The most commonly discussed ‘false positive’ source for biosignature gases is geochemical, primarily volcanism. It is quite difficult to determine whether a small amount of a gas could be produced by geochemistry. Earth’s bulk crustal geochemistry suggests that volcanism should outgas carbon exclusively as carbon dioxide; nevertheless hydrothermal systems can produce significant methane (Schwieterman et al., 2018) (although geochemical methane production is much less than biological production of methane on Earth). Therefore, we should be careful about dismissing potential sources based on planetary average properties. However, phosphine is a strongly reducing gas, and it takes quite unusual chemical circumstances to generate it from the phosphorus species expected to dominate in surface crustal rocks of rocky planets, which are P (+5) species (phosphate, phosphorus pentoxide or equivalents). The crust of Venus is known from the Soviet Venera and Vega lander missions to have an elemental abundance broadly similar to basalt (Koehler, 1982; Surkov et al., 1986; Garvin and Bryan, 1987). For this material to be able to reduce phosphate to phosphine, it would have to be 8–15 orders of magnitude more reduced than typical terrestrial crustal rocks (Bains et al., 2021b). Even the most reduced plausible surface rocks at the Iron/Wustite redox state (an equilibrium between metallic iron and FeO) would not favour phosphine production (Bains et al., 2021b). The volcanic production of phosphine would also require the rocks to have a high water content, which is unlikely given Venus’ desiccated surface and atmosphere. Bains et al. (2021b) concluded that volcanism derived from crustal rocks was extremely unlikely as a source for phosphine.

Truong and Lunine (2021) suggested that, contrary to Bains et al.‘s estimates, volcanic sources of PH₃ are plausible on Venus. Volcanic mechanisms focus on the delivery of P (+3) as phosphide in volcanic ejecta. Phosphides are known to react efficiently with dilute acids to form phosphine (Geng et al., 2010), although their reaction with concentrated sulfuric acid has not been studied; the precedent of the reaction of mineral sulfides with sulfuric acid suggests that concentrated sulfuric acid may oxidize phosphides rather than hydrolyse them, depending on the concentration of acid and other conditions (Prater et al., 1970; Bains et al., 2021b). Only rare, trace minerals contain phosphides on Earth, which argues against such a mechanism unless the geology of Venus is substantially different from Earth. However Truong and Lunine (2021) suggested that mantle plume volcanism could bring phosphorus from the deep mantle, where it is predominantly present as phosphide, to the surface. It is well known that small fragments of material a few millimeters in diameter can be dredged from the lower mantle by volcanism and appear as xenoliths in surface rocks on Earth (Sasse et al., 2020). However this potential possibility for phosphide delivery confuses two mechanisms (Bains et al., 2022b). Bains et al. (2022a) argued that the amount of phosphide surviving in the fluid magma as it makes the long, slow transit from lower mantle to crust would be trivially small. Truong and Lunine (2022) responded with the observation that reduced minerals are seen on Earth, but as noted these are all extremely rare materials, mostly present as xenoliths. Xenoliths are small fragments of minerals usually derived from the continental keel, the rigid cores of continental cratons that can reach depths of 350 km. Such xenoliths have survived as solid particles while being transported in upper mantle melts that have lower melting point than the xenolith mineral. Consequently, these highly refractory materials are very rare. The fact that xenoliths often contain high abundances of FeC or SiC does not mean that the melt contains high abundances of these minerals. For example, Truong and Lunine (2022) specifically mentioned the reduced minerals SiC and FeC found in the Luobusa ophiolite. However these are rare (and hence remarkable and publishable) xenoliths, recovered from over 1.5 tonnes of mineral that was processed to yield some 10s of kg of high density chromite, from which numerous mm sized inclusions of FeC, SiC, diamond and other highly reduced, highly refractile minerals were obtained (Huang et al., 2020). Such rarity is typical of xenoliths. This suggests an abundance of reduced species (which might contain phosphides, although no such species were reported) of <0.1%, which is in line with the calculations of Bains et al. (2022a).

We also note that no phosphide mineral has been reported in xenoliths to our knowledge. This may be because xenoliths have to be highly refractory mineral inclusions to survive in mantle melts; for example, iron phosphide (Fe₃P), with a melting temperature of ∼1200°C (American Elements, 2023), would not survive as a distinct mineral at the temperature of basaltic melts.

In conclusion, phosphides are extremely rare in crustal rocks, regardless of whether we calculate the abundance of phosphides in bulk rock from thermodynamics or estimate the presence of phosphide xenoliths. To produce sufficient phosphide to explain Venus’ phosphine, huge amounts of volcanism would be required, covering an area in the order of 2,100 km² per year with new magma (Bains et al., 2022a), which is greatly in excess of the ∼2.2 km² seen in an 8 month period by the Magellan spacecraft (Herrick and Hensley, 2023), and would resurface the whole planet in a few tens of thousands of years. In conclusion, we rule out volcanism as a source of phosphine on Venus.

