Next Article in Journal
Inclusive AI-Mediated Mathematics Education for Students with Learning Difficulties: Reducing Math Anxiety in Digital and Smart-City Learning Ecosystems
Previous Article in Journal
Virtual Reality as a Potential Cornerstone for Remote Rehabilitative Therapies
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Geochemical Modeling from the Asteroid Belt to the Kuiper Belt: Systematic Review

by
Arash Yoosefdoost
and
Rafael M. Santos
*
School of Engineering, University of Guelph, Guelph, ON N1G 2W1, Canada
*
Author to whom correspondence should be addressed.
Encyclopedia 2026, 6(2), 38; https://doi.org/10.3390/encyclopedia6020038
Submission received: 14 October 2025 / Revised: 17 January 2026 / Accepted: 28 January 2026 / Published: 3 February 2026
(This article belongs to the Section Earth Sciences)

Abstract

The high costs and time-consuming nature of space exploration missions are among the major barriers to studying deep space. The lack of samples and limited information make such studies challenging, highlighting the need for innovative solutions, including advanced data-mining techniques and tools such as geochemical modeling, as strategies for overcoming challenges in data scarcity. Geochemical modeling is a powerful tool for understanding the processes that govern the composition and distribution of elements and compounds in a system. In cosmology, space geochemical modeling could support cosmochemistry by simulating the evolution of the atmospheres, crusts, and interiors of astronomical objects and predicting the geochemical conditions of their surfaces or subsurfaces. This study uniquely focuses on the geochemical modeling of celestial bodies beyond Mars, fills a significant gap in the literature, and provides a vision of what has been done by analyzing, categorizing, and providing the critical points of these research objectives, exploring geochemical modeling aspects, and outcomes. To systematically trace the intellectual structure of this field, this study follows the PRISMA guidelines for systematic reviews. It includes a structured screening process that uses bibliographic methods to identify relevant studies. To this end, we developed the Custom Bibliometric Analyses Toolkit (CBAT), which includes modules for keyword extraction, targeted thematic mapping, and visual network representation. This toolkit enables the precise identification and analysis of relevant studies, providing a robust methodological framework for future research. Europa, Titan, and Enceladus are among the most studied celestial bodies, with spectrometry and thermodynamic models as the most prevalent methods, supported by tools such as FREZCHEM, PHREEQC, and CHNOSZ. By exploring geochemical modeling solutions, our systematic review serves to inform future exploration of distant celestial bodies and assist in ambitious questions such as habitability and the potential for extraterrestrial life in the outer solar system.

1. Introduction

Geochemistry is a field that integrates geology with chemical concepts and methods to uncover the mechanisms of major geological systems such as the Earth’s crust and seas [1]. Evidence suggests that the roots of this field date back to 1838, when the term ‘geochemistry’ was used instead of chemical geology by Christian Friedrich Schönbein [2]. Geochemistry has been recognized as a distinct field of science since the United States Geological Survey (USGS) conducted systematic investigations of rock and mineral chemistry in 1884 [2,3]. Geochemistry’s domain ranges even to space objects, including the Solar System, and has significantly contributed to expanding the knowledge of various processes in space, including planet formation [4] and cosmochemistry.
Cosmochemistry is the science of the chemical composition and processes in the universe, from the formation of the first solid bodies to the evolution of stars, galaxies, and planetary systems. It aims to understand the origins and evolution of the chemical elements, isotopes, and minerals in the universe. Cosmochemistry has been known as a distinct scientific field since 1901. However, evidence suggests that the roots of cosmochemistry may date back approximately one decade after the distinction of geochemistry; in the early 1850s, the meteorite composition was studied and compared with that of the rocks of Earth [2,5]. Space geochemistry is a subsection of cosmochemistry that focuses on the geochemical composition and processes of astronomical objects (such as planets, moons, asteroids, comets, and meteorites) to understand the origin and evolution of the chemical and isotopic compositions of these bodies, as well as the processes that have shaped their surface and interior over time.
Geochemical modeling, which is oversimplified and not exact, can be defined as a mathematically based method to explore how natural processes affect the material chemistry of a celestial body (a space object such as a planet or moon) over time and space. For example, how rocks change due to weathering, volcanic activity might affect the atmosphere, potential life-sustaining compounds may form, etc.
Geochemical modeling, also known as theoretical geochemistry, is the practice of analyzing chemical processes that impact geologic systems via chemical thermodynamics, chemical kinetics, or both, and employs mathematical equations to characterize chemical and transport processes in a geological system to make predictions that are partially visible or empirically verifiable [6]. This tool is utilized for interdisciplinary studies such as those on environmental geochemistry, petroleum geology, and geothermal and hydrothermal fluids [7,8]. Geochemical modeling also supports cosmology and cosmochemistry in studying the geochemistry of other planets and moons. The main goal of geochemical modeling is to understand the processes that govern the composition and distribution of elements and compounds in a system and to estimate the geochemical conditions on other planets and moons. Geochemical models can be used to simulate the evolution of a planet’s or moon’s atmosphere, crust, or interior, and to predict geochemical conditions at the surface or in the subsurface.
Geochemical models quantitatively analyze chemical reactions in geological systems by solving mathematical equations of thermodynamics, kinetics, mass balance, and fluid dynamics models, and assisting in predicting future events and scenarios [9]. Equilibrium, kinetic, and thermodynamic models are among the geochemical models that can be used to study space geochemistry. Equilibrium models assume that a system is in equilibrium and that the concentrations of all species are constant over time. Kinetic models consider the time dependence of reactions and the change in species concentrations over time. Thermodynamic models are based on the laws of thermodynamics and describe the behavior of a system in terms of the free energy of the species and the temperature and pressure conditions.
Given the technical barriers and the costs of space exploration, direct analysis of many astronomical objects seems unlikely in the near future [10]. Despite the number of geochemical modeling studies on the Moon and other celestial bodies, even as far as Mars, the number of studies conducted on space objects located beyond Mars is insignificant. Fewer space exploration missions about these objects lead to a lack of samples and less available information, making the geochemical modeling of such studies novel but much more challenging. Therefore, this work focused on studies of geochemical modeling of space objects beyond Mars to provide a vision of what has been done thus far by analyzing, categorizing, and providing the critical points of these research objectives, exploring geochemical modeling aspects and findings.
More precisely, this systematic review focuses exclusively on geochemical reaction modeling—thermodynamic and kinetic models of fluid–solid–gas interactions and aqueous and nonaqueous fluid speciation—applied to bodies beyond Mars, including icy moons and Kuiper Belt Objects (KBO). This systematic review intentionally restricts inclusion to studies that perform geochemical reaction modeling (thermodynamic/kinetic) of volatile–ice–brine and related solid–fluid systems on bodies beyond Mars; atmospheric circulation/photochemistry, accretion/evolution, and purely geophysical models are outside the scope unless they explicitly implement such reaction modeling.

