Electric aircraft have generated increased interest following the recent success of electric passenger vehicles. Over 4 million passenger electric vehicles have been sold,(1) and there have been numerous announcements regarding the electrification of SUVs, pick-up trucks, and other light commercial vehicles, which represent the majority of the passenger automotive market.(2,3) However, while electrification of ground vehicles is well underway, electrification of aircraft is still in its infancy. Conventional aircraft engines emit greenhouse gases such as carbon dioxide, water vapor, nitrous oxides, sulfates, and soot.(4) They also emit contrails, which could cause up to 50% of aviation-derived radiative forcing.(5) In addition, electrification of aircraft opens new architectures for improving efficiency such as distributed electric propulsion, which can increase the lift–drag ratio and decrease the weight of the propulsion system,(6,7) and boundary layer ingestion, which can increase propulsive efficiency by 8–10%.(8,9) Alongside, there is great interest in electric vertical takeoff and landing (eVTOL) aircraft for urban air mobility.(10−12) In a recent Viewpoint, we identified the challenging battery requirements for eVTOL aircraft, reiterating the obvious importance of specific energy (defined as the energy available per unit mass) and identifying the importance of power limitations and thermal management requirements during takeoff and landing.(13) While eVTOLs represent a new market for electric aircraft, electrifying existing commercial aircraft is an important step in moving the transportation sector toward net-zero emissions.(14) Efforts are underway toward the introduction and development of electric and hybrid electric commercial aircraft.(15−17) Norway has announced its intention to electrify its entire fleet of aircraft in the near future.(18) Further, the world’s largest seaplane operator, Harbor Air, announced their intention to electrify their fleet.(19) Numerous technological challenges remain in the electrification of aircraft, one of the primary uncertainties being the performance metrics required of batteries to do so. Many analyses have presented a comprehensive system-level perspective on transport-sized hybrid and electric aircraft and have identified subsystem component targets for the systems that they analyze.(20−23) Others have presented analyses on greenhouse gas emission reductions, resulting from electrification of aircraft.(24,25) Additionally, some small electric aircraft exist in various stages of the development process.(11,26,27) These analyses tend to be specific to certain classes of transport aircraft, and sometimes specific aircraft, rather than addressing the commercial aviation market as a whole. In this Viewpoint, we aim to identify a comprehensive set of performance metrics required for next-generation batteries to electrify commercial aircraft. We divide commercial aircraft into three categories: regional, narrow-body, and wide-body. Regional aircraft typically fly short missions, about 500 nautical miles (nmi) and carry low passenger loads (30–75), while wide-body aircraft carry high passenger loads (200–400) and fly much longer missions (>2000 nmi). Narrow-body aircraft fall in between, carrying medium passenger loads and flying ranges of ∼1000 nmi. We find that the major factor in determining the specific energy required of aircraft is the range that the class of aircraft typically flies, meaning that smaller, short-range aircraft will require less demanding battery performance metrics than larger, longer-range aircraft. We find that only next-generation chemistries, like Li–air or Li–CFx, may be able to meet some of the requirements needed for electric commercial aircraft to achieve the range and payloads required for adoption. In the course of a mission, an aircraft takes off from the ground, climbs to its cruising altitude, cruises to its destination, descends to near ground level, and then lands.(28) All aircraft are mandated to maintain an emergency reserve energy for contingencies such as diversions or aborted landings. The FAA (Federal Aviation Administration) requires that commercial aircraft be able to abort a landing, climb to normal cruising altitude, fly to the most distant alternate airport (here assumed to be 200 nmi), and loiter for 45 min at normal cruise fuel consumption.(29) As an alternative to the extant FAA commercial reserve, a proposed approach is to house the emergency reserve by maintaining an additional 30% battery state of charge (SoC),(20) and we use the 30% SoC reserve in our analyses. To calculate energy and power requirements in flight, we first calculate thrust. Four forces act on an aircraft in flight: thrust (force generated by the propulsion system), drag (aerodynamic force opposite to velocity), weight (gravitational force), and lift (aerodynamic force normal to velocity).(30) Neglecting acceleration, thrust can be calculated by solving the equations for each of these forces, which are a function of the geometry and operating conditions of the aircraft, including instantaneous velocity relative to the surrounding air (V), the zero lift drag coefficient (CD0), propulsive efficiency (ηprop), mechanical efficiency (ηmech), wing area (S), the aspect ratio (the ratio of the square of the wingspan to the wing area), and climb or descent angle (γ). To calculate power at any point during flight, we neglect acceleration, and thrust is multiplied by velocity, resulting in eq 1.(31) We calculate energy by integrating instantaneous power over the duration of the flight.(1) Assessing the performance of potential electric aircraft is complicated by the considerable variation of the parameters in eq 1. We calculate density (ρ) and velocity (V) at each point during the flight. We estimate the remaining parameters, namely, the zero-lift drag coefficient (CD0), propulsive efficiency(ηprop), mechanical efficiency (ηmech), wing area (S), and aspect ratio, using distributions based on current commercial aircraft. K is a function of aircraft geometry and is discussed in more detail in the SI. To estimate mass allocated to payload, energy storage, and aircraft systems, we use the empty mass fraction (ewf), the fraction of aircraft mass with no payload or energy storage to the total takeoff mass of the aircraft. The total mass of the aircraft MTO is given by eq 2, where Mpax is the payload mass, Se is specific energy, and P is the instantaneous power. Detailed descriptions of the calculations of each parameter are available in the SI.(2) Figure 1. Histograms of specific energy for regional, narrow-body, and wide-body aircraft, illustrating the uncertainty stemming from aircraft design parameters. Larger (and longer-range) aircraft require a higher specific energy than do smaller (and shorter-range) aircraft. Figure 2 shows the distributions of parameters from historical aircraft showing minimum, maximum, and mean values for each parameter and class of aircraft. These parameters were gathered from current U.S. commercial aircraft (a specific list of aircraft can be found in the SI) and were used in lieu of extensive trade studies to estimate the parameters of potential electric aircraft. Figure 2. Parameters used to estimate specific energy for various classes of commercial aircraft. The minimum, maximum, and mean of each parameter and aircraft are shown on each plot. These parameters are used to estimate the power and energy of a prospective electric aircraft of each size. The data for these parameters are in the SI. To find the distribution of specific energy resulting from the uncertain parameters in eq 1, we performed Monte Carlo simulations with predefined missions for each class of aircraft. The parameters for the simulations were sampled from the triangular distributions shown in (Figure 2). The range for regional, narrow-body, and wide-body aircraft was set at 350, 500, and 1000 nmi, respectively, and the number of passengers was set to 30, 150, and 300, while the mass for each segment was 50 000, 100 000, and 250,000 kg, chosen based on previous literature and current aircraft of each class. Then, 100 000 iterations were run for each class of aircraft. Results are shown in Figure 1. The data shown in Figure 1 have means for each segment of aircraft of ∼600, 820, and 1280 Wh/kg-pack, with standard deviations of 61, 81, and 105 Wh/kg-pack, respectively. Gnadt et. al estimated a required specific energy for a narrow-body aircraft of 800 Wh/kg-pack, which agrees well with our mean for that class of 820 Wh/kg.(20) The trend of increasing specific energy with aircraft size is not primarily due to the larger size of these aircraft but rather to the longer-range use cases for which they are typically employed. When the range for the narrow-body and wide-body cases is held constant and the same analysis is run, the mean specific energy for the wide-body is ∼1280 Wh/kg-pack, and that for the narrow-body is ∼1490 Wh/kg-pack. The resulting histograms can be seen in the SI. It should be noted that satisfying these predefined mission requirements does not guarantee that an aircraft is commercially feasible. Small aircraft, such as regional and some narrow-body aircraft, often have cruising ranges that are much lower than their maximum range. However, large aircraft often use a much larger fraction of their maximum range in a typical flight. For example, the Airbus A319, a small narrow-body aircraft, is most likely to fly a range of around 161 nmi in cruise, and the distribution of its flights is skewed toward the lower end of its range. On the other hand, a wide-body aircraft, such as the Boeing 777, nearly always flies toward the high end of its range, with a mode cruise length of 2615 nmi.