Macroscopic Nonuniformities Research Articles

(Invited) Examining Catalyst Layer Design Strategies for Improved Fuel Cell Performance and Durability

Introduction The commercialization of fuel cell products for a variety of applications including forklifts, buses, cars and other heavy duty transport is well underway. Recent studies suggest that over the next 10 years, the total cost of ownership of fuel cell vehicles will become more favourable than both battery electric vehicles and internal combustion engine commercial vehicles1,2. This is driven, in part, by a forecasted almost 50% decrease in the cost of fuel cell systems. While this achievement is mainly due to product volume increases and related economies of scale, further improvements of fuel cell technology can be achieved through next generation catalyst layer designs.Design approaches related to the individual component characteristics, optimization of the component interactions to obtain enhanced morphology and microstructure, as well as processing techniques are currently being developed. Strategies to lower catalyst loadings and move to higher current operation and power densities include incorporating high activity catalysts and improving proton conductivity and oxygen transport by integrating advanced ionomers. Advanced cathode catalysts3 such as Pt alloy (de-alloy), core-shell and shape controlled nanocrystals show more promise in achieving higher mass activity than conventional carbon supported platinum catalyst, although durability or technology maturity challenges remain. Alternatively, while performance gains through ohmic improvements can be achieved by utilizing high proton conductivity ionomers4, additional opportunities such as incorporating novel ionomer materials5, with increased oxygen permeability to allow for reduction of the oxygen thin film transport resistance are of interest.In addition to maximizing the catalyst and ionomer attributes, catalyst-ionomer interaction understanding at the molecular scale and its influence on catalyst layer morphology and transport properties is crucial. Earlier studies of ionomer thin films6, molecular dynamics modeling7 as well as catalyst layer characterization by XPS8, indicate that the ionomer morphology within confined thin films, as well as the resulting film properties, are not only dependent on the surface properties of the substrate / catalyst material and water content, but can be altered in situ as those parameters change. Therefore, the catalyst-ionomer interaction and morphology must be better understood and controlled. Although it can be optimized through design choices, processing parameters during catalyst layer fabrication can also be effective. For example, Figure 1a) shows changes to the cathode catalyst layer pore size distribution with four different processes, incorporating changes to the ink preparation, coating and drying parameters, while keeping the same catalyst layer composition and loading. The optimized process resulted in improved catalyst and ionomer dispersion, lower ionic resistivity and improved performance. Moreover, Figure 1b) shows a clear relationship of improved performance with larger catalyst layer pore size.In addition to microstructural characteristics, macroscopic non-uniformities of the catalyst layer, such as cracks and roughness variations, can also affect the performance and durability of the MEA and are an area of focus for quality control tools9. Recent four dimensional X-ray computed tomography studies have shown how pre-existing cracks within the catalyst layer grow with accelerated stress testing and can be detrimental to both the voltage and membrane durability of the MEA10-11. All in all, a multiscale research approach to gain understanding focussed on nanoscale, microscale and macroscale features of the catalyst layer for next generation designs is recommended and will be the focus of the presentation. Acknowledgement The authors would like to acknowledge Natural Sciences and Engineering Research Council of Canada (NSERC), Mitacs, Automotive Partnership Canada (APC), and the US Department of Energy (DOE) for funding various aspects of this work. References Deloitte and Ballard. (2020). Fueling the Future of Mobility: Hydrogen and fuel cell solutions for transportation. Volume 1 [White paper]. Deloitte China. https://www2.deloitte.com/content/dam/Deloitte/cn/Documents/finance/deloitte-cn-fueling-the-future-of-mobility-en-200101.pdf Hydrogen Council. (2020). Path to hydrogen competitiveness: A cost perspective. https://hydrogencouncil.com/wp-content/uploads/2020/01/Path-to-Hydrogen-Competitiveness_Full-Study-1.pdf Banham, D. and Ye, S., ACS Energy Letters 2(3): 629-638 (2017)Young, A.P. et al. 230th ECS Meeting in Honolulu, Hawaii (October 2-7, 2016)Rolfi et al. J. Power Sources, 396 (2018) 95 - 101Kusoglu, A. and Weber, A. Z. Chem. Rev. 2017, 117, 3, 987–1104Borges, D. D. et al. J. Phys. Chem. C 2015, 119, 2, 1201–1216Artyushkova, K. et al. J. Power Sources 284 (2015) 631-641Philips, A. et al. Int. J. Hydrog. Energy 2018, 43, 12, 6390-6399White, R. T. et al. J. Power Sources 350 (2017) 94-102Singh, Y. et al. J. Power Sources 412(2019) 224-237 Figure 1

