Bubble-induced versus thermodynamic voltage losses during pressurized alkaline water electrolysis
Hannes Rox, Feng Liang, Robert Baumann, Mateusz M. Marzec, Krystian Sokołowski, Xuegeng Yang, Andrés F. Lasagni, Roel van de Krol, Kerstin Eckert ∗ · 2026
Read Bubble-induced versus thermodynamic voltage losses during pressurized alkaline water electrolysis by Hannes Rox, Feng Liang, Robert Baumann, Mateusz M. Marzec, Krystian Sokołowski, Xuegeng Yang, Andrés F. Lasagni, Roel van de Krol, Kerstin Eckert ∗ (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-26855.txt.
Introduction
Rox and co-authors study how bubbles cut efficiency in pressurized alkaline water electrolysis. Making hydrogen at high pressure saves later compression costs, but bubbles raise voltage losses. Using laser-written nickel electrodes with pillars of varied size, the team changed detached bubble sizes. Larger pillars made larger bubbles at every pressure from 1 to 6 bar. At high current density, smaller bubbles at pressure cut overpotential by up to 60 millivolts. That gain offset the thermodynamic penalty from pressure.
Catalog record
| CREATOR | Hannes Rox, Feng Liang, Robert Baumann, Mateusz M. Marzec, Krystian Sokołowski, Xuegeng Yang, Andrés F. Lasagni, Roel van de Krol, Kerstin Eckert ∗ |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61784 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.26855
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bubble-induced versus thermodynamic voltage losses during pressurized alkaline water electrolysis Hannes Rox, Feng Liang, Robert Baumann, Mateusz M. Marzec, Krystian Sokołowski, Xuegeng Yang, Andrés F. Lasagni, Roel van de Krol, Kerstin Eckert ∗ Abstract Understanding how bubbles influence the efficiency of water electrolysis is crucial to achieve economically competitive hydrogen, generated by renewable energy sources, such as wind and solar power. Water electrolysis is typically performed at high pressures to reduce the cost of energy-intensive mechanical compression of the produced H 2 \text{H}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} . Thus, a better understanding of how the absolute pressure affects electrochemical performance and bubble size is necessary. In general, bubble sizes decrease as the pressure increases. Using different-sized pillar-patterned Ni electrodes generated by Direct Laser Writing, the detached bubble sizes can be modified even at elevated pressures. As the pillar size increases, the bubbles become larger at all pressures investigated from 1 to 6 bar 6\text{\,}\mathrm{bar} . At a current density of − 25 mA cm − 2 -25\text{\,}\mathrm{mA}\text{\,}{\mathrm{cm}}^{-2} , the cathodic potential increases with pressure according to the thermodynamic voltage losses given by the Nernst equation ( ≈ \approx 23 mV 23\text{\,}\mathrm{mV} at p = 6 bar p=$6\text{\,}\mathrm{bar}$ ). Surprisingly, increasing the current density to 100 mA cm − 2 100\text{\,}\mathrm{mA}\text{\,}{\mathrm{cm}}^{-2} leads to a reduction of the overpotential by up to ≈ \approx 60 mV 60\text{\,}\mathrm{mV} . Reduced bubble sizes at increased pressures minimize the losses caused by the bubbles, thereby compensating for the thermodynamic voltage penalty. Applying the Buckingham Π \Pi -theorem enables the derivation of dimensionless numbers to characterize the ratio of bubble-induced and thermodynamic voltage losses. Broader Context The Belém Declaration, recently launched at COP30, strengthens the global agenda for green industrialization to accelerate the decarbonization of heavy emitting industries. Therefore, a rapid increase in the use of low-emission technologies is needed to reduce greenhouse gas emissions by the end of the decade. One of these low-emission technologies is green hydrogen, produced by water electrolysis using renewable energy sources like solar or wind power. However, although