2. Methods and Materials

2.1. Solar System and Astronomical Objects Beyond Mars

The objects in our solar system can be broadly classified into two categories: eight planets and smaller bodies. According to the 2006 International Astronomical Union (IAU) Resolution, the eight planets listed in order from the sun are Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune. Some notable dwarf planets in our solar system include Pluto, Ceres, Eris, Haumea, and Makemake [11]. The smaller bodies include dwarf planets, moons, asteroids, comets, meteoroids, and trans-Neptunian objects (TNOs). The solar system’s planets, moons and some of the major objects are represented in Figure 1, and some details of the solar system’s planets and the most famous dwarf planet are summarized in Table 1.
According to IAU [11], a planet is defined as a celestial body that is in orbit around the Sun and has sufficient mass to assume hydrostatic equilibrium (a nearly round shape) and a cleared neighborhood around its orbit (i.e., no other similar-sized objects in its orbital path). A dwarf planet is a celestial body that orbits the sun and has sufficient mass to achieve a nearly round shape, but it shares its orbital neighborhood with other objects and has not gravitationally cleared its orbit of debris, such as asteroids or smaller objects. A moon is a natural satellite that orbits a planet, a dwarf planet, or another celestial body and is held in place by gravity. The diameter of a moon can vary in size from a few meters to several thousand kilometers. Notably, moons can have their own unique characteristics, such as different atmospheres, geological features, and even the potential for hosting life.
The asteroid belt of Mars and Jupiter contains numerous asteroids ranging in size from less than one to several hundred kilometers in diameter. Comets are icy bodies that originate from the outer solar system and have highly elliptical orbits that bring them close to the sun. Meteoroids are small, rocky or metallic fragments that originate from comets or asteroids and can range in size from tiny dust particles to large boulders. TNOs are a diverse group of small bodies that orbit the sun beyond Neptune and include objects such as dwarf planets, plutinos, and cubewanos.
To investigate the geochemical modeling of celestial bodies in the solar system beyond Mars, a list of 180 celestial objects was generated and employed as search terms. This list is represented in Table 2.

2.2. Geochemical Modeling and Celestial Objects

Geochemical modeling is a critical tool for understanding the processes that govern the composition and behavior of rocks, minerals, and fluids in the Earth and celestial bodies. It involves creating a computational representation of the chemical and physical interactions that occur within a specified geological context. While the application of geochemical models to celestial objects is a narrow area of study, various factors compound the difficulties in finding the relatively limited volume of directly relevant studies in this specific area for the purpose of reviewing the literature. The diverse array of terminologies and potential publishing outlets makes the task of identifying all relevant research particularly arduous. An additional layer of complexity is that many studies do not directly utilize the term “geochemical modeling” (or “geochemical model*” in truncated form as used in bibliographic searches). Instead, they may employ alternative terminologies, reference specific model names, or focus on individual processes or phenomena pertinent to the broader realm of geochemical modeling. This broad and varied use of language further complicates the process of pinpointing relevant research. Moreover, the general nature of some associated keywords, such as “thermodynamics” or “kinetics”, can yield an overwhelming number of search results, many of which may not be directly relevant to the geochemical modeling of celestial bodies.
To navigate these multifaceted challenges, a number of strategies can be employed. One practical solution lies in the careful crafting of a comprehensive list of relevant keywords and phrases, which include but are not limited to fundamental concepts, specific models and tools, key geochemical processes, advanced concepts, and relevant application areas, such as those listed in Table 3, that serve as valuable resources for researchers. The investigations could be narrowed down by limiting the results to those focused on celestial bodies represented in Table 2. However, even such an approach may not be specific enough to filter out unrelated works or identify a large enough fraction of the related published works. To this end, a computational method for more precise bibliographic screening, the workflow participation of which is illustrated in Figure 2, is developed and presented in Section 2.3.

2.3. Bibliographic Screening and Analysis

We employed a descriptive bibliographic approach to structure the literature, focusing on organizing existing information and identifying thematic patterns, which is distinct from complex bibliometric mapping and network analysis techniques [17,18]. We utilized Web of ScienceTM (WOS) to search for relevant papers and screen for the topic of our systematic review.
The most relevant studies were identified via traditional searches in WOS, and the resulting bibliographic data were then analyzed via computational methods. The multistep nature and recurring demand for various software, which are not essentially free and accessible for all researchers, make science mapping complex and challenging [19]. To mitigate this for our initial data handling, the Bibliometrix open-source package (available on CRAN) developed by Aria and Cuccurullo [19] was utilized for performing the initial data import and processing. The structure of scientific fields was initially visualized using Author Keywords and Keywords Plus as units of analysis. Author keywords are those given by the publication’s author(s), whereas Keywords Plus refer to a set of words or phrases that are typically found in the titles of the cited references in an article via computer algorithms [20,21]. Studies suggest that Keywords Plus could be effective in capturing scientific concepts presented in articles since it produces more descriptive terms and a broader range of results, whereas it could be less comprehensive in representing the content of an article [22].
The bibliographic screening of the identified publications on the geochemical modeling of space objects beyond Mars in the solar system assisted in identifying more than 900 related studies. A preview of the relationships among these publications, the main keywords, and Keywords Plus is presented in Figure 3.
According to Figure 3, the significant number of items in such a large and complex Sankey diagram makes it difficult to analyze for humans, and even working with such a complex diagram could be resource intensive for common computers at the time of publishing this paper. Moreover, although “geochemical modeling” is listed as the top keyword (middle), the desired and target keywords are not observed in a large portion of the identified studies. At the time of this study, Bibliometrix did not offer a solution to address such concerns and to narrow the analysis to the desired range. For this study, custom tools were developed for identifying the most relevant publications on the research topic, highlighting related studies, and focusing specifically on them via programmatic bibliographic screening and filtering. These tools are released as an open-source toolkit package named Custom Bibliometric Analyses Toolkit (CBAT) and are distributed through https://github.com/Yoosefdoost/Custom_Bibliometric_Analyses_Toolkit (accessed on 27 January 2026) under the AGPL-3.0 license. The systematic review protocol was preregistered on the Open Science Framework (OSF) at https://osf.io/7fn63 (accessed on 27 January 2026).
This toolkit integrates keyword extraction, targeted thematic mapping, and visual network representation of bibliographic records, specifically focusing on advanced screening and data refinement. Its modular system comprises three main tools, each serving a specific analytical function and designed to operate on BibTeX files commonly exported from reference management systems such as Zotero, EndNote, or Web of Science.
Common Keywords Extractor: This module processes the input. bib file and extracts all keyword fields across entries. It then standardizes the terms (lowercasing, trimming whitespace, etc.) and computes term frequencies. The output consists of two CSV files: one listing all keywords and their counts, and another sorted in descending order of frequency. This step provides a quantitative overview of recurring topics and helps inform the selection of thematic keywords for further bibliographic screening and analysis.
Target Keywords Analyzer: Based on a user-defined list of research themes or domain-specific terms, this module evaluates the presence of each term in the title, abstract, and keyword fields of each bibliographic record. The output includes two CSV files: (a) a full report listing each reference, formatted in APA style, along with binary indicators or counts for each target keyword; and (b) a matrix file representing references versus keywords, suitable for visualization. This component enables focused thematic profiling of literature collections and is especially useful for systematic screening workflows and subsequent meta-analyses.
Sankey Diagram Generator: Using the matrix file generated in the previous step, this script produces an interactive Sankey diagram via the Plotly library (version 5.0 or later). Each flow represents the association between a reference (source) and a target keyword, with optional color customization and configurable visual properties. The resulting diagram supports intuitive exploration of topical patterns and co-occurrence across the literature dataset.
Together, these tools enable a transparent and adaptable pipeline for computational bibliographic analysis that prioritizes rigorous screening and filtering and can be scaled to collections of varying size and complexity. All the scripts are open-source and designed to run in any standard Python environment (version 3.7 or later), requiring only lightweight dependencies (bibtexparser, csv, and plotly). The toolkit can be adapted for different fields of study by modifying the target keyword list and input bibliography, making it a valuable resource for both exploratory research and structured academic reviews.
This systematic review was conducted in accordance with the PRISMA 2020 guidelines [23] to ensure transparency and reproducibility. The PRISMA checklist, which maps each guideline item to its location in the manuscript, is provided in the Supplementary Materials. The checklist complements the detailed description of our screening workflow presented above and documents how the review adhered to established standards for systematic literature reviews.
The identified studies on the geochemical modeling of space objects beyond Mars in the solar system and their relationships with the desired keywords are represented in Figure 4.
In contrast to traditional keyword-based bibliographic searches that primarily seek exact keyword matches, our reinforced bibliographic screening analysis targets additional terms related to the desired concept. Analyzing publications’ titles and abstracts in addition to author keywords and focusing on target keywords could offer results closer to a semantic search. For example, for ‘geochemical modeling’, terms such as ‘equilibrium’, ‘kinetic’, ‘thermodynamic’, and ‘geochemistry’ may refer to the same underlying concept, more effectively expressing the main intent, desired context, and conceptual meanings. The analysis of the publications represented in Figure 4 suggests that although the ‘geochemical model’ is directly mentioned in more than 11.11% of the identified studies, publications that focused on ‘thermodynamic’ (7.41%), ‘equilibrium’ (7.41%), ‘kinetic’ (2.7%), and ‘geochemistry’ (4.32%) could have significant potential. More investigations have indicated that, with a total percentage of over 51.61% in identified studies, Saturn’s moons, Enceladus and Titan, and Jupiter’s moon, Europa, are the most studied celestial objects (each over 25.80%), with Ceres in the next place, accounting for more than 12.90%. Neptune’s moon Triton and Pluto’s moon Charon are also among the studied astronomical objects.