(32) Therefore, not only are the battery requirements for regional aircraft more feasible than narrow- or wide-body aircraft but the baseline cases for regional aircraft are also more practical than those for narrow- and wide-body aircraft. To compare potential electric aircraft and conventional aircraft at various battery-specific energies and empty weight fractions, we show the percentage of mean range and passenger nautical miles (pnmi) for each class of aircraft in Figure 3. The regional aircraft is able to achieve the current mean pnmi at around 1400 Wh/kg-pack, with an empty weight fraction of 0.35, while the narrow-body and wide-body aircraft are not able to achieve the current mean pnmi at any specific energy considered in this analysis. The most demanding battery requirements occur in the wide-body case, where even in the most optimistic case presented in this paper only 24% of the current pnmi and 20% of the current range are achieved. Figure 3. Range and passenger miles achieved by electric regional, narrow-body, and wide-body aircraft shown as a fraction of the current average range in (a) and passenger miles in (b) for the respective categories. We observe that for regional aircraft the current average range is achieved at a pack-level specific energy of about 2000 Wh/kg and current average passenger miles at about 1400 Wh/kg. The threshold for a feasible all-electric regional aircraft is about 500 Wh/kg, achieving about 25% of the current average range. A similar threshold is about 800 and 1700 Wh/kg for narrow-body and wide-body, respectively. However, only 12 and 16% of the current average range is achieved at threshold-specific energies for the narrow- and wide-body aircraft, respectively. At the highest pack-level specific energy considered of 2000 Wh/kg, electric wide-body aircraft can achieve only 19 and 16% of the current average range and passenger miles. On the other hand, at 2000 Wh/kg, regional aircraft achieve a much higher range and passenger miles than the current average. As mentioned above, the scaling effect is not primarily due to the larger size of the aircraft but rather due to the increased range. To illustrate this effect, consider the power profiles of each of the classes of aircraft for representative ranges flown by each (Figure 4). While the size of the aircraft results in the higher power at each point, the energy required to fly the ranges flown by aircraft (the area beneath the curve) increases as a result of both the increased power and the increased range. Figure 4. (a) Aircraft power profiles, along with conditions of flight in each segment. This figure illustrates the scaling challenges inherent in electric flight; as MTOM increases, the typical use case range also increases, causing a massive increase in the total energy needed. (b) Comparison of the power demand to energy (total energy over the trip) ratio throughout the mission. Having identified the energy and power requirements, we discuss the possible battery chemistries and materials needed to achieve the previously identified targets. The specific energy of current generation Li-ion batteries is about 250 Wh/kg-cell, which has steadily increased by about 5% over the past decade.(33) The projected maximum specific energy for future Li-ion batteries is around 400–500 Wh/kg-cell(33) with lithium metal anodes and high-voltage and high specific capacity cathodes. Accounting for packing burden, this is likely insufficient for regional aircraft, the least demanding among the three categories of aircraft considered. The maximum specific energy of a Li–S system is about 500 Wh/kg-pack,(34) which reaches the minimum threshold for regional aircraft, but does not allow for improvements beyond the baseline capability and therefore may not be practical for aircraft development. One of the most promising chemistries is Li–O2, where the projected maximum pack specific energy could potentially meet some of the targets estimated previously for narrow-body and regional aircraft and allow for improvements beyond the baseline capability in the case of regional aircraft. While Li–O2 battery systems have one of the highest specific energies among rechargeable electrochemical batteries,(34) comparable high specific energy primary batteries have been investigated for applications in space exploration.(35) At an operating temperature of about 20 °C, Li/SO2, Li/SOCl2, Li/FeS2, and Li/MnO2 systems provide specific energies in the range of 350–420 and 330–350 Wh/kg-cell at low and medium discharge rates, respectively. Li/CFx batteries provide up to 730 Wh/kg-cell at medium discharge rates.(35) It remains to be seen if these primary battery chemistries could be made rechargeable and meet the power and specific energy requirements for electric aircraft. In this study, we limit our analysis only to rechargeable batteries for aircraft propulsion, and we intend to explore the performance envelope of these primary batteries in a future study. To estimate the cell and pack-level specific energy of Li–O2 systems, we used electrochemical Li–air cell and pack models following the work of Gallagher et al.(34) Both open and closed Li–O2 systems were considered for this analysis. We chose to focus on an open system, which does not carry oxygen on-board, as opposed to a closed system wherein the O2 is contained in a pressure vessel because the open system tends to maximize specific energy, although oxygen intake over the course of a discharge will cause the mass of the system to rise over the duration of a flight, resulting in reduced effective specific energy.(34) In such a system, the battery is accompanied by a compressor to account for the changes in atmospheric pressure experienced by an aircraft in flight. The mass of the compressor and the energy that it consumes are accounted for in the model. Using the electrochemical and pack design model, we construct a Ragone plot showing the relationship between the pack specific energy and specific power, seen in Figure 5b. Li–O2 is capable of providing the specific energy required for regional and many narrow-body flights; however in some cases, the high power requirements of takeoff will limit the specific energy of the battery. Figure 5. (a) Pack-level specific energy required for various aircraft configurations as a function of power/energy ratio and specific energy achieved by a Li–air battery as a function of power/energy ratio. While for low values of the power–mass ratio (C) all three aircraft could be flown using Li–air batteries, only for regional is a meaningful percentage of current passenger nautical miles achieved. (b) Pack specific energy of Li–air open systems for different pack-level energy and power metrics. As the specific energy tends to zero, it implies that the pack power to pack energy ratio is not achievable. Figure 5a shows the specific energy as a function of power–energy ratio for the Li–O2 system. It also shows the specific energy required as a function of the peak power to energy (in W/Wh) ratio for each type of aircraft for various values of power–mass ratio (in W/kg). The intersection of these curves represents a feasible operating point for a prospective aircraft, where the battery power and energy meet the aircraft’s requirements. For all three categories of aircraft, only the lowest power–mass ratio (150) yields a feasible specific energy. For regional aircraft, the specific energy is around 900 Wh/kg-pack, meaning that a lithium air battery could achieve around 60% of the current passenger nautical miles according to Figure 3. For narrow-body aircraft, the maximum specific energy achieved is around 600 Wh/kg, achieving around 10% of the current passenger nautical miles, and for wide-body, no meaningful aircraft can be built at the specific energy identified. Therefore, Li–O2 provides a feasible route forward only for small regional aircraft. Fully electric aircraft powered by batteries face a number of challenges moving forward. The specific energy of even the most optimistic future batteries enables only small regional aircraft, while larger narrow-body or wide-body aircraft remain outside of the feasibility limits of known electrochemical rechargeable battery systems. Additionally, the achievable small electric aircraft would be heavier than conventional aircraft for comparable performance metrics. It should be noted that this analysis does not consider the energy savings through potential improvements in aircraft design such as boundary layer ingestion and distributed propulsion. Although these technologies could be achieved in conventional aircraft, electrification provides the most feasible avenue for their introduction.(7) While a fully electric aircraft requires significant innovation in battery and aircraft design, a hybrid aircraft(36) could be a potential pathway to help address some of the challenges while increasing aircraft efficiency. In any case, a fully, or at least a more electric, (hybrid) aircraft presents an opportunity to lower the climate impact of commercial aviation. While the exact extent of emission savings depends on external factors such as electricity mix, electrifying aircraft would eliminate aircraft-induced cloudiness. In the near term, hybrid electric and small fully electric aircraft can help mitigate these climatic effects of aviation. In the long term, significant technological improvements in both battery and aircraft technology will aid the further adoption of small electric and larger more electric aircraft. The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsenergylett.9b02574. Details of the aircraft performance model, the PNMi optimization, and a comparison of narrow-body and regional aircraft specific energies at constant range (PDF) Aircraft parameters (TXT) Details of the aircraft performance model, the PNMi optimization, and a comparison of narrow-body and regional aircraft specific energies at constant range (PDF) Aircraft parameters (TXT) Views expressed in this Viewpoint are those of the authors and not necessarily the views of the ACS. The authors declare the following competing financial interest(s): Venkat Viswanathan is a consultant for Pratt & Whitney. He is a technical consultant, owns stock options, and is a member of the Advisory Board at Zunum Aero. Viswanathan's group received research funding from Airbus A3 and Aurora Flight Sciences. Electronic Supporting Information files are available without a subscription to ACS Web Editions. The American Chemical Society holds a copyright ownership interest in any copyrightable Supporting Information. Files available from the ACS website may be downloaded for personal use only. Users are not otherwise permitted to reproduce, republish, redistribute, or sell any Supporting Information from the ACS website, either in whole or in part, in either machine-readable form or any other form without permission from the American Chemical Society. For permission to reproduce, republish and redistribute this material, requesters must process their own requests via the RightsLink permission system. Information about how to use the RightsLink permission system can be found at http://pubs.acs.org/page/copyright/permissions.html. A.B. and V.V. gratefully acknowledge support of part of this work from Convergent Aeronautics Solutions (CAS) project under the NASA Aeronautics Research Mission Directorate. This article references 36 other publications.

Views expressed in this Energy Focus are those of the authors and not necessarily the views of the ACS. The authors declare no competing financial interest. This article has not yet been cited by other publications.