green hydrogen reduces environmental impact, the costs of water electrolysis remains high, at up to eight times that of gray hydrogen. Therefore, reducing the operating and capital expenditures by improved electrode and cell designs is essential. Considerable losses are caused by the evolving hydrogen and oxygen bubbles by increasing ohmic resistances and blocking electrochemically active sites. Consequently, a better understanding of the effects of electrogenerated bubbles on the electrochemical reaction is needed to improve the overall efficiency and reduce the costs of green hydrogen production. 1 Introduction Understanding the ongoing processes of gas evolution on electrodes is essential for many electrochemical systems, including water splitting [ 61 ] , chloro-alkali electrolysis [ 12 ] or \text{CO}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} electroreduction [ 36 ] . Over the past two decades, many studies have focused on the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER) because they are essential to improve the overall efficiency of water electrolysis [ 17 , 39 , 60 ] . This is necessary to replace fossil fuels with hydrogen produced using low- or zero-carbon energy sources, such as solar or wind-derived electricity [ 25 ] . Nevertheless, the efficiency and economic competitiveness of large-scale green hydrogen production are still lacking. The presence of evolving hydrogen and oxygen bubbles can lead to significant losses due to increasing ohmic resistance and blocking of electrochemically active sites [ 23 , 51 , 4 ] . Additionally, the mass transfer and the actual current density are influenced by electrode coverage [ 55 , 53 , 54 , 16 ] . Various studies have shown that mitigating these bubble effects can significantly lower the electrode potentials by 9- 25\text{\,}\% [ 33 , 57 , 52 ] . The reduction of the bubble-induced losses can be achieved by optimizing the electrode’s electrocatalysts [ 27 ] , morphology [ 42 , 19 , 14 ] or surface [ 14 , 32 , 2 ] , enhancing bubble coalescence [ 6 , 38 ] , or applying external forces [ 42 , 14 , 34 , 5 , 28 ] . However, process parameters like the absolute pressure in the cell also influence the bubble evolution. Since industrial electrolyzers are usually operated under elevated pressures up to 30- 50\text{\,}\mathrm{bar} [ 17 ] , the understanding of the influence of the pressure on the gas evolution during water splitting reaction is needed. To the best of the authors knowledge, the only comprehensive study reporting the characterization of electrogenerated bubbles at higher pressures in alkaline electrolyte ( 1\text{\,}\mathrm{M} KOH) is the work of 50 [ 50 ] . Here, both oxygen and hydrogen bubbles generated under various pressures ( p=0.25-$30\text{\,}\mathrm{bar}$ ) were investigated. After this early work, 35 [ 35 ] and 37 [ 37 ] studied the impact of elevated pressure on photoelectrochemical water splitting. In general, the increase in pressure leads to an increase of the dissolved \text{H}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} concentration [ 35 ] and decrease in bubble size [ 50 , 35 , 37 ] . Additionally, the thermal behavior of the electrolyzer is significantly impacted by pressure. Increasing pressure leads to reduced water vaporization and thus reduced water losses [ 40 , 31 ] . Furthermore, the rising pressure also increases the thermodynamic voltage losses [ 35 , 40 , 31 , 22 ] . This can be understood by reformulating the Nernst equation in terms of the absolute pressure p [ 41 , 18 ] . Therefore, the dependence of the potential E on the partial pressure p_{\mathrm{{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}}} and p_{\mathrm{{}\mathrm{O}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}}} can be defined as where E^0 denotesthestandardpotential, R theuniversalgasconstant, z thechargenumber, F theFaradaicconstantand p_0 theatmosphericpressure.InEq.