The atmosphere of Venus is chemically complex, and not well understood (Petkowski et al., 2023). In principle, chemistry in the atmosphere could generate phosphine by reactions within the atmosphere, through photochemistry of atmospheric components, or through reaction of the atmosphere with solid materials on the surface or as dust. Surface reactions were deemed highly implausible as a source of phosphine by Bains et al. (2021b), a conclusion that has not been disputed. The most plausible reaction in the atmosphere would be disproportionation of P (+3) species, canonically represented as “phosphorus trioxide” (P₄O₆). Bains et al. (2021b) argued that the disproportionation of P₄O₆ was not a thermodynamically favourable mechanism to synthesize phosphine under Venus conditions. Omran et al. (2021) said that it was, and suggested that the P₄O₆ could be synthesised by reduction of P (+5) species by carbon monoxide. The two groups subsequently found that the difference in their conclusions came from differences in the ‘reference’ thermodynamic data on which their respective calculations were based, which were wildly different (Bains et al., 2023). However, the choice of reference data does not affect whether phosphine is predicted to be present on Venus. The data used by Bains et al. (2021b) predicted that P₄O₆ would be present in Venus’ lower atmosphere, but would not disproportionate to phosphine, the data used by Omran et al. (2021) suggested that P₄O₆ would disproportionate if it were present, but that P₄O₆ would not be present in Venus’ atmosphere. Bains et al. (2023) explored the effect of assuming different gas abundances, temperatures and thermodynamic ‘reference’ values, and found that there is no combination of conditions that allows phosphine to accumulate to >1 ppb in Venus’ atmosphere. The overall conclusion therefore is that a thermodynamically-driven gas phase reactions producing phosphine is implausible.

The routes for potential formation of phosphine on Venus discussed so far all assume gas phase chemistry in which the atmosphere is homogenous over a scale of meters. However, in the clouds there are at least two phases (gas/vapour and liquid droplets), and possibly solid material as well, and the abundance of minor species (such as SO₂ and H₂O) and trace species (such as the enigmatic “unknown absorber”) in the atmosphere vary substantially with altitude over distances of 100s of meters or more.

The composition of the cloud droplets or particles is not well defined. They are usually assumed to be concentrated sulfuric acid, but the VeGa 2 descent probe found high levels of a phosphorus species in some particles in the lower clouds (reviewed in Titov et al. (2016)). Moreover, measurements performed by Pioneer Venus probe suggest the larger cloud particles are not spherical (Knollenberg and Hunten, 1980), which if confirmed means they cannot be liquid. Therefore, there is substantial latitude for models that explain the production of phosphine from phosphorus species either in cloud particles or at the interface of those particles and the atmosphere.

However, as explored in detail by Bains et al. (2021b) production of phosphine within cloud particles in the absence of an external energy source to drive the reaction seems unlikely. The rapid reaction of phosphine with concentrated sulfuric acid has been known since the 19th Century, when it was used as a means of removing phosphine from acetylene used in gas lighting (Retgers, 1894; Leeds and Butterfield, 1910). This occurs rapidly at European ambient temperatures, i.e., <20°C, and would be expected to happen faster at the elevated temperatures of the lower clouds of Venus. The end product of the reaction is not known, but it is not volatile and hence is presumably phosphorus oxides or oxyacids. Therefore, the idea that reactions occurring in the presence of concentrated sulfuric acid would be nett producers of phosphine seem implausible.

Thermodynamic constraints on the formation of phosphine can be overcome if an external energy source drives a reaction. The most abundant energy source in Venus’ clouds is light (as there is little energy available in the redox disequilibria in the clouds (Jordan et al., 2022)). Potential photochemical routes to phosphine synthesis were not exhaustively explored in Bains et al. (2021b), except for an approximate gas phase kinetic model. One group has modelled potential photochemistry occurring on the acid surface of dust in the clouds as a potential source of phosphine on Venus (Kaiserová, 2021; Ferus et al., 2022). Other reduced gases, like methane and ammonia can potentially be produced photochemically under such conditions. Ferus et al. (2022) and Kaiserová (2021) argue on the basis of quantum mechanics modelling of the relevant reaction intermediates that phosphine could be made as well, via reduction of PO (generated by photochemical breakdown of H₃PO₄) by HCO radicals (generated by reduction of CO₂ on the acidic dust surface). This intriguing hypothesis would explain the detection of phosphine in the clouds, and is eminently testable with Earth-based experiments. However the mechanism by which dust can be lofted from the surface to the cloud layer remains to be explored.

