\field{journaltitle}{J. Electrochem. Soc.}
\field{month}{8}
\field{number}{12}
\field{title}{Analysis of Gas Permeation Phenomena in a {PEM} Water Electrolyzer Operated at High Pressure and High Current Density}
\field{volume}{167}
\field{year}{2020}
\field{pages}{124502}
\range{pages}{1}
\keyw{Read;Important;Reference;Thesis Proposal References;Dynamic PEM Paper}
\endentry
\enddatalist
\endrefsection
\endinput