Over the last two decades, halide perovskites (HPs) have been identified as one of the most promising materials in photovoltaic and light-emitting devices.(1,2) This has led to major breakthroughs in materials science(3,4) but has also brought about a general misunderstanding and misuse of the term “perovskite”. In this Viewpoint, we will address the definition of a perovskite, with a main focus on the subgroup of perovskites that consist of heavier halides (Cl, Br, and I), both fully inorganic and hybrid organic–inorganic ones, as well as the many variants that can be found within this subgroup. In doing so, we will clarify what defines, in our opinion, a halide perovskite and discuss the commonly misnamed nonperovskite crystal structures. A perovskite crystal lattice is defined as a network of corner-sharing BX6 octahedra that crystallize with a general ABX3 (or equivalent) stoichiometry, as is shown in Figures 1 and 2a.(5) Deviations from this ABX3 stoichiometry can be obtained when the A and B cation sites become partially or fully vacant (vacancy-ordered perovskites), or when they are replaced by a combination of other cations (with different valences but with an overall neutral charge balance), forming double or quadruple perovskites. The aristotype (the highest symmetric structure of a group of crystal structures)(6) perovskite belongs to the Pm3̅m cubic space group, with SrTiO3 (tausonite) usually considered the prime example. The vast majority of perovskites though have a reduced symmetry (hettotypes, i.e. structures that are similar to the aristotype, but with a lower symmetry)(6) due to lattice distortions, distorted octahedra, ordered cations, vacancies, or the presence of organic cations or inorganic clusters. Figure 1. Standard depiction of the aristotype cubic perovskite. Shown in display styles evidencing either all the atoms (left) or only the BX6 octahedral network and A atoms (right). Figure 2. Overview of different halide perovskites. (a) Standard ABX3 cubic halide perovskites. (b) Antiperovskites, with A being a monovalent metal (like Li+ or Ag+), X a halide, and Y a chalcogenide. (c) Common orthorombic and tetragonal disordered perovskites, arising from the tilting of the octahedra. (d) Vacant BX3 perovskites, like AlF3. (e) Ordered perovskites, where two M(II) metals are replaced by a M(I) and M(III) metal. (f) Vacancy-ordered perovskites, where a part of the B-site cations are replaced with a M(III) or M(IV) and vacancies. The term “perovskite” was first coined by Gustav Rose in 1839 for the CaTiO3 mineral, who named it after the Russian nobleman and mineralogist Count Lev Alekseyevich von Perovski. Later, in 1926, it was first used as a general term for the crystal structure group by Victor Goldschmidt.(3) In nature, perovskites are primarily found as oxides, with the majority being silicates (such as bridgmanite minerals),(7) but they also exist as fluorides, chlorides, hydroxides, arsenides, and intermetallic compounds.(5) While the number of natural perovskite minerals is limited, synthetic perovskites span across the whole periodic table in terms of elemental composition, and they can exist in many complex formulas such as metallic perovskites (e.g., MgCNi3),(8) hybrid organic–inorganic perovskites (e.g., [CH3NH3]PbI3 and [CH3NH3]Mn(HCOO)3),(9) metal-free perovskites (e.g., [DABCO][NH4]I3 with MDABCO = N-methyl-N′-diazabicyclo[2.2.2]octonium),(10) and even noble gas-based perovskites (KSrXeNaO6).(11) This indicates that the perovskite is perhaps the most adaptable type of crystal lattice. Inorganic Metal Halide Perovskites. The subgroup of perovskites (and ternary metal halides in general) that will be discussed in this Viewpoint have halides as their X anions. In the case of standard perovskites (that is, those with an ABX3 stoichiometry) the B cations are divalent (like Pb2+, or Sn2+) and the A cations are large monovalent alkali metals (most commonly cesium) or small organic cations like methylammonium (MA) or formamidinium (FA). Because of the single charge of the anions in HPs, the A and B cations are limited to M(I) and M(II), respectively (not considering ordered or vacant perovskites). While this Viewpoint will mainly concentrate on heavy metal-based halides, which generally have bandgaps in the visible spectrum and NIR (therefore they are of more interest for use in photovoltaic and light emitting devices), there is a vast group of first row transition metal fluoride perovskites, which can be interesting for lithium intercalation or as a host material for up- or down-conversion emission.(12−14) One of these fluoride perovskites, KMgF3 (parascandolaite),(15) is also one of the few HPs that can be found in nature, and it crystallizes in a perfect cubic perovskite structure (Pm3̅m). Therefore, KMgF3 is a more suitable aristotype for HPs than SrTiO3.(5) Fluoride perovskites can also crystallize in a so-called “inverse perovskite” (which is not to be confused with antiperovskite), in which the A cation has a higher oxidation state than the B cation, like in the case of BaLiF3.(16) This occurs only when the A cations (e.g., Ba2+ and Sr2+) are very large and the B cations (e.g., Li+) are very small. Moreover, these small B cations can only form perovskites with F– and H– ions. Finally, halide antiperovskites also exist, in which the A and B sites are occupied by anions (halide and chalcogenide respectively) and the X sites are occupied by a monovalent cation (as is shown in Figure 2b).(17,18) Halide antiperovskites like Li3OBr and Ag3SI will not be discussed in this Viewpoint, as the halide is not part of the BX framework, but they certainly do classify as perovskites. Fully inorganic, heavy metal HPs do not crystallize in the pure cubic perovskite crystal lattice. Rather, they exhibit a distorted lattice, resulting in hettotype structures. These distortions occur as a result of the size difference between the A, B, and X cations. Consequently, the individual octahedra tilt, which generally results in either orthorhombic (commonly referred to as the “GdFeO3-type”)(19) or tetragonal crystal structures (Figure 2c).(19) Although these distortions are responsible for the lower crystal structure symmetry, the overall perovskite framework is preserved; therefore, these materials are classified as hettotypes of the perovskite group. In fact, even the original perovskite (CaTiO3) exhibits a small distortion and does not actually crystallize in a “perfect” cubic perovskite structure.(5) Several variants of the perovskite structure, and thus of ABX3 stoichiometry, can be obtained by partially or fully replacing (ordering) or even removing A and/or B cations. Although the latter (i.e., fully removing A cations) is more frequently found in metal pnictides (e.g., skutterudites, CoAs3),(20) the A cation site in HPs can be fully vacant, resulting in the so-called “A-site vacant BX3” crystal structure, as is shown in Figure 2d. This type of structure is observed only in fluorides, like AlF3, FeF3 CoF3, and MnF3.(5) Another variant of the ABX3 perovskite is in the form of ordered perovskites (Figure 2e). Here, the B cations are heterovalently replaced by a combination of two (or more) cations that are located at specific crystallographic sites.(21) In the case of HPs, this results in an A2BCX6 elpasolite structure (which is also sometimes termed “A2BB′X6”), with its respective aristotype being the K2NaAlF6 mineral. Because of the double occupancy of the B-site cation, these perovskites are often simply called “double perovskites”, and they have been frequently studied as lead-free photovoltaic materials, albeit with limited success,(21) or for broadband light emission.(22) For heavy metal halides, the double perovskite subgroup includes materials such as Cs2AgInCl6, Cs2AgBiBr6, and MA2AgSbI6, with a crystal structure that belongs to the Fm3̅m space group.(21) It is important to note that these are not the same as mixed A, B, or X site perovskites (for example, (Cs:MA:FA)(Pb:Sn)(Br:I)3), in which the mixed cations occupying the respective A/B/X sites have the same valence, meaning that they do not occupy specific crystallographic sites. Furthermore, ordered perovskites have fixed stoichiometries, which is not the case for mixed perovskites. A variant of the double perovskite can also be obtained with a single metal in both the (I) and (III) oxidation states. These double perovskites include materials such as CsTlX3 and CsAuX3 (even though the latter has strongly distorted octahedra).(23,24) Finally, a small group of fluoride ordered perovskites crystallize in the A3BX6 cryolite (Na3AlF6) phase, where half of the B sites are occupied by the same cations that occupy the A sites (that is, the formula is the same as that of an ordered perovskite when it is written as A2ABX6).(5) While the ordering of cations in HPs occurs only with B site cations, oxide perovskites can also form quadruple perovskites when both the A and B sites are occupied by two differently charged cations: these materials are generally described as having an AA′BB′O6 formula, like KSrXeNaO6, but the CaCu3Fe2Re2O12 compound also falls under this scheme.(5,11,25) In the case of oxide perovskites, there are also examples of ordered perovskites that consist of either alternating layers or columns of different metal octahedra. One example of this type of ordered perovskites is Sr2La2CuTi3O12, which is formed by three layers of TiO6 octahedra alternated with a layer of CuO6 octahedra. These structures have not yet been reported for HPs, but any advances in this direction will certainly lead to interesting new halide materials.(26) The final subgroup of halide perovskites are vacancy-ordered perovskites (Figure 2f).(27) These are similar to ordered perovskites, but the B-site cation is partially replaced with a vacancy. Among the most studied vacancy-ordered perovskites are those belonging to the Cs2BX6 group, in which half of the B-site cations are occupied by M(IV) cations and the other half by vacancies. Because these are crystallographically identical to ordered perovskites (A2B[V]X6, in which [V] is a vacancy), they share the same Fm3̅m space group. Vacancy-ordered perovskites in the M(IV) group include Cs2SnI6, Cs2PdBr6, Cs2TeI6, and Cs2TiBr6.(27) Vacancy-ordered perovskites can also form with trivalent metals like Sb3+ and Bi3+. In this case, the B sites are not occupied by M(III) and by a vacancy in equal ratios; instead, they are occupied at a ratio of 2:1, resulting in an A3B2X9 stoichiometry (A3B2[V]X9) such as Cs3Sb2I9.(28) In these ordered perovskites, the vacancies are ordered along the [111] planes, giving rise to a two-dimensional (2D) layer of BX6 octahedra. By mixing these vacancy-ordered perovskites with Cd2+, the fraction of vacancies can be further lowered to 25%, leading to a Cs4CdSb2Cl12 (A4BC2[V]X12) vacancy-ordered triple perovskite.