### Recent Submissions

• #### Au···H-C Hydrogen Bonds as Design Principle in Gold(I) Catalysis

(Angewandte Chemie, Wiley, 2021-07-27) [Article]
Secondary ligand metal interactions are decisive in many catalytic transformations. While arene-gold interactions have repeatedly been reported as critical structural feature in many high-performance gold catalysts, we herein report that these interactions can also be replaced by Au···H-C hydrogen bonds without suffering any reduction in catalytic performance. Systematic experimental and computational studies on a series of ylide-substituted phosphines featuring either a PPh 3 ( Ph YPhos) or PCy 3 ( Cy YPhos) moiety showed that the arene-gold interaction in the aryl-substituted compounds is efficiently compensated by the formation of Au···H-C hydrogen bonds. The strongest interaction is found with the C-H moiety next to the onium center, which due to the polarization results in remarkably strong interactions with the shortest Au···H-C hydrogen bonds reported to date. Calorimetric studies on the formation of the gold complexes further confirmed that the Ph YPhos and Cy YPhos ligands form similarly stable complexes. Consequently, both ligands showed the same catalytic performance in the hydroamination, hydro­phe­noxy­lation and hydrocarboxylation of alkynes, thus demon­strating that Au···H-C hydrogen bonds are equally suited for the generation of highly effective gold catalysts than gold-arene interactions. The generality of this observation was confirmed by a comparative study between a biaryl phosphine ligand and its cyclohexyl-substituted derivative, which again showed identical catalytic performance. These observations clearly support Au···H-C hydrogen bonds as fundamental secondary interactions in gold catalysts, thus further increasing the number of design elements that can be used for future catalyst construction.
• #### Recovery of assessed global fish stocks remains uncertain

(Proceedings of the National Academy of Sciences, Proceedings of the National Academy of Sciences, 2021-07-26) [Article]
Concerns over overexploitation have fueled an ongoing debate on the current state and future prospects of global capture fisheries, associated threats to marine biodiversity, and declining yields available for human consumption. Management reforms have aimed to reduce fishing pressure and recover depleted stocks to biomass and exploitation rates that allow for maximum sustainable yield. Recent analyses suggest that scientifically assessed stocks, contributing over half of global marine fish catch, have, on average, reached or even exceeded these targets, suggesting a fundamental shift in the effectiveness of fisheries governance. However, such conclusions are based on calculations requiring specific choices to average over high interstock variability to derive a global trend. Here we evaluate the robustness of these conclusions by examining the distribution of recovery rates across individual stocks and by applying a diversity of plausible averaging techniques. We show that different methods produce markedly divergent trajectories of global fisheries status, with 4 of 10 methods suggesting that recovery has not yet been achieved, with up to 48% of individual stocks remaining below biomass targets and 40% exploited above sustainable rates. Furthermore, recent rates of recovery are only marginally different from zero, with up to 46% of individual stocks trending downward in biomass and 29% of stocks trending upward in exploitation rate. These results caution against overoptimistic assessments of fisheries writ large and support a precautionary management approach to ensure full rebuilding of depleted fisheries worldwide.
• #### The Ionic Liquid–H2O Interface: A New Platform for the Synthesis of Highly Crystalline and Molecular Sieving Covalent Organic Framework Membranes

(ACS Applied Materials & Interfaces, American Chemical Society (ACS), 2021-07-26) [Article]
Covalent organic frameworks (COFs) are highly porous crystalline polymers with uniform pores and large surface areas. Combined with their modular design principle and excellent properties, COFs are an ideal candidate for separation membranes. Liquid-liquid interfacial polymerization is a well-known approach to synthesize membranes by reacting two monomers at the interface. However, volatile organic solvents are usually used, which may disturb the liquid-liquid interface and affect the COF membrane crystallinity due to solvent evaporation. Simultaneously, the domain size of the organic solvent-water interface, named the reaction zone, can hardly be regulated, and the diffusion control of monomers for favorable crystallinity is only achieved in the water phase. These drawbacks may limit the widespread applications of liquid-liquid interfacial polymerization to synthesize diverse COF membranes with different functionalities. Here, we report a facile strategy to synthesize a series of imine-linked freestanding COF membranes with different thicknesses and morphologies at tunable ionic liquid (IL)-H<sub>2</sub>O interfaces. Due to the H-bonding of the catalysts with amine monomers and the high viscosity of the ILs, the diffusion of the monomers was simultaneously controlled in water and in ILs. This resulted in the exceptionally high crystallinity of freestanding COF membranes with a Brunauer-Emmett-Teller (BET) surface area up to 4.3 times of that synthesized at a dichloromethane-H<sub>2</sub>O interface. By varying the alkyl chain length of cations in the ILs, the interfacial region size and interfacial tension could be regulated to further improve the crystallinity of the COF membranes. As a result, the as-fabricated COF membranes exhibited ultrahigh permeance toward water and organic solvents and excellent selective rejection of dyes.
• #### Phase Transitions in Poly(vinylidene fluoride)/Polymethylene-Based Diblock Copolymers and Blends