Macroscopic Nonuniformities in Metal Grids Formed by Cracked Film Lithography Result in 19.3% Efficient Solar Cells

Cracked film lithography (CFL) is an emerging method for patterning transparent conductive metal grids. CFL can be vacuum- and Ag-free, and it forms more durable grids than nanowire approaches. In spite of CFL's promising transmittance/grid sheet resistance/wire spacing tradeoffs, previous solar cell demonstrations have had relatively low performance. This work introduces macroscopic nonuniformities in the grids to improve the short-circuit current density/fill factor tradeoff in small area Cu(In,Ga)Se2 cells. The performance of optimized baseline grids is matched by CFL grids with microscopic openings and macroscopic patterns, culminating in a 19.3%-efficient cell. Simulations show that uniform CFL grids are enhanced by patterning because it leads to better balance among shadowing, grid resistance and transparent conductive oxide resistance losses. Thin-film module efficiency calculations are performed to highlight the performance gains that metal grids can enable by eliminating the transparent conductive oxide losses and widening monoliths. Adding the patterned CFL grids demonstrated in this work to CIGS modules is predicted to reach 0.7% higher efficiency (absolute) than screen-printed grids.

Plastic flow instability: Chernov–Lüders bands and the Portevin—Le Chatelier effect

We have studied the regularities in the evolution of macroscopic nonuniformities in the plastic flow of metals in the form of the Chernov–Luders bands and the Portevin—Le Chatelier effect. The regularities in the evolution of strain inhomogeneity in these two cases have been established, and the kinetics of motion of the Chernov–Luders fronts and abrupt deformation in the Portevin–Le Chatelier effect has been analyzed. It has been shown that the Chernov–Luders fronts and Portevin–Le Chatelier jumpwise straining can be treated as macroscopic autowave processes of switching and excitation, respectively, in deformed media of various origins.

Estimation and Reduction of Macroscopic Nonuniformity in CNT Thin Film Resistors

Carbon nanotube (CNT) thin film-based devices are expected to have better uniformity and reproducibility compared with single-tube devices due to statistical averaging. In this work, thin film resistors (TFR) were prepared from CNT dispersions and their nonuniformity estimated by multipoint, two-terminal resistance (TTR) measurements. I-V measurements were done for determining the nature of metal-CNT contact. For estimation of contact resistance, transmission line measurements (TLM) were carried out. Reduction of CNT-TFR nonuniformity was achieved using di-methyl formamide (DMF). Scanning electron microscopy and electrical measurements imply that DMF improves debundling of single-walled nanotubes, reduces the resistance and improves homogeneity of subsequently prepared thin films.

Correlations Between Mapping Spectroscopic Ellipsometry Results and Solar Cell Performance for Evaluations of Nonuniformity in Thin-Film Silicon Photovoltaics

An understanding of the relationship between materials property and thin-film solar cell performance variations over large areas is of interest for evaluating the impact of macroscopic nonuniformities in scale-up from laboratory cells to production modules. In this study, we have spatially correlated the properties of the hydrogenated silicon (Si:H) i- and p-layers-as mapped over a 13 cm×13 cm substrate area-with device performance parameters from an array of a-Si:H based n-i-p dot cells. To evaluate materials and device nonuniformities, a 16 × 16 array of dot cells has been fabricated over the substrate area, and this same area has been mapped by spectroscopic ellipsometry (SE).Analysis of the SE data over the full area provides maps of i-layer thickness and band gap, p-layer thickness and band gap, and p-layer surface roughness thickness for the n-i-p solar cell structure. The mapped values adjacent to the devices have been correlated with photovoltaic (PV) device performance parameters. When sufficient nonuniformity exists, these correlations enable optimization based on specific values of the fundamental properties. Alternatively, if the optimum set of properties has been identified, the impact of deviations due to macroscopic uniformities can be evaluated.