\penalty\ \ref{eq:nernst},itisassumedthatthefugacitycoefficientsof{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}and{}\mathrm{O}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}areequalto1.Thisisareasonableassumptionsince,atapressureof$10\text{\,}\mathrm{bar}$,thedeviationfromtheNernstpotentialusingtheactualfugacitycoefficientsfor{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}(1.03)andfor{}\mathrm{O}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}(0.92)isonly$-0.5\text{\,}\mathrm{mV}$\cite[cite]{[\@@bibref{Number}{Vallejos-Burgos2019}{}{}]}.\par Inthepresentcaseofasymmetricalwaterelectrolyzerwithaconstantpressureatbothanodeandcathodeduetotheabsenceofaseparator,itcanbeassumedthat p_ \text{H}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} = p_ \text{O}{\vphantom{\text{X}}}_{\vphantom{\text{2}}\smash[t]{\text{2}}}^{\vphantom{\smash[t]{\text{2}}}\hphantom{\text{2}}\text{}} = p \cite[cite]{[\@@bibref{Number}{Millet2022}{}{}]}.Here, p denotestheabsolutepressureoftheelectrolyzer.Furthermore,byneglectingthesaturatingpressureofwatervaporin{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}and{}\mathrm{O}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}},Eq.\penalty\ \ref{eq:nernst}canbereformulatedas\begin{aligned} E=E^{0}+\frac{\mathscr{R}T}{zF}\ln\left(\left(\frac{p}{p_{0}}\right)^{3/2}\right).\end{aligned}Assumingaconstantambienttemperature T= 298.15\text{\,}\mathrm{K} andpressure p_0 = 1\text{\,}\mathrm{bar} ,andthusaconstantstandardpotential E^0 ,thetheoreticalpressure-inducedthermodynamiclosses E_theor canbecalculatedwiththesecondtermoftheaboveequation\cite[cite]{[\@@bibref{Number}{Millet2022}{}{}]}:\begin{aligned} E_{\mathrm{theor}}=\frac{\mathscr{R}T}{zF}\ln\left(\left(\frac{p}{p_{0}}\right)^{3/2}\right)\end{aligned}Thisleadstoadditionalcellvoltagesof ≈ $66\text{\,}\mathrm{mV}$whenincreasingtheoperatingpressurefrom$1\text{\,}\mathrm{bar}$to$30\text{\,}\mathrm{bar}$.\par Incontrast,bubble-inducedlossesareloweredasthebubblesizesaredecreasedsimultaneously\cite[cite]{[\@@bibref{Number}{Sillen1983, Liang2024}{}{}]}.However,thiseffecthasnotbeenthoroughlyinvestigatedforalkalinewaterelectrolysis,whichisstillthemostmaturetechnologytoproducegreenhydrogenandistypicallyperformedatcurrentdensitiesof100to$500\text{\,}\mathrm{mA}\text{\,}{\mathrm{cm}}^{-2}$\cite[cite]{[\@@bibref{Number}{ElEmam2019}{}{}]}.\cite[citeauthor]{\@@bibref{Authors}{Sillen1983}{}{}}focusedprimarilyontheevolvingbubblesanddidnotdiscusstheinfluenceoftheoperatingpressureontheoverpotential\cite[cite]{[\@@bibref{Number}{Sillen1983}{}{}]}.Inaddition,photoelectrochemicalwatersplittingrunswitharathersmallcurrentdensityof$10\text{\,}\mathrm{mA}\text{\,}{\mathrm{cm}}^{-2}$\cite[cite]{[\@@bibref{Number}{Liang2024}{}{}]}.Themostrecentworkisfrom\cite[citeauthor]{\@@bibref{Authors}{Wu2025}{}{}}\cite[cite]{[\@@bibref{Number}{Wu2025}{}{}]},investigatingthegrowthofasinglehydrogenbubbleonaPtelectrodewithanaerophilicPTFEspotasnucleationsite.Incontrasttoallpreviouswork,largerbubblesarereportedatelevatedpressuresupto$20\text{\,}\mathrm{bar}$.\cite[citeauthor]{\@@bibref{Authors}{Wu2025}{}{}}relatethistoanincreaseintheaerophilicityofelectrodesurface\cite[cite]{[\@@bibref{Number}{Wu2025}{}{}]}.However,as{}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{2}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{2}}\mathrm{}}\mathrm{SO}{\vphantom{\mathrm{X}}}_{\vphantom{\mathrm{2}}\smash[t]{\mathrm{4}}}^{\vphantom{\smash[t]{\mathrm{2}}}\hphantom{\mathrm{4}}\mathrm{}}wasusedaselectrolyteandonlysinglebubblesarestudied,theseresultsaredifficulttotransfertoindustrialalkalineelectrolysis.Consequently,littleisknownabouttheinfluenceofabsolutepressureonbubble-inducedlossesinalkalinewaterelectrolysis.