(Polymers, MDPI AG, 2021-07-24) [Article]
The crystallization and morphology of two linear diblock copolymers based on polymethylene (PM) and poly(vinylidene fluoride) (PVDF) with compositions PM23-b-PVDF77 and PM38-b-PVDF62 (where the subscripts indicate the relative compositions in wt%) were compared with blends of neat components with identical compositions. The samples were studied by SAXS (Small Angle X-ray Scattering), WAXS (Wide Angle X-ray Scattering), PLOM (Polarized Light Optical Microscopy), TEM (Transmission Electron Microscopy), DSC (Differential Scanning Calorimetry), BDS (broadband dielectric spectroscopy), and FTIR (Fourier Transform Infrared Spectroscopy). The results showed that the blends are immiscible, while the diblock copolymers are miscible in the melt state (or very weakly segregated). The PVDF component crystallization was studied in detail. It was found that the polymorphic structure of PVDF was a strong function of its environment. The number of polymorphs and their amount depended on whether it was on its own as a homopolymer, as a block component in the diblock copolymers or as an immiscible phase in the blends. The cooling rate in non-isothermal crystallization or the crystallization temperature in isothermal tests also induced different polymorphic compositions in the PVDF crystals. As a result, we were able to produce samples with exclusive ferroelectric phases at specific preparation conditions, while others with mixtures of paraelectric and ferroelectric phases.
• #### Pushing the Limits of Flexibility and Stretchability of Solar Cells: A Review

Emerging forms of soft, flexible, and stretchable electronics promise to revolutionize the electronics industries of the future offering radically new products that combine multiple functionalities, including power generation, with arbitrary form factor. For example, skin-like electronics promise to transform the human-machine-interface, but the softness of the skin is incompatible with traditional electronic components. To address this issue, new strategies toward soft and wearable electronic systems are currently being pursued, which also include stretchable photovoltaics as self-powering systems for use in autonomous and stretchable electronics of the future. Here recent developments in the field of stretchable photovoltaics are reviewed and their potential for various emerging applications are examined. Emphasis is placed on the different strategies to induce stretchability including extrinsic and intrinsic approaches. In the former case, engineering and patterning of the materials and devices are key elements while intrinsically stretchable systems rely on mechanically compliant materials such as elastomers and organic conjugated polymers. The result is a review article that provides a comprehensive summary of the progress to date in the field of stretchable solar cells from the nanoscale to macroscopic functional devices. The article is concluded by discussing the emerging trends and future developments.
• #### Ambiguities in solvation free energies from cluster-continuum quasichemical theory: lithium cation in protic and aprotic solvents.

(Physical chemistry chemical physics : PCCP, Royal Society of Chemistry (RSC), 2021-07-22) [Article]
Gibbs free energies for Li+ solvation in water, methanol, acetonitrile, DMSO, dimethylacetamide, dimethoxyethane, dimethylformamide, gamma-butyrolactone, pyridine, and sulfolane have been calculated using the cluster-continuum quasichemical theory. With n independent solvent molecules S initial state forming the "monomer" thermodynamic cycle, Li+ solvation free energies are found to be on average 14 kcal mol-1 more positive compared to those from the "cluster" thermodynamic cycle where the initial state is the cluster Sn. We ascribe the inconsistency between the "monomer" and "cluster" cycles mainly to the incorrectly predicted solvation free energies of solvent clusters Sn from the SMD and CPCM continuum solvation models, which is in line with the earlier study of Bryantsev et al., J. Phys. Chem. B, 2008, 112, 9709-9719. When experimental-based solvation free energies of individual solvent molecules and solvent clusters are employed, the "monomer" and "cluster" cycles result in identical numbers. The best overall agreement with experimental-based "bulk" scale lithium cation solvation free energies was obtained for the "monomer" scale, and we recommend this set of values. We expect that further progress in the field is possible if (i) consensus on the accuracy of experimental reference values is achieved; (ii) the most recent continuum solvation models are properly parameterized for all solute-solvent combinations and become widely accessible for testing.
• #### Bayesian seismic inversion: A fast sampling Langevin dynamics Markov chain Monte Carlo method