Terraced spreading of nanofilms under a nonmonotonic disjoining pressure

A thin (∼nanometer) film of a viscous, essentially nonvolatile liquid spreads over a substrate controlled by the disjoining pressure Π(h) exerted by the two interfaces on one another. Such films are commonly used as hard disk lubricants in the magnetic recording industry. Macroscopic nonuniformities in the film caused by a perturbation of the uniformly spread state flow away and the film is “healed” in a time frame governed by the appropriate hydrodynamic equations. Lubrication theory may be used to derive a diffusion equation for the local film height h(x,t) as a function of position and time which shows that an effective height-dependent diffusion coefficient D(h)=−[h3/3μBξ(h)][dΠ(h)/dh] controls the spreading dynamics, where μB is the bulk liquid viscosity and ξ(h) is a function accounting for any variation of local viscosity near the substrate due to molecularity of the liquid. Such an approach is possible due to the very small ratio of the film height to the in-plane length scale of the disturbance. Provided the disjoining pressure is positive and monotonically decreasing with film thickness, the motion of the film is unexceptional, exhibiting the usual smooth profiles associated with diffusive flow with time. However, for nonmonotonic disjoining pressures, the film is experimentally observed to exhibit vertical terraces. These abrupt jumps in height do not disappear with time and they move slowly or are stationary. This phenomenon is investigated here. We demonstrate how a physically consistent “weak” solution of the diffusion equation can be constructed, where only positive values of the diffusion coefficient are sampled. The film heights at the jump discontinuity are determined by an equal area rule for the disjoining pressure. Numerical simulations for a realistic nonmonotonic disjoining pressure exhibit finite termination on the low-side and vertical terraces, thereby matching the behavior observed in experimental systems.

Repetitively pulsed atmospheric pressure discharge treatment of rough polymer surfaces: I. Humid air discharges

Plasmas generated at atmospheric pressure are used to functionalize the surfaces of polymers by creating new surface-resident chemical groups. The polymers used in textiles and biomedical applications often have non-planar surfaces whose functionalization requires penetration of plasma generated species into sometimes complex surface features. In this regard, the atmospheric pressure plasma treatment of a rough polypropylene surface was computationally investigated using a two-dimensional plasma hydrodynamics model integrated with a surface kinetics model. Repetitively pulsed discharges produced in a dielectric barrier–corona configuration in humid air were considered to affix O. Macroscopic non-uniformities in treatment result from the spatial variations in radical densities which depend on the polarity of the discharge. Microscopic non-uniformities arise due to the higher reactivity of plasma produced species, such as OH radicals, which are consumed before they can diffuse deeper into surface features. The consequences of applied voltage magnitude and polarity, and the relative humidity on discharge dynamics and radical generation leading to surface functionalization, are discussed.

High sensitivity radiochromic film dosimetry using an optical common‐mode rejection and a reflective‐mode flatbed color scanner

A novel method that can greatly improve the dosimetric sensitivity limit of a radiochromic film (RCF) through use of a set of color components, e.g., red and green, outputs from a RGB color scanner has been developed. RCFs are known to have microscopic and macroscopic nonuniformities, which come from the thickness variations in the film's active radiochromic layer and coating. These variations in the response make the optical signal-to-noise ratio lower, resulting in lower film sensitivity. To mitigate the effects of RCF nonuniform response, an optical common-mode rejection (CMR) was developed. The CMR compensates nonuniform response by creating a ratio of the two signals where the factors common to both numerator and denominator cancel out. The CMR scheme was applied to the mathematical operation of creating a ratio using two components, red and green outputs from a scanner. The two light component lights are neighboring wavebands about 100 nm apart and suffer a common fate, with the exception of wavelength-dependent events, having passed together along common attenuation paths. Two types of dose-response curves as a function of delivered dose ranging from 3.7 mGy to 8.1 Gy for 100 kV x-ray beams were obtained with the optical CMR scheme and the conventional analysis method using red component, respectively. In the range of 3.7 mGy to 81 mGy, the optical densities obtained with the optical CMR showed a good consistency among eight measured samples and an improved consistency with a linear fit within 1 standard deviation of each measured optical densities, while those with the conventional analysis exhibited a large discrepancy among eight samples and did not show a consistency with a linear fit.