3. Literature Review of Geochemical Modeling Studies

Following the PRISMA-compliant screening process (Figure 2), we included 41 studies from 903 records identified in WoS. Across Section 3 and Section 4, we cite these 41 included studies together with 32 supporting sources (methods/model papers and mission/overview items), totaling 73 unique references in these sections.

3.1. Ceres

Once designated an asteroid, Ceres (Figure S1) was reclassified to a dwarf planet in 2006 because of its significant size and unique characteristics. Located between Mars and Jupiter, Ceres is the largest object in the asteroid belt, the only dwarf planet within the inner solar system, and the first dwarf planet explored by a spacecraft (NASA’s Dawn) in 2015 [24,63]. Formed approximately 4.5 billion years ago, Ceres is an “embryonic planet” that did not fully develop owing to Jupiter’s gravitational influence. Its structure is similar to that of terrestrial planets (Mercury, Venus, Earth, and Mars), featuring a layered interior with a likely solid core and a mantle made of water ice. Its crust is rocky and dusty and is enriched with large deposits of salts. Ceres has a thin atmosphere with evidence of water vapor, which is likely produced by ice volcanoes or ice sublimation. However, a magnetosphere is not believed to exist on Ceres. Although Ceres’s radius is approximately 7.69% of Earth’s radius, it may be made from up to 25% water, exceeding all Earth’s water quantity, which makes it a potential location for alien life (although microscopic, like bacteria) in our solar system [63].
Ceres’ volatile-rich interior and long history of water–rock interactions make it an ideal candidate for geochemical modeling. Dawn mission constraints, including indications of remnant brines, hydrated minerals, and possible clathrate-rich crustal layers, provide boundary conditions for equilibrium and reaction-path models that explore carbonate and sulfate chemistry at cryogenic temperatures and moderate pressures [25,64]. These models are well-suited for testing hypotheses about brine preservation, carbonate speciation, and organic formation in differentiated bodies with relict oceans while highlighting data gaps in thermodynamic properties near eutectic points and in the low-temperature kinetics of water–rock reactions [65].
Castillo-Rogez and McCord [66] conducted a comprehensive literature review and utilized various methods to explore and analyze the geochemical aspects of Ceres, including silicate chemistry, ocean chemistry, ammonia, organic chemistry, long-lived radioisotopes, and long-term evolution. For silicate chemistry, Castillo-Rogez and McCord [66] explored the potential of CO2-involved reactions involving serpentine, olivine, and plagioclase, which can produce a variety of minerals. They also examined whether silicate core internal heating could reach temperatures high enough to dehydrate hydrated silicates. The results suggest that these reactions could lead to the formation of minerals such as talc, magnesite, orthopyroxene, hornblende, and spinel.
Exploring long-lived radioisotopes, Castillo-Rogez and McCord [66] suggested that silicate alteration would lead to the depletion of key radioactive elements as a result of low-temperature mobilization, which has been identified as a probable mechanism for explaining the scattering in the Th/U ratios measured at various carbonaceous chondrites (Cc). The long-term evolution of the core and hydrosphere is considerably influenced by their respective compositions following differentiation. For example, silicate hydration is not only a major heat source, but also the permeability of the core is altered by hydrated silicates, which could lead to the transfer of key elements such as long-lived radioisotopes to the hydrosphere from the silicate core.
Figure 5 represents evolutionary paths for Ceres’ core as a function of initial conditions and the possible extent of hydrothermal circulation. According to this figure, for formation timeframes less than five million years after the formation of calcium-aluminum-rich inclusions (CAIs), it is possible that a substantial amount of the silicate phase is hydrated as a result of hydrothermal activity induced by 26Al decay in planetesimals and/or the accreting planet. Then, the presence of hydrothermal minerals may have influenced core permeability and constrained hydrothermal cooling. As a result, hydrated minerals play various roles in heating the core. The temperatures might be high enough for dehydration, even the melting of silicate. Longer formation timeframes are less likely to harbor early conditions favorable for silicate hydration, which makes it more probable that few silicate minerals were hydrated by the accretion end. In this scenario, hydration of silicates might take place afterwards because fractures form through cooling of the core.
To explore ocean chemistry, they employed a theoretical model for the chemistry of icy satellites to indicate that sulfates, hydrates, carbonates, and chlorides of magnesium, potassium, and sodium are the primary accumulating components in the ocean. Furthermore, a hydroxide layer can be deposited at the surface of the Ceres core. These authors also suggested that clathrate hydrates could be stable under Ceres’ conditions and could store methane, H2S, argon, etc. [66].
Castillo-Rogez and McCord [66] also explored reactions involving ammonia, which could lead to the production of molecular nitrogen, ammonium salts, and/or amino acids. The existence of an ammonium compound might also assist in constraining model parameters because secondary temperatures less than 400 K inside Ceres are necessary, at least in some regions. The authors also interpreted the quantity of organic compounds in Cc as evidence of geochemistry in the parent meteorite. Carbon species are expected to lead to complex chemistry. The hydrothermal activity of ice–rock objects could pave the way for the formation of carbonates, as suggested by Europa’s chemical models.