(29) Because of the high concentration of vacancies, vacancy-ordered perovskites often exhibit low conductivity and thus have limited applications.(27) It is important to note that not all A3B2X9 halides crystallize in this vacancy-ordered perovskite phase: for example, Cs3Bi2I9 crystallizes in a nonperovskite phase (as will be discussed below). Nonperovskite Inorganic Metal Halides. Not all ternary metal halides can form a stable perovskite structure. In general, the HP framework collapses if the A-site cation is either too small or too large for the metal halide network(19) (the reader should refer to the Goldsmith tolerance factor, which is a figure of merit that indicates whether ternary metal halides can form geometrically stable perovskites or not).(12,30,31) As is shown in Figure 3a, ABX3 metal halides that cannot form geometrically stable perovskites crystallize in either 1D or 2D edge-sharing octahedral lattices (like CsPbI3 or RbPbBr3, in which the A-site cation would be too small to stabilize the hypothetical perovskite structure) or a 1D face-sharing crystal lattice (like FAPbI3, in which the A-site cation would be too large for the perovskite structure). The structural stability of ternary metal halide perovskites depends not only on the size of the ions but also on temperature and pressure:(19) ternary metal halides that do not form stable perovskites under ambient conditions may do so at high temperatures or pressures; therefore, they are often called “post-perovskites”. In the case of LHPs, the two most studied post-perovskites are CsPbI3 and FAPbI3, both of which are post-perovskites under ambient conditions, but they form a perovskite phase when they are annealed at around 200–300 °C.(19) Figure 3. Overview of ternary metal halides not crystallizing in a perovskite structure. (a) Post-perovskites formed by ABX3 ternary metal halides that would form stable perovskites at higher temperatures/pressures. (b) Ternary M(II) halides with nonperovskite stoichiometries, including inorganic Ruddlesden–Popper metal halide (A2BX4), A4BX6, and AB2X5 phases. (c) Ternary bismuth halides not crystallizing in vacancy-ordered perovskites, including the isolated dimer structure (Cs3Bi2I9) and isolated octahedra (Cs3BiX6). (d) Several ternary M(I) and M(II) (mainly transition metal and F– and Cl–) phases based on MX4 tetrahedra. Inorganic ternary M(II) halides (like those based on Sn2+ and Pb2+ions) can also crystallize in several other stoichiometries (as shown in Figure 3b) with phases that can be very different from perovskites. One of these phases, the A4BX6 one (such as Cs4PbBr6), has often been called a “0D perovskite”.(31,32) Although this phase consists of individual disconnected octahedra, similar to the A2BX6 vacancy-ordered perovskites, the BX6 octahedra are no longer in the crystallographic positions that correspond to those of a perovskite. Moreover, the A-site cations now occupy two different crystallographic sites.(33) Therefore, an A4BX6 phase cannot be considered a perovskite. One ternary phase which is often called a “2D perovskite” is the AB2X5 phase (the most known example of which is CsPb2Br5). CsPb2Br5 is generally described as a stack of layers of connected [B2X5]+ clusters, separated by layers of Cs+ ions.(34) A close inspection of this structure (Figure 3b, right structure) reveals that it does not even contain MX6 octahedra; therefore, it falls short of being considered a perovskite. Although thus not perovskites, these materials recently gained interest for their bright and narrow green photoluminescence (Cs4PbBr6 and CsPb2Br5) and for their use as a remote thermography material (Cs4SnBr6).(31,32,35) Heavy metal (Bi and Sb) based A3B2X9 compounds usually crystallize in a vacancy-ordered perovskite structure (as mentioned above). However, ternary bismuth iodides with large A-site cations, like Cs3Bi2I9, MA3Bi2I9, and FA3Bi2I9, crystallize in a different phase.(36) In this A3B2X9 phase, [Bi2I9]3+ dimers are formed by two face-sharing BiI6 octahedra, similar to Cs3Cr2Cl9.(37) These dimers are separated by Cs+ ions, resulting in a hexagonal 0D ternary metal halide, as is shown in Figure 3c. Thus, this phase does not much resemble a perovskite. Finally, Cs–Bi–X compounds can also crystallize in the Cs3BiX6 phase featuring single BX6 octahedra which are no longer in the crystallographic positions that correspond to those of a perovskite. Therefore, they are not a A3BX6 cryolite, as described above. Consequently, they are not (double) perovkites.(38) In addition to the tolerance factor, another important parameter that dictates the stability of perovskites is the octahedral factor,(12) as it relates to the stability of the BX6 octahedron. For instance, small cations that are coordinated by large anions (such as first row d metals and large anions like Br– and I–) generally prefer tetrahedral coordination. Therefore, HPs containing first row transition metals are primarily limited to fluorides, and there are only a few examples of HPs based on chlorides.(12) For example, ternary compounds like A3Cu2X5 and ACu2X3,(39) or compounds like A2BX4 and A3BX5 in which B is a first row transition metal (such as Zn2+, Ni2+, Cu2+, and Mn2+)(40) all consist of tetrahedral BX4 clusters surrounded by A+ cations. Therefore, they cannot be classified as perovskites (see Figure 3d). However, they can be interesting nontoxic broad emitters (or at least less toxic than Pb-based ones).(40) Inorganic Layered Metal Halides. Among the various ternary metal halides, Ruddlesden–Popper (RP) phases (and layered metal halides in general, which will be discussed below) present a structurally interesting case. In the RP phase, which has an A2BX4 stoichiometry, a single 2D layer of corner-sharing M(II) halide octahedra is separated by a layer of A atoms (Figure 3b).(41) Even though oxide RP phases can be synthesized using a wide variety of metals (such as Sr2RuO4, La2NiO4, and LaSrCoO4),(42) inorganic RP halide phases have, thus far, been limited to Cs2PbI2Cl2 and Cs2SnI2Cl2.(43) In an inorganic halide RP phase, the separation into layers is driven by the size difference between the two halides.(41) Although (in oxides) these RP phases are often characterized as perovskites (“Ruddlesden–Popper perovskites”), it is our opinion that they should be classified only as RPs, and not as perovskites. We justify this opinion on the basis of the following two points: (i) These phases already have a common name (i.e., “Ruddlesden–Popper”). (ii) Starting from a perovskite structure, one cannot derive such phases without having to severely dismantle the perovskite lattice, at least locally. Furthermore, RP halides exhibit electronic and optical properties that are markedly different from their perovskite counterparts.(41) The same reasoning applies to other phases consisting of stacks of layers made of corner-sharing octahedra, like Dion–Jacobson phases and the Aurivillius phase (both of which have been reported only for oxides to date),(44) as well as hybrid organic–inorganic layered metal halides, which will be discussed next. Hybrid Organic–Inorganic Metal Halides. Hybrid organic–inorganic metal halides have gained an increasing amount of attention over the past decade.(19) These are often also referred to as “hybrid perovskites” (Figure 4).(9,45) In these materials, the A site is occupied by an organic cation, most commonly an alkyl ammonium cation. These types of materials are not organometallic compounds,(46,47) as they do not contain any carbon–metal bonds, therefore they should not be called “organometallics” or “organolead” perovskites, but rather “hybrid organic–inorganic perovskites” (HOIPs).(9) In the case of HOIPs, the A cations can consist of only small (pseudo spherical) MA and FA cations and the larger B cations, like Pb2+ and Sn2+,(9) as is shown in Figure 4a. Here, it must be noted that not all of the hybrid organic–inorganic Pb and Sn halides form perovskites,(19) as some (like FAPbI3) are stable only at high temperatures, thus forming “post-perovskites” at ambient conditions.(19) Small transition metal fluorides can also form stable HOIPs with NH4+, like NH4MnF3 and NH4ZnF3.(48) Because of the anisotropy in the inorganic cations, the HOIPs have a lower symmetry than their inorganic counterparts, even in their pure cubic crystal structure.(19) As the size of the A cation increases, layered compounds are formed, and these compounds usually crystallize in an A2BX4 (or ABX4 in the case of cations with two ammonium groups) stoichiometry, in which layers of corner-sharing octahedra are separated by a layer of interlocking organic hydrocarbon chains (Figure 4c).(45,49) Following the same line of reasoning as in the RP case above, these materials should not be considered perovskites either. In our opinion, these compounds should be simply classified as “layered hybrid organic−inorganic metal halides”. One subclass of these layered hybrid organic–inorganic metal halides that has gained considerable interest because of their better moisture stability and higher degree of optoelectronic fine-tuning than conventional lead halide perovskites is the one containing several consecutive layers of 3D connected metal halides alternated with a single layer of organic cations (often called “2D perovskites” or “layered perovskites”).(19) These compounds can be synthesized by using a mixture of a small cation that can form a perovskite (e.g., Cs+ or MA) and a large organic cation that can form only layered metal halides. These layered materials have a general [L]An–1BnX3n+1 formula, in which L is the large organic cation (sometimes referred to as a “ligand”) and n is the number of layers of 3D connected octahedra.(45,46,49) In this case, the line between a layered metal halide (n = 1) and a perovskite (n ≈ ∞) remains a topic of debate and, to a certain extent, a matter of opinion. Finally, hybrid organic–inorganic metal halides have been reported using bulky organic cations or organic cations consisting of multiple ammonium groups.(46) These compounds often are either formed by single chains of metal halide octahedra (Figure 4c) or by completely isolated metal halide clusters (Figure 4D), and they are sometimes referred to as “1D” or “0D organic perovskites”.(46) These types of compounds have very little in common with the perovskite crystal structure, therefore they should not be identified as perovskites, but rather as “0D” or ‘1D hybrid organic–inorganic metal halides’. Figure 4. Overview of layered and hybrid organic–inorganic metal halides. (a) Hybrid organic–inorganic perovskite structure, with the A cation being a small organic cation like methylammonium. (b) Layered (2D) hybrid organic–inorganic metal halides, consisting of planar or zigzag corner-sharing octahedra. (c) 1D metal halides, consisting of chains of facet or corner-sharing octahedra. (d) 0D hybrid organic–inorganic metal halides, consisting of isolated metal halide octahedra. Antiperovskite with Halide B Sites. Although the aforementioned AB2X5 and A3BX5 compositions do not consist of octahedrally coordinated metal cations (and are therefore strictly not perovskites), they can qualify as inverse antiperovskites, because they have XB6 octahedra and small X or BX clusters occupying the A site. They are not the same as antiperovskites, in which the halides generally occupy only the A site, and are not part of the perovskite framework. In the case of an A3BX5 phase, using Cs3ZnBr5 as an example, the Br– anion can be interpreted as being octahedrally coordinated by Cs+ cations, resulting in a [BrCs3]2+ cubic framework.