Simple model of halogen depletion in XeCl discharge pumped lasers

The aim of this work is to study the development of the macroscopic non-uniformity of the electron density of high pressure discharge for excimer lasers and eventually its propagation because of the medium kinetics phenomena. This study is executed using a transverse mono-dimensional model, in which the plasma is represented by a set of resistances in parallel. The time variations of the electron density in each plasma element are obtained by solving a set of ordinary differential equations describing the plasma kinetics and external circuit. The use of the present model allows a good comprehension of the halogen depletion phenomena, which is the principal cause of laser operation premature termination and allows a simple study of a large-scale non-uniformity in preionization density and its effects on plasma electrical and chemical properties. The results obtained indicate clearly that about 50% of halogen is consumed at the end of the pulse.

Evolution of the temperature field and macrolocalization of strain under intermittent yield

Results of a study of special features of macrolocalization of strain under intermittent yield obtained with the help of thermal imaging and analysis of thermograms are presented. The stage of initiation of strain bands is discovered and detected. It is shown that the beginning of intermittent yield may be preceded by macroscopic nonuniformity of strain, which does not disturb the stability but activates the beginning of band formation. Static loading in the elastoplastic range is shown to be accompanied by a decrease in the temperature. Specific features of the evolution of temperature fields are determined for different serration patterns. It is shown that local heating due to the formation of one strain band can activate the formation of the next band. It is inferred that the method of thermal imaging is very effective for studying special features of plastic strain.

Bioconvective dynamics: dependence on organism behaviour.

Bioconvection occurs when a macroscopic nonuniformity of the concentration of microbial populations is generated and maintained by the directional swimming of the organisms. This study investigated the properties of the patterns near the onset of the instability and later during its evolution into a fully nonlinear convection regime. In suspensions of the bacteria Bacillus subtilis, which tend to swim upwards in a gradient of oxygen concentration that they create by consumption, we discovered that the dominant wavelength at the onset of the instability is determined primarily by the cell density and is influenced only weakly by the fluid depth. This observation contrasts strongly with previous observations on the gravitactic alga Chlamydomonas nivalis, in which the opposite dependence was found. Considerable differences were also found in the long-term evolution of the convection patterns. These results demonstrate the existence of readily distinguishable types of bioconvection systems, even at early stages of the instability. The observed differences are clearly and causally correlated with disparate reasons for upward swimming by these micro-organisms, leading to different geometric distributions of the density of the suspension.

Effects of bandgap, lifetime, and other nonuniformities on diode laser thresholds and slope efficiencies

We examine the effect of various types of spatial nonuniformities on the performance of semiconductor diode lasers. Light versus current (L-I) characteristics are modeled by solving the coupled rate equations for the carrier density, photon density, carrier temperature, and lattice temperature. In cases of macroscopic nonuniformities, that is, when the fluctuation correlation length (/spl Lambda//sub F/) is longer than the carrier diffusion length (L/sub D/), the cavity is divided into a large number of segments, and separate rate equations are solved for each segment. Type-II antimonide quantum wells are used as the model diode laser system. Macroscopic nonuniformities of the nonradiative lifetime are shown to have little effect on the slope efficiency at high injection levels, but produce a signature softening of the threshold abruptness if the nonuniformity is large enough. This results from the nonuniformity of the carrier concentration and consequent mixture of gain and loss regions. For the case of microscopic bandgap fluctuations (L/sub D//spl Gt//spl Lambda//sub F/), inhomogeneous broadening of the gain spectrum leads to an increase of the threshold current density, which is accompanied by only a small reduction of the efficiency. However, in the complementary limit of macroscopic bandgap fluctuations (L/sub D/</spl Lambda//sub F/), the regions with distinct bandgaps become isolated, and photons emitted from the regions with larger bandgaps tend to be absorbed by the regions with smaller bandgaps. For an example with a Gaussian bandgap distribution of width /spl sigma/=30 meV, the nonuniformities have the effect of increasing the threshold current by a factor of 2.6, while reducing the slope efficiency by a factor of 2. We also consider lateral nonuniformities due to the temporal instability of the lasing mode, which tend to have a rather modest effect on the threshold current, slope efficiency, and even on the maximum output power despite the severe degradation of the beam quality.