\par Toaddressthisgap,differentbubblesizesareneededtoseehowtheyaffecttheoverallelectrodepotentialatelevatedpressures.In\cite[citeauthor]{\@@bibref{Authors}{Rox2025b}{}{}}weshowedthatthedetachedbubblesizeissignificantlychangedbypillar-likesurfacepatternsproducedviaDirectLaserWriting(DLW)\cite[cite]{[\@@bibref{Number}{Rox2025b}{}{}]}.Thisisduetothedualwettingbehaviorofthesesurfaces,whichcanbeunderstoodasahierarchicalsurfacewithdefinedhydrophobicnucleationsitesincorporatedintoasystemofhydrophilicmicrochannels.Basedonthecurrentunderstandingofthebubbleevolutiononthepillar-patternedelectrodes,thebubblesarepinnedtothehydrophobicpatchesduringtheirgrowth\cite[cite]{[\@@bibref{Number}{Rox2025b}{}{}]}.Inaddition,theforcesactingonthegrowingbubbles\cite[cite]{[\@@bibref{Number}{Bashkatov2022, Hossain2022}{}{}]}aremodifiedbythesesurfacepatterns.Here,especiallytheinterfacialtensionforce F_S (see.Eq.\penalty\ \ref{eq:interfacial_tension_force})andthecontactpressureforce F_CP (see.Eq.\penalty\ \ref{eq:contact_pressure_force})areinfluencedbythemodifiedcontactangle θ andcontactradius r_c ofthebubble:\begin{aligned} F_{\mathrm{S}}&=-2\cdot\pi\cdot r_{\mathrm{c}}\cdot\gamma\cdot\sin\theta\cdot f_{\mathrm{S}}\\ F_{\mathrm{CP}}&=\frac{2\cdot\gamma}{r_{\mathrm{B}}}\cdot\pi\cdot r_{\mathrm{c}}^{2}\end{aligned}AschematicdrawingofthesetwoforcesactingonabubblegrowingatahorizontalelectrodecanbefoundintheESI(seeFig.\penalty\ \ref{fig:appendix_bubble_forces}). F_S actsasaretardingforceatthethree-phasecontactlinewithradius r_c ,drawingthebubblewithradius r_B towardstheelectrodesurfaceduetothesurfacetension γ \cite[cite]{[\@@bibref{Number}{Bashkatov2022, Demirk\T1\i r2024, Meulenbroek2021}{}{}]}.Onroughsurfaces,thethree-phasecontactlinecanbecutintoadiscontinuousstatebyelectrolyte-filledpocketsbelowthebubble,whichcanbetakenintoaccountbythesurfaceareafraction f_S \cite[cite]{[\@@bibref{Number}{Li2023}{}{}]}.Incontrast, F_CP actsasadetachingforce\cite[cite]{[\@@bibref{Number}{Bashkatov2022}{}{}]},whichdominateswhen sin θ approaches0.Consequently,changingthesizeofthehydrophobicnucleationsizesresultsinachangeof r_c andthereforeisexpectedtochangethesizeofthedetachedbubbles.\par\par Inthepresentstudy,differentsizedpillar-patternsareappliedtohigh-purityNi-foilsusingDLWtomodifythedetachedbubblesize.ThisenablesthestudyoftheinfluenceofdetachedbubblesizeandabsolutecellpressureonthecathodicpotentialduringHERinanalkalineelectrolyteatelevatedcurrentdensitiesandpressures.Linkingtheseresultswiththetheoreticalprinciplesofalkalinewaterelectrolysisprovidesnewinsightsintotheimportanceofminimizingbubble-inducedlossestoimprovetheoverallefficiency.\par 2 Results and discussion Electrode characterization To ensure maximum flexibility in terms of structure size and thus, detached bubble size, DLW was used to texture the electrode surfaces. The side length of the pillars, here referred to as spatial period \Lambda , was changed from 30\text{\,}\mathrm{\SIUnitSymbolMicro m} to 100\text{\,}\mathrm{\SIUnitSymbolMicro m} at a constant depth of \approx 7\text{\,}\mathrm{\SIUnitSymbolMicro m} . A total of 4 different laser structures were studied, as shown in Fig. 1 . In addition, one non-structured electrode (NSE) was used as comparison. A summary of all characteristics of the studied electrodes is provided in Table 1 . Table 1 : Spatial period of pillar structure \Lambda , average surface roughness S_{\mathrm{a}} , developed interfacial area ratio S_{\mathrm{dr}} , and water contact angle \theta of all electrodes in comparison to the non-structured Ni-foil (NSE). \Lambda S_{\mathrm{a}} S_{\mathrm{dr}} \theta ( \mathrm{\SIUnitSymbolMicro m} ) ( \mathrm{\SIUnitSymbolMicro m} ) (%) ( \mathrm{\SIUnitSymbolDegree} ) —— NSE —— 55.2\pm 0.5 30 1.26 9.12 114.7\pm 3.4 40 1.16 6.70 81.3\pm 3.7 60 0.927 3.64 77.0\pm 3.2 100 0.627 1.49 86.7\pm 1.5 Figure 1 : Overview of all laser-structured electrodes: (a) Microscopic and (b) confocal images of electrode structures with definition of the