Thermal evolution is a key aspect of the geochemical modeling of Ceres because of its influence on the geochemical processes that occur within Ceres, such as the hydration and dehydration of silicates, the potential presence of ammonia, and the role of long-lived radioisotopes. Castillo-Rogez and McCord [66] studied thermal evolution models for Ceres. The models show temperature changes over time and across Ceres’ radius. The models consider different times of formation after the creation of calcium–aluminum–rich inclusions (CAIs), a key event in the early solar system. The model also considers the composition of certain elements in Cc, a type of meteorite, to compute the thermal evolution of the core, which influences these geochemical processes [66]. Castillo-Rogez and McCord [66] also discussed the implications of their findings. These findings suggest that extensive liquid water was likely present inside Ceres during its evolution. Prospective observations from the Dawn mission help constrain the character of Ceres’ interior.
Launched in 2007, NASA’s Dawn mission aimed to explore the protoplanets Vesta and Ceres, remnants of the early solar system. The composition and structure of spacecraft, which are equipped with various instruments, such as a framing camera, a visual and infrared spectrometer, and a gamma-ray and neutron detector, have been studied. From 2011 to 2012, Dawn orbited Vesta, and from 2015 to 2018, Ceres. For geochemical modeling purposes, this mission led to valuable information, such as confirming Vesta as the source of howardite-eucrite-diogenite (HED) meteorites. At Ceres, evidence of water ice and a potential subsurface ocean was found, significantly contributing to our understanding of its geochemical evolution.
To explore how Ceres has changed and developed over time, Castillo-Rogez et al. [25] introduced a new model for the internal evolution of Ceres with high agreement with Dawn’s mission observations. This study used a one-dimensional thermal modeling approach and incorporated several geochemical aspects, including the redistribution of potassium, the freezing of the ocean, and the preservation of brines in Ceres’ mantle.
Considering the observations of the Dawn mission, Castillo-Rogez et al. [25] assumed that the crust was dominated by volatile substances and that the rocky mantle contained remnant brines. The results suggest the possibility of a current warm crust containing a high volume of clathrate hydrates. For many scenarios, the crust’s base estimated temperature exceeds −53.15 °C (220 K), allowing for the retention of a modest quantity of brines there. They proposed that a small amount of these brines could be preserved at the base of the crust if the crust is rich in clathrate hydrates (compounds where water forms a cage-like structure around another molecule, usually a gas). Nevertheless, it requires a minimum temperature of −23.15 °C (250 K), and over 55% of these clathrate hydrates are present. However, the presence of more than 55 vol.% clathrate hydrates in the crust is essential for maintaining a minimum temperature of −23.15 °C (250 K) to harbor 1 wt.% sodium carbonate in solution, which is a limited situation in evolutionary scenarios. On the basis of geochemical predictions and geophysical observations (Figure S2), a current relict ocean in Ceres is expected if the crust has more than 40% insulating minerals such as hydrated salts and clathrates.
Neveu et al. [26] proposed the explosive ejection of cold material from the interiors of icy celestial bodies, which should be observed on Ceres, Enceladus, Europa, and Triton. They studied this phenomenon by developing a novel geochemical equilibrium model that bridges geophysics and geochemistry to explore the realm of subzero-temperature liquid water. More details are presented in Section 3.6 (Pluto).
Using geochemical modeling, Diab et al. [24] explored the bulk composition and thermal evolution effects on Ceres’ subsurface ocean organics. Considering that significant quantities of carbon and organic compounds on the surface of Ceres were found through the Dawn mission, they developed a thermodynamic model to constrain the speciation, stability, and abundance of organic compounds produced through abiotic reactions in Ceres’ mud-bearing mantle and primordial subsurface ocean. To evaluate the variables contributing to the speciation, abundance, and optimum production of organics, they adjusted environmental parameters, including temperature, pressure, pH, solution components, and redox state.
Carbonaceous Ivuna-type (CI) chondrites are extremely rare meteorites and are known to be rich in volatile elements and water. They are characterized by a very close match to the solar system’s elemental abundance pattern, making them valuable for studying early solar systems. Diab et al. [24] utilized thermal profiles of ocean depths produced from previous thermal evolution models. Their preliminary findings suggest that local organic formation within the subsurface ocean is mostly insignificant if Ceres is primarily formed from CI Cc’s. However, the scenario substantially changes if Ceres originates from a commercial material and leads to considerably more organics, including alcohols, amines, and carboxylic acids, due to a higher carbon and hydrogen content and lower oxygen content than the chondritic composition (Note S1). They highlighted the significance of temperature and pH as major influencing factors in organic formation through water–rock equilibrium, with temperature exerting the most profound impact. They also explored the relative molal abundances of the formed carbon-bearing species as a function of the mass percentage of the mixed CI chondrite composition according to the cometary composition (Note S2). The authors concluded that a mixed source might explain Ceres’ surface organics since neither pure CI chondritic nor pure cometary compositions could.
Across the reviewed Ceres studies, objectives converge on constraining internal thermal evolution and brine chemistry, typically using equilibrium thermodynamics with Dawn-derived inputs. Differences arise from the assumed bulk composition and redox state, with CI chondrite versus cometary endmembers leading to divergent predictions for organic speciation and carbonate abundance. These findings consistently support the plausibility of relict brines and carbonate buffering while emphasizing uncertainties in low-temperature datasets and kinetics that limit definitive conclusions on present-day aqueous activity [25,65].