(Geophysical Journal International, Oxford University Press (OUP), 2021-07-22) [Article]
Summary In this study, we aim to solve the seismic inversion in the Bayesian framework by generating samples from the posterior distribution. This distribution incorporates the uncertainties in the seismic data, forward model, and prior information about the subsurface model parameters; thus, we obtain more information through sampling than through a point estimate (e.g., Maximum a Posteriori method). Based on the numerical cost of solving the forward problem and the dimensions of the subsurface model parameters and observed data, sampling with Markov chain Monte Carlo (MCMC) algorithms can be prohibitively expensive. Herein, we consider the promising Langevin dynamics MCMC algorithm. However, this algorithm has two central challenges: (1) the step size requires prior tuning to achieve optimal performance, and (2) the Metropolis-Hastings acceptance step is computationally demanding. We approach these challenges by proposing an adaptive step-size rule and considering the suppression of the Metropolis-Hastings acceptance step. We highlight the proposed method’s potential through several numerical examples and rigorously validate it via qualitative and quantitative evaluation of the sample quality based on the kernelized Stein discrepancy (KSD) and other MCMC diagnostics such as trace and autocorrelation function (ACF) plots. We conclude that, by suppressing the Metropolis-Hastings step, the proposed method provides fast sampling at efficient computational costs for large-scale seismic Bayesian inference; however, this inflates the second statistical moment (variance) due to asymptotic bias. Nevertheless, the proposed method reliably recovers important aspects of the posterior, including means, variances, skewness, and one- and-twodimensional marginals. With larger computational budget, exact MCMC methods (i.e., with a Metropolis-Hastings step) should be favored. The results thus obtained can be considered a feasibility study for promoting the approximate Langevin dynamics MCMC method for Bayesian seismic inversion on limited computational resources.
• #### Effects of Ecological Restoration Using Non-Native Mangrove Kandelia obovata to Replace Invasive Spartina alterniflora on Intertidal Macrobenthos Community in Maoyan Island (Zhejiang, China)

(Journal of Marine Science and Engineering, MDPI AG, 2021-07-22) [Article]
Spartina alterniflora has extensively invaded the coastline of China, including in Maoyan Island of Zhejiang Province. Ecological restoration has been conducted using non-native mangrove Kandelia obovata to replace S. alterniflora in an attempt to restore the impacted intertidal zones. To illustrate the ecological effectiveness of the restoration projects, macrobenthos communities were studied among different habitats within the restored areas, including one non-restored S. alterniflora marsh (SA) and three differently-aged restored K. obovata stands planted in 2003, 2009, and 2011 respectively (KF14, KF8, and KF6). Besides, one unvegetated mudflat (MF) adjacent to the non-restored S. alterniflora marsh and one K. obovata forest transplanted in 2006 (RKF) at a previously barren mudflat without invasion history of S. alterniflora were set as reference sites. A total of 69 species of macrobenthos were collected from Maoyan Island, and the species richness was dominated by gastropoda (23 species), polychaeta (18 species), and malacostraca (16 species). There was no significant difference between the six sites in terms of the abundance of macrobenthos, with the average values of abundance peaking in KF6 (734.7 ind m−2) and being lowest in RKF (341.3 ind m−2). The six sites had significant differences in terms of the biomass of macrobenthos. The KF8 site contained the highest average biomass (168.3 g m−2), whereas the MF site had the lowest (54.3 g m−2). The Shannon-Wiener diversity index and Pielou’s evenness index of the macrobenthos did not exhibit significant differences among the six sites. However, the results of permutational multivariate analysis of variance (PERMANOVA) revealed significant spatial differences in the macrobenthos community structure between the sites. Since KF14 shared a similar macrobenthos community structure with RKF, while representing a strikingly different structure from SA, we infer that ecological restoration using K. obovata can restore the macrobenthos community to resemble to a normally planted K. obovata forest about 15 years after restoration.
• #### Re-infection with different SARS-CoV-2 clade and prolonged viral shedding in a patient with hematopoietic stem cell transplantation: SARS-CoV-2 Re-infection with different clade.