(50) Within this framework, the [ZnBr4]2– clusters occupy the A sites, as is shown in Figure 5a. This is also similar to a double antiperovskite Na6FCl(SO4)2 (sulphohalite, a natural mineral), in which the A site is occupied by a SO42– cluster and the perovskite framework consists of FNa6 and ClNa6 octahedra, as well as Ba3(FeS4)Br, which has a [FeS4]−5 A-site cluster.(5,51) Small inorganic clusters in A sites can also be found in a standard perovskite structure, like Tl4SnTe3, which is a chalcogenide perovskite (consisting of SnTe6 octahedra) with an A site that is occupied by a tetrahedral Tl44+ cluster.(52) Similarly, the AB2X5 phase (for example, CsPb2Br5, as is shown in Figure 5b) can also be considered an antiperovskite consisting of octahedrally coordinated halide anions. Here, [BrPb2Cs]4+ octahedra form a perovskite framework, with the A cation site being occupied by a cluster of 4 Br– ions. This is analogous to the TlSn2I5 case.(53) The possibility for small inorganic clusters to occupy the A site of a standard perovskite was recently demonstrated with the vacancy-ordered halide perovskite structure of Cs3Cu4In2Cl13, in which 25% of the A sites are occupied by [Cu4Cl]3+ clusters and half of the B sides are vacant (a structure that can be more appropriately described as ([Cu4Cl]Cs3In2[V]2Cl12), is shown in Figure 5c.(54) These cases of inverse antiperovskites, together with the inverse perovskites like BaLiF3, demonstrate that the A-site cations in perovskites are not simply limited to alkali metals (or small organic molecules), but they can also consist of small inorganic clusters and metal cations with different charges, a consideration that should inspire the discovery of new halide perovskites. Figure 5. Antiperovskites of A3BX5 and AB2X5 phases. (a) Cs3ZnBr5 and (b) CsPb2Br5 with (left) polyhedra centered on the halides and (right) top view shown along the c-axis. (c) Cs3Cu4In2Cl13 vacancy-ordered perovskite, consisting of InCl63– octahedra and Cu4Cl3+ tetrahedra occupying the A sites. Conclusions and Recommendations. Although there are many different types of HPs, the recent and rapid advancements in the field has led to a confusion on the correct way of naming all the different compounds that have been reported. In our opinion, if a compound is to be considered a “perovskite”, it must have an ABX3 (or equivalent) stoichiometry consisting of a cubic network of corner-sharing BX octahedra, with the following exceptions: (i) the A and B sites can be partially or fully vacant or ordered; (ii) the A cations can consist of small organic molecules or inorganic clusters; (iii) the formation of anti-halide perovskites, with a halide occupying either the A or B site. 1D and 0D hybrid organic–inorganic metal halides do not satisfy these conditions; therefore, they should not be considered as perovskites. Although this is a somewhat controversial subject, layered Ruddlesden–Popper (and similar) phases, as well as 2D hybrid organic–inorganic metal halides, do not fit these criteria either. Thus, it is our opinion that they should not be referred to as perovskites. As a final comment, we stress the importance of stating whether a perovskite is vacant, ordered, anti, or a combination of these. This not only identifies the deviations from a typical perovskite stoichiometry but also, in most cases, explains the completely different properties of the materials compared to their perovskite equivalent. Q.A.A.: Department of Chemistry and Applied Biosciences, ETH Zürich, Vladimir Prelog Weg 1, CH-8093 Zürich, Switzerland. The views expressed in this Viewpoint are those of the authors and not necessarily the views of the ACS. The authors declare no competing financial interest. We thank Emma De Cecco for proofreading the viewpoint. The research leading to these results has received funding from the seventh European Community Framework Programme under Grant Agreement No. 614897 (ERC Consolidator Grant “TRANS-NANO”). This article references 54 other publications.

In order to become a viable solution to compete with fossil fuel, solar energy-based chemical production should be robust, reliable, and economically feasible. A strategy for minimizing the technical and financial risk inherent in such an economic transition is to develop a process based on abundant raw material resources and known chemistries. Today it is technically possible to produce a transportable gas (H2), a transportable liquid fuel, or petrochemical feedstock (methanol) with minimal byproduct carbon dioxide (CO2), the primary global warming culprit. Solar energy can be harnessed directly by photovoltaic or photoelectrochemical technologies to produce H2.(1−3) One approach that has been under much discussion recently is linked to the so-called “H2 economy”. Although H2 is indeed a clean energy source, generating it is a high-energy consuming process releasing large amounts of CO2 into the atmosphere. Currently, more than 90% of H2 is produced by steam methane reforming (SMR) in industry, where around 20% of the energy from fossil fuel is lost as heat in addition to generating CO2 (stoichiometrically 5.5 kg of CO2 is produced per kg of H2).(4) Furthermore, H2 storage is one of the key challenges because of the low volumetric energy density and its explosive nature. Therefore, converting hydrogen to other compounds for transport and storage is a much needed approach. Yet, it is obvious that in this case hydrogen needs to originate from water. Consequently, a methanol economy was proposed as a reasonable alternative.(4) Methanol, in addition to its energy storage property, is used as a convenient fuel as well as a raw material for various hydrocarbons.(5) Moreover, methanol is used to generate electricity using direct methanol fuel cell (DMFC) technology.(6−9) Methanol has a potential market of more than 80 million tons per year globally.(10) Conventionally, methanol is produced using syngas generated from natural gas steam reforming or partial oxidation.(11) Reforming accounts for more than 60% of the capital cost and requires substantial energy generated from fossil fuels.(12) For example, a typical methanol plant releases about 0.5 kg of (CO2)/kg of (MeOH).(13) Hence, incorporating renewable energy in methanol synthesis is a necessary step to produce green methanol. Methanol in this scenario can be produced either by using CO generated from photochemical reduction of CO2(14) or directly by H2 made from solar water splitting.(13) For the first approach, Kim et al. reported a comprehensive technoeconomic analysis using the photothermal approach to activate CO2 to CO and then use the CO for methanol production.(14) The break-even price of 1.22 $/kg for the methanol was obtained by their approach.(14) As for utilizing renewable H2, Alsayegh et al. demonstrated the process and economic viability of a new production route for methanol, utilizing captured CO2 and H2 produced from photocatalytic water splitting.(13) The break-even price for the methanol was 0.94 $/kg. H2 was produced using a “Baggies” approach described by James et al.(15) The main cost driver was to recover the diluted H2 from CO2, which was used to prevent explosion hazard. These two studies revealed that the main cost is in the H2 plant and gas separation. In this work, we have identified a conceptual process that utilizes ultrahigh concentrated photovoltaic (UHC-PV)-based water splitting for H2 production and a carbon capture and reuse (CCR) process to create a new production route for methanol through direct CO2 hydrogenation. The words ultrahigh concentrated (UHC) refer to sun concentration of a few hundred and above using multijunction solar cells. Multijunction GaAs-based cells, while expensive, can be made practical and therefore economical when used under very high light fluxes (hence the acronym UHC), as presented in this work. The process consists of three main sections, namely, H2 plant, CO2 capture plant, and methanol plant (Figure 1). The hybrid H2 plant will operate using the UHC-PV power generation unit during the day followed by direct electrolysis using electricity coming from a grid at night. Dried/compressed H2 stream leaving the H2 plant is mixed with CO2 supplied from CO2 capture units from fossil fuel power plants at a ratio of 3:1 (H2:CO2) mole-based. The H2/CO2 mixture is fed to the methanol plant, where a Cu/ZnO/Al2O3 catalyst is used to produce methanol.(16,17) Figure 1. General layout for the proposed methanol production plant from water and CO2. Optimum or ideal case studies regarding all plants were extracted from the literature and were assimilated into the overall economic and energetic analyses. All process and economic parameters along with detailed plant descriptions are provided in the Supporting Information S.1 and S.2. Process Assumptions. The results obtained from this study show the break-even value for the production of both the H2 and methanol from the UHC-PV hybrid system under the following assumptions: Each process section is separated with no heat integration. The process is based on a net H2 production rate of 2.0 tons/h. There is no interruption in H2 production while switching between solar and grid power. For the base case, the hybrid mode of H2 production is assumed, where the UHC-PV unit will be utilized for 11.3 h/day (daytime), followed by direct electrolysis, using electricity from the grid at night (12.7 h/day). The cost of feed from the CO2 plant is considered a purchased raw material. The CO2 stream from the capture section is pure CO2. Each process section is separated with no heat integration. The process is based on a net H2 production rate of 2.0 tons/h. There is no interruption in H2 production while switching between solar and grid power. For the base case, the hybrid mode of H2 production is assumed, where the UHC-PV unit will be utilized for 11.3 h/day (daytime), followed by direct electrolysis, using electricity from the grid at night (12.7 h/day). The cost of feed from the CO2 plant is considered a purchased raw material. The CO2 stream from the capture section is pure CO2. H2 Plant Analysis. For the base case (Supporting Information S.2.1), the estimated capital cost for the plant is 398.4 million $, and the operating cost is 26.2 million $/year for the desired H2 production of 2 tons/h. Table 1 shows the breakdown of both CAPEX and OPEX of the H2 plant. The minimum levelized selling price (break-even value, NPV = 0) of H2 is at 4.79 $/kg, which is around 3 times higher than the conventional H2 price of 1.39 $/kg from steam reforming.