3.2. Jupiter

Jupiter (Figure S3) is the fifth planet from the Sun and the largest planet in our solar system; it is estimated to be twice as massive as all other planets combined. Its atmosphere consists mainly of hydrogen (H2: 89.8% ± 2.0%) and helium (He: 10.2% ± 2.0%) [67,68]. While lacking an Earth-like surface, its core may be around Earth’s size if it is solid. Jupiter has a strong magnetosphere with an 800 million km magnetic tail [67,68], leading to a complex system of moons, rings, and asteroids. It has an estimated 80–95 moons; owing to their number, small moons are named only if they are scientifically significant [15]. The largest moons by orbit proximity are Io, Europa, Ganymede, and Callisto (Figure S4).
Ganymede is Jupiter’s largest moon and the largest in our solar system, larger than Mercury and Pluto, and approximately 75% size of Mars. Its magnetic field makes it unique among moons. Ganymede’s metallic iron core generates this field and is surrounded by rocky and icy layers. It has a thin oxygen atmosphere, but it is not sufficient for human respiration. A subterranean saltwater ocean may contain more water than Earth’s surface water [15,69].
Callisto is Jupiter’s second-largest moon and third-largest in our solar system at approximately 38.46% Earth’s size. Its thin exosphere contains carbon dioxide, hydrogen, and oxygen. Spacecraft images suggest the presence of ice on Callisto’s surface, which has more impact craters than any other object in our solar system. A salty ocean beneath its icy surface could harbor life.
Io is slightly larger than Earth’s moon, with a diameter of approximately one-fourth that of Earth. One side always faces Jupiter due to tidal lock. Its thin atmosphere mainly consists of sulfur dioxide (SO2). Io is our solar system’s most volcanically active object with lakes of molten silicate lava; some volcanoes erupt tens of kilometers high lava, making it inhospitable for known life forms [15,70].
Europa is smaller than the Earth’s Moon, with one quarter of the Earth’s diameter, but has a thin oxygen atmosphere that is insufficient for human respiration. An iron core, the rocky mantle, and saltwater ocean make Europa an intriguing target for the search for life beyond Earth, as its icy surface likely harbors a subsurface ocean twice Earth’s ocean volume, which interacts with rocky surfaces since solar system formation provides ample time for the emergence of life [15,71]. Several studies have focused on the geochemical modeling of Jupiter’s famous moon Europa.
Europa’s ocean is anticipated to reside under its thick icy shell and is among the most likely alien habitable habitats or existing life in the solar system [72]. However, owing to the technical barriers and expenses of space exploration, direct analysis of this ocean seems unlikely in the near future [10]. However, various studies have attempted to infer Europa’s ocean chemical composition by analyzing its icy surface via remote sensing, landed surface missions [10], and other approaches, such as geochemical modeling.
The Jupiter system presents diverse physicochemical settings for geochemical modeling, from high-pressure ocean environments in icy moons to sulfur-rich volcanism on Io. Equilibrium and reaction path models that incorporate pressure effects, multicomponent brine chemistry, and water–rock interactions are suitable for evaluating ocean compositions, freezing sequences, and mineral precipitation in Europa and Ganymede, whereas complementary models of volatile speciation and redox cycling inform hypotheses about oxidant delivery through ice shells and potential habitability [27,73]. Laboratory-validated chemical-divide approaches and mission-driven flowcharts further link surface mineral detection to subsurface fluids, offering predictions for system-wide signatures that future observations can test across multiple Jovian moons [10,74].
One of the earliest studies, McCollom [28], utilized geochemical models to study the possibility of lithoautotrophic methanogenesis to produce methane (CH4) via CO2 and H2 as a source of chemical energy for metabolic processes in hydrothermal systems to produce biomass. By developing numerical geochemical models, the author studied the fluid composition changes throughout hypothetical seawater and basalt reactions at high temperatures (Figure S5) and assessed the available energy from methanogenesis throughout mixing seawater with this hydrothermal fluid. Such geochemical models are typically based on thermodynamic data and chemical reaction kinetics. These calculations aimed to characterize fluid–rock reactions within Europa’s hydrothermal systems, focusing on environments around hydrothermal vents where chemical energy is generated by mixing hydrothermal fluids and seawater. These models propose methanogenesis as a potential metabolic energy source for hypothetical hydrothermal systems in Europa [28]. The cross-section of this conceptual hypothetical hydrothermal system in Europa is represented in Figure S5.
The reactions of ocean water with igneous rocks at high temperatures in the subsurface could lead to the conversion of dissolved methane to a source of H2 as well as CO2 for methanogenesis. This process could provide ingredients to support metabolic processes [28], such as hydrogenotrophic methanogenesis, where H2 is oxidized to H+, and CO2 is reduced to CH4 [75]. High H2 concentrations in hydrothermal fluids in a rather oxidized ocean could provide energy from methanogenesis or sulfate reduction, similar to Earth’s hydrothermal systems, where methanogenesis provides comparable quantities of energy to the abundant ecosystems that surround subsurface hydrothermal vents (Figure 6).
McCollom [28] concluded this study by suggesting the possibility of enough metabolic energy to support life and localized ecosystems in Europa’s hydrothermal systems. However, even in the best-case scenario, the total supportable biomass by lithoautotrophic microbes on Europa will be negligible compared with Earth’s photosynthetically produced biomass.
The detection of Europa’s subsurface ocean could be performed via an orbiting radar sounder. Moore [29] explored the effects of possible ice chemistries in Europa on radar attenuation, utilizing temperature profiles. Ice chemicals were estimated via geochemical models for oceans dominated by sulfate or chloride, experimental data for the Earth’s marine ice, and rock and ice mixture models. This model was founded on several assumptions, such as a surface temperature of 80 K, a total ice thickness of 5 km, and chloride ions as dominant radar absorption materials due to incorporation into the ice lattice, 1% NaCl in pure ice and the tendency of saltwater to drain salinities similar to those of Baltic Sea ice (approximately 1 ppt, or 0.1%). Additionally, it was assumed that the chloride ions in the ice lattice, or brine pockets in cases of rapid freezing or sea ice conditions, lead to main radar absorption. Ignoring the interpretation challenges posed by scattering, this study concluded that realistic ice-penetrating radars could penetrate several kilometers into the ice (Note S3).
FREZCHEM is an “equilibrium chemical thermodynamic model for concentrated electrolyte solutions” that is widely utilized for studying cold geochemical processes and potential life conditions in the Arctic, Antarctica, and even celestial bodies such as Mars and Europa (Marion et al., 2010) [30]. This model incorporates methods for modeling variations in temperature, evaporation, and pressure within environmental contexts, can be utilized even for extremely concentrated solutions with molality up to moles/kg (H2O), and can apply the Pitzer approach for temperatures ranging from −70–25 °C and pressures ranging from 1 to 1000 bar [30].
Geochemical modeling applications to subzero temperatures have sparked substantial interest in geochemical processes in cold environments on Earth, Mars, and Europa. However, Marion [31] reported that only a handful of geochemical models directly include acid chemistry, and even those models are largely restricted to 0 °C and above or are dependent on the less usual mole-fraction scale instead of the molal scale. To address this, Marion [31] focused on developing, validating and evaluating the limitations of a molal-based model for sulfuric, nitric, and hydrochloric acids at low temperatures via Clegg mole-fraction acid models as well as the inclusion of acid chemistry and nitrate minerals in the FREZCHEM model for simulating Europa’s possible acidic brines.
Marion [31] utilized Clegg mole-fraction acid models for determining mean ionic activity coefficients and water activities to develop a database for predicting molal Pitzer equation parameters for HCl, HNO3, and H2SO4 for −85.15 °C, −45.15 °C, and −65.15 °C, respectively, to 24.85 °C. A comparison of the experimental measurements and model eutectics suggest reasonable agreement for HNO3 and H2SO4 within ±0.2 °C, whereas the results of pure HCl at subzero temperatures were identified as unreliable according to the experimental freezing point depression data. Therefore, alternative methods have been suggested for HCl chemistry at subzero temperatures. In addition to solid-phase acid solubility, this study expanded the FREZCHEM model by offering three nitrate minerals and acid salts. In addition, it refined equilibria between water phases (solid, liquid, and water vapor) between −93.15 and 24.85 °C. The system was then parameterized for sodium, potassium, magnesium, calcium, hydrogen, chloride, sulfate, nitrate, hydroxide, bicarbonate, carbonate, carbon dioxide, and water [31].