A second route is the photochemical reduction of P₄O₆ in the clouds, which has been modelled by Wunderlich et al. (2023). This route is only valid if P₄O₆ actually exists in the atmosphere at the altitude of the clouds, which Bains et al. (2021b) and Omran et al. (2021) concluded that it did not. The mechanism analysed by Wunderlich et al. (2023) assume an efficient net production of PO from P₄O₆ at rates comparable to the production of methane on modern Earth. They do not suggest a mechanism for this unprecedented chemistry, and indeed thermodynamic calculation suggests that it is extremely unlikely (Bains et al., 2021b). Wunderlich et al. (2023) also assume that the kinetics of reaction that break the phosphorus-oxygen double bond could be modelled with reference to the kinetics of analogous chemistry of the nitrogen-oxygen double bond, a parallel that Bains et al. (2021b) has shown is very inaccurate. However Wunderlich et al. (2023)’s overall conclusion, that the quantitation of the phosphorus photochemistry in Venus’ atmosphere is uncertain by many orders of magnitude, and that more experiments and observations are essential for resolving whether PH₃ can be explained abiotically, is consonant with the conclusions of Bains et al. (2021b, 2023) and Omran et al. (2021).

Given these uncertainties, a photochemical source therefore remains an open option as a source of Venusian phosphine. We note that McTaggart (2022) explores whether cosmic rays could power the production of PH₃ in Venus’ upper clouds, and conclude that they could not.

Other considered sources of Venusian PH₃ have included lightning (Bains et al., 2021b) and meteoritic infall (Bains et al., 2021b; Omran et al., 2021). The mineral schreibersite ((Fe,Ni)₃P) reacts efficiently with aqueous sulfuric acid to form phosphine (Geng et al., 2010); its reaction with concentrated sulfuric acid has not been explored to our knowledge. Schreibersite is found in meteorites, and so meteoritic infall could be a source of phosphine through this reaction. Bains et al. (2021b) estimated that the rate of infall was insufficient to explain phosphine, and Omran et al. (2021) estimate the infall rate of small particles and dust falls short of the required rate by at least 10-fold. However, Omran et al. (2021) also suggested that a single, large impactor could deliver phosphide: a 10¹² kg object would ablate primarily 30–60 km above the surface, and would not form any significant crater. Such an object would be a one in a million year event, and so it is unlikely that Greaves et al. were observing soon after such an impact (the half-life of phosphine is 100s of years in the deeper clouds, reducing to minutes or seconds above the clouds.) It is also improbable that the Akatsuki spacecraft, which has been studying Venus’ atmosphere since 2015, would not have seen such an event. Meteoritic infall, either constant or a single large event, is therefore unlikely as a realistic scenario for phosphine production on Venus.

The lower clouds of Venus may contain substantial phosphorus (Titov et al., 2018), unlike the clouds of Earth, and so lightning passing through those clouds could be a source of phosphine. Bains et al. (2021b) argued on semi-quantitative grounds against lightning as a source. We can also note that lightning on Earth produces no detectable ammonia (the nitrogen analogue of phosphine) despite clouds on Earth being made of water vapour and the atmosphere being 80% nitrogen. Lightning chemistry strongly favours oxidized species rather than the more labile hydrides as products; essentially all the nitrogen compounds made by lightning on Earth are nitrogen oxides (Schumann and Huntrieser, 2007).

By far the most contentious candidate source for the phosphine is the idea that the cloud droplets are inhabited by microorganisms that are making it (Benner, 2021). Bains et al. (2021b,c) showed that biochemical production of phosphine was not thermodynamically impossible, based on a model where an organism maintained a neutral aqueous interior while living in a sulfuric acid droplet, but did not speculate where the energy to do this might come from. Jordan et al. (2022) show that some limited sulfur-based metabolic strategies either cannot provide enough energy, or would produce chemical signatures that contradict what is currently observed in Venus’s atmosphere, but Bains et al. (2024) show that more than enough solar energy is available for a biosphere in the clouds, and alternative photosynthetic metabolisms may not create the same anomalies in Venus’s atmosphere as the metabolisms explored by Jordan et al. Such an organism is therefore not ruled out by what we know of life, but nevertheless seems intuitively implausible.

The principle argument against life being a source of phosphine is evolutionary, not chemical. To make phosphine from the P (+5) compounds that are likely to be the dominant phosphorus chemical species in the clouds, the putative organism has to capture substantial energy, source hydrogen atoms from the cloud droplet, and perform the highly endothermic overall transformation H 3 PO 4 + 8 H →   PH 3 + 4 H 2 O

Hydrogen-bearing species are rare in the atmosphere of Venus, so finding eight hydrogen atoms per phosphine molecule produced would require investment in energy. The reaction is highly endergonic, requiring energy to reduce P (+5) to P (-3). Having invested substantial energy in making PH₃, life then allows it to diffuse away into the clouds. There seems to be no evolutionary rationale for doing such a wasteful process (Benner, 2021), unlike on Earth where PH₃ production is of thermodynamic benefit in a highly reducing environment (Bains et al., 2019a). This may be a failure of our imaginations, but absent any rationale for spending so much resource of making a product, an evolutionary biologist would be highly skeptical that life is the source of phosphine.