(18,19) As can be seen from Table 1, 91% of the capital cost is associated with the solar tracker-based UHC-PV power generation unit. Looking at the OPEX analysis, the majority of the cost is related to electricity, as expected, with more than 84% of the total cost. Using grid electrolysis for more than half of the H2 production time significantly contributed to this high cost. The sensitivity analysis, with respect to the selected parameters, is shown in Figure 2. The break-even value for H2 was calculated, and the results revealed a strong relationship between the UHC-PV power generation unit cost and solar-to-hydrogen (STH) factor. The levelized cost of H2 (LCH) increases by 47% when the electricity price changes from 0.02 to 0.09 $/kWh. This strong dependence makes a significant impact on the OPEX analysis where the electricity is the major OPEX cost. Changing the STH to a lower value of 10% results in an increase in solar collection area and subsequently requires more land as well as more UHC-PV units. Hence, the LCH increases by 116% when we compare the lower and upper STH values. However, when an STH is 30%, only 8% decrease in LCH is found. As for the best case scenario, where the STH value is 30%, with electricity at 0.02 $/kWh, UHC-PV unit cost at 0.40 $/W, and electrolyzer cost at 0.2 $/kgH2, the levelized cost for H2 would drop to 2.2 $/kg. Figure 2. Sensitivity analysis of the H2 plant based on the UHC-PV hybrid system. For the base case parameters, Figure 3 shows the cost range of H2 production from both the UHC-PV system and traditional grid electrolysis. Note that the levelized cost is lower for the hybrid system than grid-based electrolysis, when the grid electricity price is higher than 0.13 $/kWh. The H2 price obtained from direct electrolysis is in good agreement with the literature data.(15) The generally accepted cost for industrial electricity is 0.07 $/kWh in developed economies, which makes the H2 price from UHC-PV at 37% more expensive than direct electrolysis. This gap in cost can be reduced by improving the main variables influencing the cost, i.e., STH and the UHC-PV unit cost. Figure 3. H2 production price comparison between a direct electrolyzer and the proposed hybrid system. MeOH Plant Analysis. As for the whole methanol process, the resulting capacity is at 77 738 tons/year with a purity of 99.4 wt %. The overall process and economic results are summarized in Table 2, which shows a break-even price of methanol at 1.12 $/kg for the base case parameters. H2 plant capital and operating costs account for more than 95% of the entire plant cost. It must be noted that the methanol operating and capital cost is lower than that of a typical methanol plant. This can be explained by the fact that both gases (H2 and CO2) are delivered to this plant at high pressure and the compression section is not required. The sensitivity analysis for the break-even value of methanol is shown in Figure 4. The results show similar behavior to the H2 sensitivity graph where both UHC-PV cost and STH influence methanol price by up to 78%. The CO2 price has low influence, demonstrated by the 9% increase in methanol price when the capture cost is at 0.1 $/kg. This can be explained by the fact that the CO2 cost as a raw material accounts for about 13% of the total operating cost of the whole plant. Beside UHC-PV cost and STH influences, the electricity price has a considerable influence on the load control module (LCM) as it did for the H2 price above. Moreover, the effect of the electrolyzer capital cost is relatively insignificant compared to its effect on the H2 value. For the best case scenario, the LCM would drop to 0.54 $/kg. Figure 4. Sensitivity analysis of the methanol plant based on the UHC-PV hybrid system. Process Energy Analysis. Net energy efficiency (NEE), gross energy efficiency (GEE), and primary energy efficiency (PEE) calculations were used for the comparison with other solar energy-based methanol production routes.(13,14) Detailed calculation and required values are listed in the Supporting Information (2.5). Table 3 summarizes these energy analyses. The high value of the GEE (19.9%) is a direct result of the advantage of the hybrid system where the electrolyzer efficiency is at 75% compared to the STH efficiency of 23.3%. When all of a utility’s energy requirement is converted to primary energy, the efficiency drops to 14.9% because most of the utilities needed for the H2 plant are for electricity. Further improvement in both efficiencies can be realized by optimizing the heat integration among the three plants and improved STH values. Summary and a Path Forward. A sustainable economy for the future needs to be suitable for the needs of an ever-expanding population with reduced environmental impact. The scientific know-how is currently available to begin meeting those future needs with a smaller environmental footprint. Technical (scientific and engineering) risks remain to be addressed, though these pale in comparison to the financial risks of establishing a completely new energy and petrochemical infrastructure for a more sustainable future. Large investments must be made to establish manufacturing capability for requisite materials having the appropriate quality and form factor, such as III–V semiconductors, which might be the key enablers for the desired future, because of their present high efficiency. A large portion of the existing petroleum refining and electrical power generation capability, which has taken over a century to be developed, would need to be replaced and/or reconfigured. As a consequence, transportation systems will be affected along with them is the fuel delivery system. The cost of hydrogen and methanol production from renewable sources eventually will become competitive with the present ones. UHC-PV-based technology either directly connected to a matching electrocatalyst performance or as a one system in the form of a photocatalyst(20) is poised to be part of the solution. A roadmap for implementation is at present needed considering first a hybrid system that can be incorporated into an existing technology and then followed by larger-scale fully renewable technology upon completion of the technical challenges. Here we propose an alternative route for sustainable methanol production. Captured CO2 and hybrid solar-driven H2 (ultrahigh concentrated solar module and water electrolysis) are utilized for direct CO2 hydrogenation. On the basis of optimal or ideal case studies regarding all plants involved, we were able to estimate the break-even value of methanol (1.12 $/kg) for the base case parameters (23.3% STH and 0.05 $/kWh). Through sensitivity analysis, the cost of methanol can be reduced significantly by reducing the cost of the water-splitting unit including the UHC-PV and electrolyzer. This is entirely possible by applying good engineering principles and invoking the economy of scale to the UHC-PV unit, electrolyzer stack, and their balance of systems. Optimal power matching between the UHC-PV and electrolyzer is the key step to achieve a high STH, which is the second significant factor in both hydrogen and methanol sensitivity tables. The energy efficiency and methanol cost estimated for the process route suggested in this study have been compared to those for other proposed alternatives using similar assumptions and involving photoelectrochemical or photothermal process steps. It was found that the process described in this report was slightly more energy efficient than that for the other methods described in the literature.(13,14) Though we have envisioned a system where CO2 is captured from fossil fuel power plant exhaust, it should be mentioned that CO2 can be extracted from air using similar known technologies. Thus, the vision of a (near) carbon neutral economy is at hand with PV (and possibly wind)-only powered electrolysis of water to produce hydrogen, which, in combination with CO2, in the air can provide a sustainable future. Once carbon neutral methanol is available at a competitive price, it can be used as a fuel itself or be an intermediate for liquid hydrocarbon fuels and chemicals in production today. Nevertheless, all green methanol is possible if the electricity exclusively originates from a PV farm (or other fully renewable systems). Some attractive PV electricity prices are floating around in global bids.(21) For example, Saudi Arabia received a 1.78 ¢/kWh bid (declined), Mexico saw a 1.97 ¢/kWh bid, and Chile got a 2.15 ¢/kWh bid. These projected prices from pure renewables are far cheaper than the crossing point (13.2 ¢/kWh), where the hybrid system would be a better choice (Figure 3). Furthermore, traditional water electrolysis is a mature technology, and less risk is associated with the investment. Therefore, the hybrid system may not compete with the solar electrolysis system if the above electricity costs become indeed feasible. The current electricity price (excluding CO2 sequestration costs and associated environmental issues) is also more in favor of direct electrolysis unless the cost of the UHC-PV system comes down significantly in the future or carbon taxation is put in place on hydrogen originating from methane reformers. The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsenergylett.9b02455. Detailed plant descriptions and process and economics parameters/calculations (PDF) Detailed plant descriptions and process and economics parameters/calculations (PDF) Views expressed in this Viewpoint are those of the authors and not necessarily the views of the ACS. The authors declare no competing financial interest. Independent consultant, Seattle, Washington, United States. Electronic Supporting Information files are available without a subscription to ACS Web Editions. The American Chemical Society holds a copyright ownership interest in any copyrightable Supporting Information. Files available from the ACS website may be downloaded for personal use only. Users are not otherwise permitted to reproduce, republish, redistribute, or sell any Supporting Information from the ACS website, either in whole or in part, in either machine-readable form or any other form without permission from the American Chemical Society. For permission to reproduce, republish and redistribute this material, requesters must process their own requests via the RightsLink permission system. Information about how to use the RightsLink permission system can be found at http://pubs.acs.org/page/copyright/permissions.html. This work was financially supported by the SABIC-Corporate Research and Development (CRD) Center at King Abdullah University for Science and Technology (KAUST). This article references 21 other publications.