For Europa, Marion [31] reported that the application of the developed model for theoretical brines consisted of magnesium sulfate, sulfuric acid, and water, as well as sodium sulfate, magnesium sulfate, sulfuric acid, and water, indicating the freezing effects in converting a predominantly saltwater solution into a predominantly acidic solution at subzero temperatures. These findings imply that salinity, acidity, and temperature may affect the possibility of life on Europa and that high acidity could limit life forms to very acidophilic ones (Figure 7).
Considering geochemical models, geophysical evidence, and meteorite investigations, McCord et al. [32] identified new spectroscopic data for analyzing Europa (Figure S6). This study’s hypothesis was based on evidence from the ocean under Europa’s frozen crust and the reflectance spectra of disturbed surface areas that suggest hydrated substances such as salts. The simulation of flash-freezing sulfate and carbonate salty brines when interacting with Europa’s frigid surface suggested substances with near-infrared reflectance spectra close to those of non-ice zones of Europa and different from those of crystalline minerals. These new spectroscopic data suggest that the non-ice materials in the disrupted areas of Europa contain substantial quantities of endogenically disordered and significantly hydrated MgSO4 and possibly Na2SO4.
To estimate Europa’s surface composition and the possible subsurface ocean, Orlando et al. [33] used a combination of geochemical modeling, atmosphere information and infrared reflectance spectra. The approach of Orlando et al. [33] suggests similarities to the Moore [29] method in remote sensing and the McCollom [28] and Marion [31] methods in utilizing geochemical modeling. The Galileo mission’s Near-Infrared Mapping Spectrometer detected distorted water vibrational overtone bands, indicating highly concentrated solvated contaminants that suggest the presence of hydrated salt minerals or sulfuric acid. Studies of flash-frozen acid and salt mixtures as Europa surface analogs suggest that sodium and magnesium-bearing sulfate salts mixed with sulfuric acid best correlate with the observed spectra. This indicates higher sodium and proton contents, which is consistent with low-temperature aqueous differentiation and hydrothermal processing of carbonaceous chondrite-forming materials during Europa’s formation and early evolution (Note S4).
Europa’s hypothetical ocean bottom pressure could be greater than 1.2 × 10 8 Pa, which is comparable to Earth’s deepest ocean basin, the Mariana Trench ( 1.1 × 10 8 Pa). Chemical thermodynamic relationships must explicitly account for pressure at such high pressures. In addition to the role of pressure on equilibrium constants, activity coefficients, and water activity, models need to consider processes at subzero temperatures in cold environments [27].
To simulate pressure-dependent processes on Europa, Marion et al. [27] developed a geochemical model by adapting the FREZCHEM model to include pressure dependency and parameterizing it for subzero temperatures. After evaluation, the model was utilized to investigate at what distance above Europa’s presumed ocean in the ice layer the thermodynamically stable brine pockets that support life habitats could be expected (Figure 8). For a theoretical non-convecting 20 km thick ice shell, such a life zone could exist within 2.8 km before reaching the eutectic, but the formation of stable aqueous phases for supporting life near the surface is not possible. Moreover, for a 100 km deep ocean in Europa, comparing chemical equilibria at pressures equal to 10 5 Pa (1 bar) with 1.46 × 10 8 Pa (1460 bars) indicates a 12 °C decrease in the ice-forming temperature, an 11 °C increase in the MgSO4·12H2O-forming temperature, and only a 1.2 °C decrease in the eutectic temperatures. The author concluded that the effects of pressure on chemical systems are dependent on the volumetric properties of the ingredients.
Notably, Marion et al. [30] continued their studies on the FREZCHEM to mitigate the convergence issues that plagued older versions of the model by introducing a more practical mathematical algorithm based on Gibbs free energy to determine chemical equilibrium. They released this model as a web app and a website to host codes and instructions for using various model versions, but unfortunately, it is no longer accessible.
To connect observations of surface chemistry with Europa’s theoretical subsurface ocean chemical environment and to address the endogenic and exogenic origins of detected species, Johnson et al. [10] examined the freezing behavior of brines representative of Europa’s ocean composition. Their analysis focused on estimating the sequence in which hydrated minerals precipitate as a function of relative ionic concentrations and pH during progressive freezing of solutions containing sodium (Na+), chloride (Cl), magnesium (Mg2+), and sulfate (SO42−). Figure 9 synthesizes these pathways by depicting the evolution of ionic concentrations across successive freezing stages for multiple solution types (Oxidized Salty, Saturated, Equimolar, and experimental Pathways A and B), with corresponding mineral phases annotated. This representation highlights how minerals such as Na2SO4·10H2O, MgSO4·7/11H2O, MgCl2·12H2O, and acid hydrates emerge as brines concentrate, supporting interpretations of mineral assemblages observed on Europa’s surface and their potential linkage to subsurface ocean chemistry.
The authors claimed that the flowchart estimations are validated experimentally and have reasonable agreement with the published results of FREZCHEM for related brines, and the discrepancies are attributed to potential metastable effects, limitations of Raman spectroscopy, and the inherent complexities of thermodynamic models [10].
According to these flowcharts, significant amounts of magnesium (Mg) on Europa’s surface suggest an ocean with a pH < 8.4, whereas its absence indicates a pH ≥ 8.4. While the detection of sodium sulfate is difficult, it confirms the presence of sodium (Na) in the ocean. In contrast, the identification of endogenously derived sodium sulfate in addition to magnesium sulfate suggests a low-pH ocean with low sodium and high magnesium and sulfate contents. While the presence of sodium chloride with magnesium sulfate could suggest an exogenic origin for the magnesium sulfate, observing it with magnesium chloride could indicate an endogenic origin for magnesium sulfate or a recent formation of the chloride collection [10].
The authors concluded that while these species could be detected via a mapping imaging spectrometer, ultraviolet spectrometer, and mass spectrometer, the vibrational spectrometer would be the most suitable instrument among the nine selected instruments for the Europa Clipper mission. To improve the interpretation of ocean composition on the basis of surface expression and address species that have overlapping characteristics in Raman signatures, a complementary technique to the vibrational spectrometer for a potential Europa lander would be helpful [10].
On the basis of NASA’s Galileo spacecraft reports, Europa could have an iron-rich core, which only forms at temperatures above 1250 K or more. On the basis of this, Daswani et al. [34] suggested that Europa’s potential ocean beneath its icy exterior could have formed through metamorphism: partial devolatilization of Europa’s interior minerals, including carbonates and phyllosilicates, released large volumes of volatile compounds, such as CO2 and solutes such as Ca2+, SO42−, and HCO3 and the formation of an ocean with CO2 levels higher than current estimates and even an early CO2 atmosphere. This implies that ocean formation may not have required external water inputs, such as from comets [34].
This hypothesis suggests that CO2 is the most abundant solute in Europa’s ocean and that gypsum precipitation (Figure 10) could reduce the dissolved S/Cl ratio in the upper ocean layers. This could explain the observed chlorine-rich salts on the surface of Europa, suggesting that chlorine could be more prevalent at shallower depths. The precipitation of gypsum could also lead to the formation of a 3–10 km thick sedimentary layer on the seafloor [34].
The study concluded that the volatile mass from the deep interior of Europa could be equal to or exceed its current hydrosphere mass. The resulting ocean, created through thermal evolution, could be rich in carbon and sulfur. A CO2–rich ocean might even facilitate the emergence of life by creating a proton gradient between acidic ocean water and alkaline hydrothermal fluids [34].
The authors also mentioned that pressure changes impact the ocean composition by promoting gypsum precipitation and increasing the chlorine concentration in the upper water column. The originally high volatile mass in Europa must have been effectively reduced through metamorphism and fluid migration, with the excess volatile mass potentially escaping into space or being retained in clathrates. Despite the robustness of the current models, further research is necessary, particularly in relation to refining thermodynamic data and improving our understanding of fluid percolation processes. This will further inform our knowledge of ocean worlds’ habitability before the arrival of missions such as Jupiter Icy Moons Explorer (JUICE) (by the European Space Agency (ESA)) and Europa Clipper (by NASA) [34].
Across the reviewed studies in the Jupiter system, modeling objectives commonly seek to connect surface spectroscopic signals to ocean chemistry and seafloor reactions, with thermodynamic frameworks handling pressure and low temperature. Methods differ in the degree of emphasis on carbonate versus acid systems and in how oxidants and sulfate are treated. Findings converge on the diagnostic value of freezing pathways and mineral sequences but diverge regarding the expected pH, sulfate abundance, and balance between endogenous and exogenous sources of salts, highlighting the need for integrated geochemical models and targeted measurements spanning Europa, Ganymede, and Io [10,27,73].