(International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases, Elsevier BV, 2021-07-21) [Article]
Immunocompromised patients who have SARS-CoV-2 infection pose many clinical and public health challenges. We describe a patient with hematopoietic stem cell transplantation and lymphoma with protracted illness requiring 3 consecutive hospital admissions. Whole genome sequencing confirmed two different SARS-CoV-2 clades. Clinical management issues, and unanswered questions arising are discussed.
• #### CRISPR-Based Crop Improvements: A Way Forward to Achieve Zero Hunger

(Journal of Agricultural and Food Chemistry, American Chemical Society (ACS), 2021-07-21) [Article]
Zero hunger is one of the sustainable development goals set by the United Nations in 2015 to achieve global food security by 2030. The current harvest of crops is insufficient; feeding the world's population and meeting the goal of zero hunger by 2030 will require larger and more consistent crop production. Clustered regularly interspaced short palindromic repeats-associated protein (CRISPR-Cas) technology is widely used for the plant genome editing. In this review, we consider this technology as a potential tool for achieving zero hunger. We provide a comprehensive overview of CRISPR-Cas technology and its most important applications for food crops' improvement. We also conferred current and potential technological breakthroughs that will help in breeding future crops to end global hunger. The regulatory aspects of deploying this technology in commercial sectors, bioethics, and the production of transgene-free plants are also discussed. We hope that the CRISPR-Cas system will accelerate the breeding of improved crop cultivars compared with conventional breeding and pave the way toward the zero hunger goal.
• #### Observation of Negative Photoconductivity in Lead-Free Cs3Bi2Br9Perovskite Single Crystal

(ACS Photonics, American Chemical Society (ACS), 2021-07-21) [Article]
Light exposure usually causes an increase in photoconductivity in perovskite semiconductors. However, we report here light-induced negative photoconductivity, followed by slow dark self-recovery in a lead-free Cs3Bi2Br9 perovskite single crystal. Femtosecond transient reflection (fs-TR) spectroscopy studies further reveal hole self-trapping at the Vk center (Br2– dimer) in the midband states of this vacancy-ordered perovskite. Subsequently, these charged defect states (Vk) trap photogenerated charge carriers and produce an internal electrical field, which essentially opposes the externally applied field, leading to negative photoconductivity. A highly sensitive prototype photodetector was fabricated with figure of merits estimated as responsivity (6.42 mA/W), detectivity (2.51 × 1012 Jones), and current in a dark to light ratio (∼20). Our observation of this retrospective photocurrent in optically active perovskite materials can be applied for developing highly sensitive detectors.
• #### The impact of COVID-19 lockdowns on surface urban heat island changes and air-quality improvements across 21 major cities in the Middle East.

(Environmental pollution (Barking, Essex : 1987), Elsevier BV, 2021-07-20) [Article]
This study investigates changes in air quality conditions during the restricted COVID-19 lockdown period in 2020 across 21 metropolitan areas in the Middle East and how these relate to surface urban heat island (SUHI) characteristics. Based on satellite observations of atmospheric gases from Sentinel-5, results indicate significant reductions in the levels of atmospheric pollutants, particularly nitrogen dioxide (NO2), sulfur dioxide (SO2), and carbon monoxide (CO). Air quality improved significantly during the middle phases of the lockdown (April and May), especially in small metropolitan cities like Amman, Beirut, and Jeddah, while it was less significant in “mega” cities like Cairo, Tehran, and Istanbul. For example, the concentrations of NO2 in Amman, Beirut, and Jeddah decreased by −56.6%, −43.4%, and −32.3%, respectively, during April 2020, compared to April 2019. Rather, there was a small decrease in NO2 levels in megacities like Tehran (−0.9%) and Cairo (−3.1%). Notably, during the lockdown period, there was a decrease in the mean intensity of nighttime SUHI, while the mean intensity of daytime SUHI experienced either an increase or a slight decrease across these locations. Together with the Gulf metropolitans (e.g. Kuwait, Dubai, and Muscat), the megacities (e.g. Tehran, Ankara, and Istanbul) exhibited anomalous increases in the intensity of daytime SUHI, which may exceed 2 °C. Statistical relationships were established to explore the association between changes in the mean intensity and the hotspot area in each metropolitan location during the lockdown. The findings indicate that the mean intensity of SUHI and the spatial extension of hotspot areas within each metropolitan had a statistically significant negative relationship, with Pearson's r values generally exceeding - 0.55, especially for daytime SUHI. This negative dependency was evident for both daytime and nighttime SUHI during all months of the lockdown. Our findings demonstrate that the decrease in primary pollutant levels during the lockdown contributed to the decrease in the intensity of nighttime SUHIs in the Middle East, especially in April and May. Changes in the characteristics of SUHIs during the lockdown period should be interpreted in the context of long-term climate change, rather than just the consequence of restrictive measures. This is simply because short-term air quality improvements were insufficient to generate meaningful changes in the region's urban climate.
• #### CANITA: Faster Rates for Distributed Convex Optimization with Communication Compression