Theoretical studies on microscopic and macroscopic nonuniformities in electric-discharge-excited XeCl lasers

The satbilit r of the pumping discharge for XeCI (Ne/Xe/HCI) lasers is investigated by using different one-dimensional models, considering both bulk discharge instabilities and filamentary ones. It is found that the bulk stability in a XeCl laser discharge can he improved, under certain conditions, due to the vihrational excitation of HCl molecules: this effect is studied by using different models for the HCl vihrational kinetics. It is established that the very short instability time found in experiments cannot he explained as a result of hulk instability of the discharge. On the other hand, a satisfactory comparison with experiments can he obtained by adding to the model a simplified microinstability which develops in the discharge.

On the Method of Determination of the Noise Levels of Black-and-White Photographic Materials

According to the recent studies on the noise characteristics of photographic materials, two kinds of noise can be defined: the macronoise, trD (mac), and the micronoise, aD (mic). The macronoise is the equivalent of the RMS-granularity of the photographic material, and is believed to include the contribution of the macroscopic non-uniformity of the emulsion layer. On the other hand, the micronoise is thought to be determined by the ultimate limits of the statistical fluctuation in the process of photon recoding on the emulsion layer. In this work, the above mentioned hypothesis was verified experimentally and also by computer simulation. The results showed that σD (mac) is generally greater than σD (mic), and the standard correlation coefficient, r, is greater than zero. This is obviously not due to the overlapping of the successive sampling spots. When the sampling rate was reduced from 50 s-l to 11 s-1 which is equivalent to increase the distance between the successive sampling spots from 52.4 microns to 238 microns, the macronoise remained unchanged, whereas the micronoise became close or equal to the macronoise. The computer simulation showed that the micronoise approaches the macronoise as the distance between the successive sampling spots increases. The authors suggested that the macronoise, i.e. the RMS-granularity, can be used as a figure of merit in the evaluation of photographic materials, whereas the micronoise may be more meaningful when one’s interest is on the influence of AgX emulsion on the image quality. The r-value may serve as a measure of the large-scale non-uniformity of a photographic layer.

Effect of surfactant concentration on the critical thicknesses of liquid films

It is shown that an increase in surfactant concentration at first causes the critical thicknesses of rupture on microscopic films to decrease but gradually a nearly constant value is reached. With films of aqueous solutions of fatty acids (valeric, caproic, caprylic and capric acids), this dependence correlates well with the effect of the surfactant concentration on the damping of capillary waves. With surfactants of the detergent type (OPE-7, OPE-20) in addition to the change of the critical thickness, the transition from rupture to formation of black spots is described, as the surfactant concentration exceedsC bl. The remarkable fact in the latter case is the independence of the critical thickness of the final state, be it rupture or formation of first or second black films. On the basis of the experimental data the assumption is put forward that the critical thickness of rupture or the critical thickness of formation of black spots is substantially affected by macroscopic non-uniformities in the film thickness. Thus the conclusion is reached that the critical thickness of an ideally plane parallel film which is the object of the theory, must be obtained by extrapolation of the measured value toward extremely small radii.

Macroscopic Quasilinear Theory of High-Frequency Radiation in a Cold Plasma

The nonlinear fluid equations for a cold electron-proton plasma are averaged over short time and length scales to obtain high-frequency “radiation” terms as well as the usual macroscopic electromagnetic fields. The result is a self-consistent, quasilinear theory in which the radiation terms can be calculated directly from the plasma equations and Maxwell's equations without reference to an independent radiative transfer equation. Mechanisms for the nonlinear saturation of the absorption of resonant modes of the radiation field can be included in the formulation by introducing a small, finite temperature. When the plasma is steady, uniform, and nonstreaming, with no applied fields, it is possible to derive a radiative intensity function which includes the collective effects of the plasma and obeys an integral form of the classical radiative transfer equation. However, in general there are averaged, high-frequency terms which have no counterpart in the usual radiation theory. These new radiation terms come from waves in the presence of applied magnetic fields or macroscopic nonuniformities in the plasma.

High-Frequency Resistance of Photoconducting Films

Measurements of high-frequency (hf) resistance and photoconductivity have been performed on PbTe infrared photocells, and interpreted in the light of a recent analysis of hf resistance of thin films. Hf data show the presence of macroscopic nonuniformity in the photoconducting layers which decreases upon illumination. Since this technique will not detect the presence of inter crystallite barriers, their action in photoconductivity can neither be proved nor disproved by this method.