At present, the replacement of fossil fuels by alternative CO2-emission-free energy sources, such as solar and wind, is substantially hindered by the lack of low-cost and large-scale energy storage technologies. Although lithium-ion batteries (LIBs) represent the most mature and widely deployed electrochemical energy storage technology for mobility and portable electronics, the uneven worldwide distribution of known lithium reserves and the high production costs greatly reduce the economic appeal of LIBs for large-scale stationary storage.(1−5) In this regard, inexpensive Al–graphite dual-ion batteries (AGDIBs) have attracted great attention over the past few years.(6−26) With energy densities of 30–70 Wh kg–1, AGDIBs are suitable for stationary storage.(27−30) The constituents of AGDIBs include highly abundant elements (H, O, N, C, and Al) and are easy to manufacture. While recent research efforts on AGDIBs have been mainly focused on testing various graphite cathodes, further progress of this technology is inherently limited by the low charge storage capacity of the chloroaluminate ionic liquids used as anolytes (often but incorrectly called electrolytes). In this Viewpoint, we discuss the critical interplay between the capacity of the anolyte and the energy density of AGDIBs along with their measurements at different current densities. The basic configuration of an AGDIB contains a graphite cathode, AlCl3-EMIMCl (1-ethyl-3-methylimidazolium chloride) ionic liquid anolyte, and metallic aluminum current collector, as demonstrated in Figure 1a. AGDIBs operate as an electrochemical energy storage system by employing the reversible intercalation of AlCl4– anion species into the positive graphite electrode upon charging (i.e., the oxidation of the graphite network). Concurrently, the aluminum electroplating reaction takes place on the negative side of AGDIBs. The working principle of AGDIBs can be represented by the following cathodic and anodic half-reactions during charging: On the negative electrode(I) On the positive electrode(II) On the negative electrode On the positive electrode Figure 1. (a) Schematic of the charging process of aluminum GDIBs. (b) Charge storage capacity of the chloroaluminate ionic liquid anolyte (AlCl3:EMIMCl) versus its acidity (r). The curve is computed from eq 1. The chloroaluminate ionic liquid anolyte used in AGDIBs is defined as a mixture of aluminum chloride and other chlorides countered by a bulky organic cation, such as 1-ethyl-3-methylimidazolium chloride (EMIM) or 1-butyl-3-methylimidazolium chloride (BMIM). As a result of the acid–base reaction between AlCl3 (Lewis acid) and Cl– (Lewis base), the solid mixture liquefies at room temperature, forming AlCl4– anions that are charge-balanced with asymmetric organic cations, such as EMIM+. The chloroaluminate ionic liquid with an excess of AlCl3 over EMIMCl is composed of both AlCl4– and Al2Cl7– ions. Importantly, only Al2Cl7– ions enable the electroplating of aluminum, which, therefore, occurs only in acidic chloroaluminate melts (i.e., an excess of AlCl3).(31−40) Consequently, the specific charge storage capacity of the acidic melt is a function of the concentration of Al2Cl7– ions in the ionic liquid. Electroplating, and consequently the charging process, stops when no Al2Cl7– ions remain in the ionic liquid, at which point the melt becomes neutral (AlCl3:EMIMCl = 1). The highest molar ratio (r) of AlCl3 and EMIMCl that forms an ionic liquid is ca. 2:1. AlCl3 does not dissolve in the chloroaluminate melts at higher molar ratios. In this context, we note that the mechanism of AGDIBs is substantially different from the operation of “rocking-chair” metal-ion batteries: there is no unidirectional flow of Al3+ ions from the cathode to the anode and vice versa. In fact, the Al species are depleted from the chloroaluminate ionic liquid during the charge of AGDIBs and are being consumed by both electrodes (Figure 1a). Notably, the electrochemistry of AGDIBs is not restricted to AlCl3/EMIMCl ionic liquids. Other possible ionic melts have also been recently tested in AGDIBs, such as AlCl3/1-methyl-3-propylimidazolium chloride (MPIMCl),(41) AlCl3/benzyltriethylammonium chloride (TEBACl),(42) AlCl3/1,2-dimethyl-3-propylimidazolium chloroaluminate (DMPICl),(43) AlCl3/NaCl,(44) and AlCl3/LiCl/KCl(45) or AlCl3/urea/EMIMCl,(46) AlCl3/urea,(47−49) and AlCl3/Et3NHCl.(50,51) Figure 1b illustrates the impact of acidity (r) on the charge storage capacity of the chloroaluminate ionic liquid anolyte. The theoretical capacity of the ionic liquid (Can) as an anolyte, considering its whole mass/volume, can be calculated as follows: Gravimetric(1) Volumetric(2)where F = 26.8 × 103 mAh mol–1 (the Faraday constant), and (number of electrons used to reduce 1 mol of the Al2Cl7– ions); MAlCl3 is the molar mass of AlCl3 in g mol–1, MACl the molar mass of the Cl– source (for example, EMIMCl) in g mol–1, r the AlCl3:ACl molar ratio, and ρ the density of the chloroaluminate-based anolyte in g mL–1. Gravimetric Volumetric The theoretical gravimetric charge storage capacities of AlCl3:EMIMCl ionic liquids equal, for instance, to 48 mAh g–1 and 19 mAh g–1 for r = 2 and r = 1.3, respectively. These capacities determine and limit the overall energy density of AGDIBs, as previously pointed by us(7,27) and others.(52−59) To the best of our knowledge, it had not been verified whether these capacities are achievable experimentally, i.e. whether Al2Cl7– ions can be fully depleted for Al electroplating at practically relevant experimental conditions. For this, the anolyte-limited cell must be assembled, that is, the cell with the significant excess of graphite cathode. On the contrary, the commonly reported AGDIB tests employ an up to 10-fold excess of anolyte (or even higher). We note that such a 10-fold anolyte access can be recalculated into an impractical overall charge-storage capacity of the cell (<5 mAh g–1). In the anolyte-limited cell, depletion of Al2Cl7– ions r will eventually cause a drop in the potential at the negative electrode of the battery. It is thus also important to conduct tests in a three-electrode configuration in order to differentiate between changes in the potentials originating at positive and negative electrodes. Following these considerations, we prepared an anolyte-limited full cell, in which an excess of graphite flakes was used in combination with the anolyte, i.e. having roughly an order of magnitude higher cathodic capacity in comparison to that needed to match the theoretical charge storage capacity of the anolyte. AGDIBs were assembled using a three-electrode cell composed of a glassy carbon current collector, graphite flake cathode, Al foil reference electrode, separator impregnated with chloroaluminate ionic liquid anolyte, and Al foil current collector (Figure 2a). The cells were cycled between 2.5 and 0.1 V vs Al3+/Al. An example of the electrochemical measurement at a current density of 60 mA g–1 is shown in Figure 2b using a chloroaluminate ionic liquid with a molar ratio of 2.0. In these measurements, the simultaneous acquisition of the potential profiles of both positive and negative electrodes was recorded, in addition to that of the full cell. Figure 2. (a) Three-electrode cell configuration of the AGDIBs (the location of the reference electrode is shown on the left). (b and c) Galvanostatic charge curves for the chloroaluminate ionic liquid (r = 2) anolyte (ECE), graphite (EWE), and full cell (ECell) measured versus the Al foil reference electrode in anolyte-limited (b) and graphite-limited (c) cell configurations. The cells were charged at room temperature with a current density of 60 mA g–1 and an upper voltage limit of 2.5 V vs Al3+/Al. In the case of the graphite-limited cell configuration, a 5-fold excess of the anolyte over graphite was used. As follows from Figure 2b, the voltage profile at the negative electrode (ECE) remains relatively stable during charging for 15 min, with a small overpotential of up to 200 mV. However, upon further charging, this voltage drops sharply when the Al plating process ends. The plating termination is also reflected in the sharp increase in the overall potential of the cell of up to 2.5 V vs Al3+/Al. The potential at the positive electrode (EWE), however, is relatively stable for the full charging cycle. In contrast to the anolyte-limited cell, the voltage profile at the negative electrode (ECE) for the graphite-limited cell is very stable during the entire charge, with a minimal overpotential of <50 mV (Figure 2c). As indicated above, the graphite-limited cell configuration is used for the assessment of the charge storage capacity of graphite in the majority of research studies on AGDIBs. Using the ionic liquid formulations with r = 1.3–2.0, we performed rate capability measurements of anolyte-limited full cells at different current densities ranging from 5 A g–1 to 20 mA g–1. Figure 3a presents the galvanostatic curves of the anolyte with r = 2 during charging (see Figure S1 for details). The electrochemical data for r = 1.3 and 1.8 are shown in the Supporting Information (Figures S2 and S3). Figure 3b summarizes these results in terms of the obtained charge storage capacities at different current densities. The results reveal two major points. First, a more acidic anolyte yields, as expected, a higher anolyte capacity. For instance, the charge storage capacity of the most commonly used anolyte with r = 1.3 is ca. 21 mAh g–1 at 20 mA g–1. In contrast, the anolyte with the highest acidity (r = 2.0) possesses a capacity of ca. 46 mAh g–1. These results point to the fact that the highest energy density of the AGDIBs can be obtained using a 2.0 molar ratio, and therefore, future works on AGDIBs should be focused on the most acidic formulations. Second, the charge storage capacity of the anolyte is highly dependent on the applied current density. For instance, at a high current density of 1 A g–1, minimal charge storage capacities are obtained (ca. 10–14% from theoretical values). The latter point is reflected in the pronounced deviation of the voltage profiles at the negative electrode at high current densities (Figures 3a and S1–S3). Conversely, minimal polarization is observed at the graphite positive electrode. These results indicate that the frequent statements regarding the high power density of AGDIBs need to be taken with care. Specifically, we note that at high current densities a significant drop in the energy density of AGDIBs is expected and originated from the rate capability limitations of the chloroaluminate ionic liquid anolyte when its acidity (concentration of Al2Cl7–) is drastically reduced. In fact, our observations show that the charge storage capacities of the anolyte significantly deviate from the theoretical value at charge current densities greater than 20 mA g–1. Figure 3. (a) Galvanostatic Al plating (discharge) curves of the chloroaluminate ionic liquid anolyte with r = 2.0 measured at different current densities in combination with graphite as the working electrode and Al foil as the reference electrode. (b) Charge storage capacities of the chloroaluminate ionic liquid anolyte with r = 1.3, 1.8, and 2.0 measured at different current densities. The green line shows the theoretical charge storage capacity of the anolyte computed from eq 1. In conclusion, we reiterate that commonly reported tests of AGDIBs employ large excess of ionic liquid anolyte (cathode-limited cells) and, despite nominally dealing with full cells, do not provide correct and practically relevant information on achievable energy and power densities, as well as cycling stability, energy and Coulombic efficiencies of these batteries. At best, such tests yield the theoretical capacities and other characteristics of the cathode material only, similar to, for instance, Li-ion half-cell tests of novel cathodes (with thick Li foil as a counter electrode). We further note that ionic liquids used in AGDIBs are not just electrolytes (ion-conductors) but represent an electrochemically active, capacity- and rate-limiting battery component. It is also apparent that future work should focus on finding ways of minimizing the amount of this anolyte toward the capacity-matched quantity, ideally with just ca. 10% of the excess anolyte, without sacrificing the power density and cyclability of the battery. Most likely, a successful solution to this problem will invoke a radically new battery design, vastly different from that employed in commercial Li-ion batteries and nearly all tests of new battery materials. All these advancements will need to come at low capital costs in order to retain the overall cost-competitiveness that is commonly attributed to AGDIBs (based on low costs of electrochemically active constituents). Similar considerations apply also to other nonrocking-chair batteries, such as magnesium–sodium and magnesium–lithium dual-ion batteries. The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsenergylett.9b02832. Additional electrochemical data (PDF) Additional electrochemical data (PDF) Views expressed in this Viewpoint are those of the authors and not necessarily the views of the ACS. The authors declare no competing financial interest. Electronic Supporting Information files are available without a subscription to ACS Web Editions. The American Chemical Society holds a copyright ownership interest in any copyrightable Supporting Information. Files available from the ACS website may be downloaded for personal use only. Users are not otherwise permitted to reproduce, republish, redistribute, or sell any Supporting Information from the ACS website, either in whole or in part, in either machine-readable form or any other form without permission from the American Chemical Society. For permission to reproduce, republish and redistribute this material, requesters must process their own requests via the RightsLink permission system. Information about how to use the RightsLink permission system can be found at http://pubs.acs.org/page/copyright/permissions.html. This research is part of the activities of SCCER HaE, which is financially supported by Innosuisse - Swiss Innovation Agency. This article references 59 other publications.