(arXiv, 2021-07-20) [Preprint]
Due to the high communication cost in distributed and federated learning, methods relying on compressed communication are becoming increasingly popular. Besides, the best theoretically and practically performing gradient-type methods invariably rely on some form of acceleration/momentum to reduce the number of communications (faster convergence), e.g., Nesterov's accelerated gradient descent (Nesterov, 2004) and Adam (Kingma and Ba, 2014). In order to combine the benefits of communication compression and convergence acceleration, we propose a \emph{compressed and accelerated} gradient method for distributed optimization, which we call CANITA. Our CANITA achieves the \emph{first accelerated rate} $O\bigg(\sqrt{\Big(1+\sqrt{\frac{\omega^3}{n}}\Big)\frac{L}{\epsilon}} + \omega\big(\frac{1}{\epsilon}\big)^{\frac{1}{3}}\bigg)$, which improves upon the state-of-the-art non-accelerated rate $O\left((1+\frac{\omega}{n})\frac{L}{\epsilon} + \frac{\omega^2+n}{\omega+n}\frac{1}{\epsilon}\right)$ of DIANA (Khaled et al., 2020b) for distributed general convex problems, where $\epsilon$ is the target error, $L$ is the smooth parameter of the objective, $n$ is the number of machines/devices, and $\omega$ is the compression parameter (larger $\omega$ means more compression can be applied, and no compression implies $\omega=0$). Our results show that as long as the number of devices $n$ is large (often true in distributed/federated learning), or the compression $\omega$ is not very high, CANITA achieves the faster convergence rate $O\Big(\sqrt{\frac{L}{\epsilon}}\Big)$, i.e., the number of communication rounds is $O\Big(\sqrt{\frac{L}{\epsilon}}\Big)$ (vs. $O\big(\frac{L}{\epsilon}\big)$ achieved by previous works). As a result, CANITA enjoys the advantages of both compression (compressed communication in each round) and acceleration (much fewer communication rounds).
• #### Fire in paradise: mesoscale simulation of wildfires

(ACM Transactions on Graphics, Association for Computing Machinery (ACM), 2021-07-19) [Article]
Resulting from changing climatic conditions, wildfires have become an existential threat across various countries around the world. The complex dynamics paired with their often rapid progression renders wildfires an often disastrous natural phenomenon that is difficult to predict and to counteract. In this paper we present a novel method for simulating wildfires with the goal to realistically capture the combustion process of individual trees and the resulting propagation of fires at the scale of forests. We rely on a state-of-the-art modeling approach for large-scale ecosystems that enables us to represent each plant as a detailed 3D geometric model. We introduce a novel mathematical formulation for the combustion process of plants - also considering effects such as heat transfer, char insulation, and mass loss - as well as for the propagation of fire through the entire ecosystem. Compared to other wildfire simulations which employ geometric representations of plants such as cones or cylinders, our detailed 3D tree models enable us to simulate the interplay of geometric variations of branching structures and the dynamics of fire and wood combustion. Our simulation runs at interactive rates and thereby provides a convenient way to explore different conditions that affect wildfires, ranging from terrain elevation profiles and ecosystem compositions to various measures against wildfires, such as cutting down trees as firebreaks, the application of fire retardant, or the simulation of rain.
• #### A Nonionic Alcohol Soluble Polymer Cathode Interlayer Enables Efficient Organic and Perovskite Solar Cells