3.3. Saturn

Saturn (Figure S7) is the sixth planet from the Sun. An average radius and mass of approximately 9.14 and 95.16 times Earth, respectively [76], make Saturn the second-largest planet in the solar system. Hydrogen (H2: 96.3% ± 2.4%) and helium (He: 3.25% ± 2.4%) are major constituents of this gas giant’s atmosphere [76]. Currently, 63 of 83 Saturn moons have been confirmed and named, and the discovery of the remaining moons should be confirmed and officially named by the International Astronomical Union (IAU). The size of Saturn’s moons ranges from as small as a sports arena to its giant moon Titan, which is larger than the planet Mercury [77]. The largest Saturn moons by the proximity of their orbits are Mimas, Enceladus, Tethys, Dione, Rhea, Titan, Hyperion, and Iapetus. Saturn’s largest and sixth-largest moons are critical science targets: Titan has an Earth-like cycle of liquids flowing across its surface, and Enceladus is one of the most scientifically compelling bodies in our solar system [77].
The Saturn system offers complementary geochemical modeling targets that include alkaline carbonate oceans inferred for Enceladus and cryogenic hydrocarbon lakes on Titan. Equilibrium carbonate-speciation and reaction-path models use Cassini plume and grain data to constrain pH, CO2 activity, and water–rock buffering in Na–Cl–CO3 systems, whereas non-ideal solution models at 90–110 K quantify the composition and solubilities of Titan’s lake liquids and their coupling to the atmosphere [35,36,78]. Together, these approaches enable system-level questions about redox gradients, organic synthesis, and mineral precipitation, and they provide boundary conditions for assessing habitability and tracing the provenance of salts and organics across Saturn’s moons [37,38].