One of the major challenges facing science and society in the 21st century is sustainable energy. The dwindling supply of fossil fuels and their adverse effect on the environment make it imperative to replace these fuels with alternate forms of energy that are clean, renewable, and affordable. To meet this challenge, materials need to be developed that can efficiently harness, store, and convert energies from the sun and the wind, in addition to other renewable sources. Atomic clusters and nanostructured materials, because of their unique size and composition-specific properties, are ideally suited to address the many challenges. Note that atomic clusters are the ultimate nanoparticles where structure–property relationships can be studied one atom and one electron at a time and, hence, can address fundamental questions. The objective of the International Symposium on Clusters and Nanomaterials (ISCAN2019), held at the historic Jefferson hotel in Richmond, Virginia on November 3–7, 2019, was to gain a fundamental understanding of the materials issues in energy production, storage, and conversion. This quadrennial symposium, founded in 1982, was hosted by Virginia Commonwealth University (VCU) and supported by VCU, the Office of Basic Energy Sciences of the Department of Energy, Army Research Office, and ACS Energy Letters. Realizing that many of the materials issues in sustainable energies and environmental impacts of current energy technologies are intertwined and solving these problems requires a multidisciplinary team of scientists and engineers, ISCAN2019 brought together researchers from the fields of physics, chemistry, materials science, and engineering to share their ideas and results, identify outstanding problems, and develop new collaborations. A special session was also held in honor of Will Castleman, Millie Dresselhaus, Walter Kohn, and John Yates who were notably associated with The Richmond Symposium series for many years and recently passed away. (See Figure 1.) Figure 1. ISCAN2019 group photo. (Photo Courtesy: Thomas M. Kojscich) ISCAN2019, the 10th Richmond symposium, consisted of 14 plenary sessions with 41 invited speakers from 11 countries. In total, 142 conferees from 21 countries and five continents took part in this symposium. In addition to the invited talks, two poster sessions of contributed works as well as 10 hot topics talks, selected from the contributed abstracts, were presented. ISCAN2019 also covered topics in medicine and the unique role that nanomaterials can play in this area. The topics included Clean and Sustainable Energy: Production, Storage, Conversion, and Efficiency (solar, hydrogen, energy storage such as batteries and photothermal energy conversion) and Medicine (bioactive, bio responsive and biomimetic materials, nanotoxicity, bio engineering and regenerative medicine, diagnostic devices such as sensors and image enhancement, therapeutic devices such as drug design and delivery, and noninvasive cancer treatment). Crosscutting topics included nanocatalysis (e.g., new developments in core–shell catalysts), properties (electronic, optical, and magnetic), hybrid nanoparticles, carbon nanostructures, polymer nanoparticles, and nanoparticle Ferro fluids. The Richmond symposium series is a standalone event that is always held in Richmond, Virginia. There are no parallel sessions so that the interaction between all of the participants can be maximized. In addition to invited talks, each session also features a “hot topics” presentation, which is selected from the contributed abstracts. This makes it possible to bring the latest breakthroughs to the participants. Although the focus of this symposium series changes each time, it always addresses cutting edge issues important to science and society. It attracts world leaders in the area of clusters and nanostructures, and nine Nobel Laureates have been associated with this series from its inception. More about ISCAN2019 can be found from the Web site https://iscan.vcu.edu/. Important issues and challenges in energy production and storage such as solar cells, thermoelectric materials, Li-, Na-, and K-ion batteries and the role catalysts play in converting harmful gases to usable fuels were discussed. Ways to overcome materials challenges in making safer and high-energy-density batteries, solar cells stable against moisture and intense heat, and thermoelectric materials with high ZT values were discussed, and the role of nanomaterials in addressing these challenges was brought into focus. In the nanobio field, talks covered application of nanomaterials for drug design and delivery, wireless electronic nanobio sensors for global health and security, nanomaterials for the treatment of cancers, detection of intracellular analytes using nanoparticle platforms, artificial muscles, cross-talk in the amyloid assembly of peptides and proteins, etc. Potential toxicity of nanoparticles in diagnostic and therapeutic devices remains an area of concern. The symposium also featured talks on the unique potential and challenges in synthesizing cluster-assembled materials and the role of ligands and substrates on their properties. One of the highlights of the symposium was the prediction and synthesis of a new form of metastable phase of carbon that is metallic and magnetic. While more experiments are needed to confirm all of the reported properties, the prospect of creating metastable phases with molecular precursors opens new possibilities for materials genome. This work is highlighted in an article published in Science https://www.sciencemag.org/news/2019/11/next-graphene-shiny-and-magnetic-new-form-pure-carbon-dazzles-potential. Views expressed in this Energy Focus are those of the author and not necessarily the views of the ACS. The author declares no competing financial interest. This article has not yet been cited by other publications. Figure 1. ISCAN2019 group photo. (Photo Courtesy: Thomas M. Kojscich)

We discovered a technical problem with one of our laboratory balances affecting the molal concentrations of the NaFSI solutions used for this study. We report the corrected data in Figure 1. The trend in conductivity vs molality now resembles that found for LiTFSI solutions (Figure 1a), with a maximum at ∼4 mol kg–1 (4m). Since the original publication of this study, we recorded the phase diagram for the H2O–NaFSI system and found that the room-temperature solubility of NaFSI is lower than that previously reported by us. It is in fact similar to that of LiTFSI (∼21m at 23 °C).(1) Hence, NaFSI solutions with concentrations > 21m are only metastable at room temperature. In light of the lower solubility and for easy comparison with 21m LiTFSI, the original water-in-salt electrolyte, we now include data for 21m NaFSI in Figure 1a–c. The narrow peak visible in the Raman spectrum of 21m NaFSI indicates that NaFSI solutions also have a high stability at this concentration (Figure 1b). The linear sweep voltammetry data shown in Figure 1c confirms that the stability of 21m NaFSI approaches that of 21m LiTFSI. Figure 1. Conductivity, structural characterization, and electrochemical stability of aqueous electrolytes based on NaFSI: (a) conductivity at 20 °C, (b) Raman spectra in the wavenumber region corresponding to the OH stretching modes of water, and (c) electrochemical stability on stainless steel evaluated using linear sweep voltammetry at a scan rate of 0.1 mV s–1. Cyclic voltammograms of NaTi2(PO4)3 and Na3(VOPO4)2F based electrodes measured in 35m NaFSI at a scan rate of 0.05 mV s–1 are also shown. The current densities for the active material measurements were scaled to fit to the electrolyte stability measurements. The thermodynamic potentials for the hydrogen and oxygen evolution reactions at pH = 7 are shown as vertical dashed lines labeled HER and OER, respectively. For comparison, data for aqueous LiTFSI solutions are also shown. We emphasize that the main conclusions of our paper, i.e., the much higher solubility of NaFSI (21m) vs NaTFSI (8m) and the extended electrochemical stability window of NaFSI of 2.6 V near the solubility limit are still valid. Figure 1 should appear as follows: This article references 1 other publications.

Views expressed in this Energy Focus are those of the authors and not necessarily the views of the ACS. The authors declare no competing financial interest. This article has not yet been cited by other publications.