(Chemistry of Materials, American Chemical Society (ACS), 2021-07-19) [Article]
The choice of interfacial materials and their properties play a critical role in determining solar cell performance and stability. For compatibility with roll-to-roll printing, it is desirable to develop stable cathode interface layers (CILs) that can be processed over the photoactive layer using orthogonal solvents. In this study, an n-type naphthalene diimide core and oligo (ethylene glycol) side-chain-based conjugated polymer is reported as a universal, efficient CIL for organic and perovskite photovoltaics. Besides good thermal stability and easy processing in alcohol/water, the new CIL is found to possess electron transport properties with an electrical conductivity of 2.3 × 10–6 S cm–1, enabling its use as a CIL with a film thickness of up to ∼35(±2) nm. Utilizing the new CIL, 16% power conversion efficiency (PCE) is achieved for organic solar cells (OSCs) based on the PM6-Y6 photoactive layer (8.9% PCE for no CIL and 15.1% with state-of-the-art CIL, PDINO), and perovskite solar cells from methylammonium lead iodide yielded a PCE of 17.6%. Compared to the reference devices, the new CIL reduced trap-assisted carrier recombination and increased the built-in potential by 80 mV, simultaneously enhancing all photovoltaic parameters. Moreover, new CIL based devices had better photostability with no burn-in losses.
• #### Seasonal M2 Internal Tides in the Arabian Sea

(Remote Sensing, MDPI AG, 2021-07-18) [Article]
Internal tides play a crucial role in ocean mixing. To explore the seasonal features of mode-1 M2 internal tides in the Arabian Sea, we analyzed their propagation and energy distribution using along-track sea-level anomaly data collected by satellite altimeters. We identified four primary source regions of internal tides: Abd al Kuri Island, the Carlsberg Ridge, the northeastern Arabian Sea, and the Maldive Islands. The baroclinic signals that originate from Abd al Kuri Island propagate meridionally, whereas those originating from the west coast of India propagate southwestward. The strength and energy flux of the internal tides in the Arabian Sea exhibit significant seasonal and spatial variability. The internal tides generated during winter are more energetic and can propagate further than those generated in summer. Doppler shifting and horizontal variations in stratification can explain the differences in the internal tides’ seasonal distributions.
• #### Flame flow field interaction in non-premixed CH4/H2 swirling flames

(International Journal of Hydrogen Energy, Elsevier BV, 2021-07-17) [Article]
Alternative fuels and stocks like biomass or chemical and refinery waste, may potentially be used in gas turbines and industrial applications after gasification. Thus, understanding the role of hydrogen in these fuels is critical to the broader aim of utilising alternative fuels for power generation. In this work, the interaction between the flame and the flow field was studied in a quarl-stabilised swirl non-premixed flame burning CH4 and H2–enriched CH4. Simultaneous high-speed OH-PLIF/PIV imaging at 5 kHz was carried out on these flames to explore the flame-flow interaction. The instantaneous flow fields in the CH4 or CH4+H2 flames showed a small scale vortical structure near the shear layers, which were not apparent in the time-averaged flow fields. Increasing H2% in the fuel jet was observed to dampen the velocity fluctuations. The fuel composition affected the spatial location of the reaction zone; in the CH4 flames, the axial position of the reaction zone is seen to track the relatively large-magnitude axial velocity fluctuations while remaining in locally low-speed regions of the flow. In contrast, in H2-enriched flames, where the flame is more robust, the reaction zone was able to survive longer, in terms of axial distance, in the vicinity of high swirling jet velocity, with less sensitivity to velocity fluctuations. With increasing the H2%, the reaction zone steadily leaves the IRZ towards the swirling jet flow and localised between its outer and inner vortices. This acts as a stabilisation factor where the internal vortices convect hot product towards the fresh mixture. Moreover, the flame curvatures, the vorticity and compressive strain fields interactions with the reaction zone are presented and discussed. This article outlines results that yield more in-depth insight into hydrogen-enriched hydrocarbon non-premixed swirling flames' combustion, which is essential to accelerate the fuel switching from hydrocarbons to hydrogen.