3.3.1. Enceladus

Enceladus is among the research targets of alien life in our solar system because of its potentially habitable subsurface ocean. This small moon’s ice-covered water ocean eruption in space forms a plume that, on the basis of Cassini’s spacecraft mission, may contain almost all the basic life requirements known on Earth, including inorganic and organic carbon, NH3, potentially H2S, and chemical energy (disequilibria for methanogenesis) [79]. This geologically active water world [78] and its subsurface ocean characteristics, such as interior processes, conditions, and potential for harboring life, could be explored by analyzing the composition of a plume that erupts gas and frozen sea spray of salt-rich ice grains into space [35] and performing geochemical modeling.
By employing geochemical modeling and Cassini data, Hao et al. [79] explored the presence and quantity of a missing crucial life element, phosphorus, in the Enceladus Ocean. They employed thermodynamic and kinetic modeling to explore phosphorus geochemistry, considering discoveries about the ocean-seafloor system on Enceladus. The results suggest a relatively rich dissolved phosphorus ocean (Figure 11), increasing the habitability potential of Enceladus’s ocean.
According to Hao et al. [79], Enceladus’s ocean aqueous phosphorus is assumed to be orthophosphate (e.g., HPO42−). The total dissolved inorganic phosphorus is expected to increase with increasing dissolved carbon dioxide and lower pH, which also depends on dissolved NH3 and SiO2. The reported levels are much higher than those of previous estimations and close to those of current Earth seawater. A high phosphorus content is typically correlated with high levels of (bi)carbonate. Kinetic modeling of phosphate mineral dissolution implies that the rate of geologically fast release of phosphorus from seabed weathering of a chondritic rocky core is substantially lower than the estimated age of the Enceladus ocean.
Robinson et al. [39] employed geochemical modeling to explore organic reactions that might occur within the subsurface ocean of Enceladus, an environment of significant astrobiological interest. They performed hydrothermal experiments with protonated benzylamines to determine how these compounds undergo deamination—a reaction where an amine group is removed—to form benzyl alcohols and ammonium at various temperatures. By understanding the rates of these reactions, they aimed to identify organic species that could serve as tracers for environmental conditions such as temperature, pH, and redox state in remote planetary settings. These results suggest that these deamination reactions can reach equilibrium over geological timescales, even at temperatures approaching freezing (Figure S12). This implies that similar organic processes could occur in Enceladus, making certain compounds valuable targets for future exploration of its potentially habitable subsurface environment.
To gain insights into the water chemistry of subsurface oceans on icy celestial bodies such as Enceladus and to explore the total dissolved carbonate concentrations in Enceladus’ ocean and other alkaline–carbonate ocean worlds, Fukushi et al. [40] investigated mineral precipitation in terrestrial alkaline soda lakes (Figure 12).
Fukushi et al. [40] provided evidence of in situ monohydrocalcite (MHC) formation in alkaline lakes on the basis of observations of the water chemistry and mineralogy of suspended matter in Olgoy, Boon Tsagaan, and Orog Lakes, which are located in the Valley of Gobi Lakes, Mongolia. By analyzing the dissolution equilibria of MHC and amorphous magnesium carbonate (AMC), Fukushi et al. [40] developed geochemical models to predict the concentrations of key ions in Enceladus’s ocean. The solution chemistries of these lakes were found to be near saturation with respect to both the MHC and amorphous Mg-carbonate (AMC), similar to other alkaline lakes worldwide. These findings suggest that the MHC is a direct authigenic product of evaporation in alkaline lakes, offering a potential analog for understanding similar processes in extraterrestrial alkaline–carbonate ocean environments [40]. They also estimated that the ocean contains approximately 1 mmol/kg magnesium ions (Mg2+), 10 µmol/kg calcium ions (Ca2+), and between 0.06 and 0.2 mol/kg total dissolved carbonates. These estimations highlight the importance of measuring the magnesium content in Enceladus plumes, as they could provide critical insights into the total dissolved carbonate concentrations and chemical characteristics of subsurface oceans in Enceladus and similar alkaline–carbonate ocean worlds.
To estimate the activity of carbon dioxide and the dissolved CO2 concentration of the Enceladus ocean, Glein & Waite [35] utilized plume mass spectrometry data gathered by Cassini for novel geochemical interpretations. The results indicate similarities between expectations from the dissolution and formation of certain mixtures of silicon- and carbon-bearing minerals at the seafloor.
Reaction path modeling (Figure S9) indicated that reduced or oxidized seafloor rocks containing quartz, talc, and carbonate minerals in the MgO–FeO–SiO2–CO2–H2O system, in addition to silica and molecular hydrogen formerly found at Enceladus, reinforce the idea of a rocky core with a heterogeneous structure with a carbonated upper layer and a serpentinized interior that hosts geochemically diverse environments and offers a geochemical gradient for harboring life [35].
To explore the ocean pH of Enceladus to establish a better understanding of the chemical processes occurring there, Glein et al. [78] employed a thermodynamic model of carbonate speciation and Cassini’s mission data. These authors reported a high pH in the Enceladus’ ocean (Note S5). The high pH and lack of sulfuric acid suggest that strong oxidizing substances from the surface do not significantly reach the ocean, which aligns with episodic resurfacing activity geophysical models on Enceladus. While this basic nature might be related to the release of CO2 gas from the ocean due to ocean mixing, the results are in agreement with those of previous geochemical reaction path models that suggested that this phenomenon could be due to the serpentinization of chondritic rocks, which produces hydrogen (H2), which serves as a chemical fuel for the synthesis of organic compounds similar to those found in the plumes of Enceladus through abiotic and/or biological processes, such as those occurring in meteorites altered by water. Nevertheless, it is not clear whether these processes are still ongoing or if past hydrothermal activity completely altered the rocky core of Enceladus. The presence of native hydrogen in the plumes may suggest that water-related chemical reactions still occur, providing a continuous energy source that could support possible life [78].
The developed chemical model of Enceladus’ water ocean suggests a solution of Na–Cl–CO3 with a pH of approximately 11–12. While the Enceladus ocean is rich in salt (NaCl), similar to the Earth’s oceans, it also contains many dissolved sodium carbonate (Na2CO3), such as soda lakes. This highly basic environment naturally implies a high amount of hydroxide ions, leading to low concentrations of divalent metals since they may be absorbed by carbonates and phyllosilicates on the ocean floor. Organic molecules with carboxyl groups (acidic parts) carry a negative charge, whereas amino groups (basic parts) are mostly in a neutral state [78].
To explore the potential of serpentinization and the rate of hydrogen production on Enceladus, Nissen et al. [81] employed geochemical modeling via the EQ3/6 software package and thermodynamic databases created by the DBCreate program (Nissen et al., 2015). EQ3/6 is a software package for modeling geochemical interactions between aqueous solutions, solids, and gases that employs thermodynamic and kinetic chemical principles. It helps interpret chemical compositions and predict reactions with minerals, solids, and gases. The EQPT preprocesses thermodynamic data, EQ3NR handles speciation-solubility, and EQ6 simulates reaction pathways [82]. Nissen et al. [81] assumed that the chemical composition of plumes is similar to that of Enceladus’ subsurface aquifer and explored a water/rock boundary in various ranges of pH: 7–9, pressure: 25–50 bars, and temperature: 0–50 °C. The reported results (Figure 13) suggest the possibility of serpentinization at the water/rock boundary in the case of an iron-rich end-member in the Enceladus core [81].
The explosive ejection of cold material from Enceladus’ icy interior was explored by Neveu et al. [26] through the development of a novel geochemical equilibrium model, which bridges geophysics and geochemistry to explore the realm of subzero-temperature liquid water. Details are presented in Section 3.6 (Pluto).
To explore the aqueous fluid composition of CI chondritic materials on carbonaceous asteroids, trans-Neptunian objects, and icy moons such as Enceladus, Zolotov [41] employed simulated chemical equilibrium in closed water–rock–gas systems under varying conditions of initial fluids: water/rock mass ratios of 0.1–1000, temperatures below 350 °C, and pressures below 2 kbar. In the modeled alkaline solutions (Note S6), the major species include Na+, K+, Cl, HCO3, CO32−, OH, H2, and CO2. Aqueous forms of Mg, Fe, Ca, Mn, Al, Ni, Cr, S, and P are less prevalent, as their corresponding solids have little solubility [41].
The results suggest that fluid compositions are influenced primarily by the solubility of solids, initial speciation of chlorine in water–rock mixtures, and Na-bearing secondary minerals, such as saponite. NaCl-type alkaline fluids, along with saponite-bearing mineralogies, are available in aqueously altered chondrites, which is consistent with the composition of grains emitted from Enceladus. The results of this study also suggest that the fluids on Enceladus, which are rich in Na+, CO32−, and HCO3, likely interact with Na-depleted secondary minerals, providing clues about the moon’s subsurface water chemistry and potential habitability. Na-rich solutions with considerable CO32−, HCO3, and OH occur alongside secondary mineralogy without Na. Cometary ice accretion might lead to the formation of Na2CO3 and NaHCO3. NaOH follows in reduced conditions and might be formed on the original CR Cc’s. Melting HCl-bearing ice is thought to produce initial acidic solutions rich in Mg and Fe, along with other metals, which could result in basic Na-rich solutions when neutralized. Sulfate species do not develop in closed systems that remain strongly reducing, because the accumulation of H2 gas increases internal pressure and helps maintain isolation from oxidizing conditions. While the quantity of CH4 cannot be accurately determined, it is estimated that the gaseous phase is dominated by H2, CO2, and H2O [41].

3.3.2. Titan

Hundreds of radar-dark areas in Titan’s northern and southern polar areas are assumed to be lakes [36]. In addition to hydrocarbon lakes, solid organics form due to photochemical reactions where they precipitate on their surface and can interact with liquids, and geochemical processes may also occur [38]. Moreover, the dense atmosphere of Titan, which is made of N2 and CH4, is so unique in solar systems that its potential origin is still uncertain [42].
To explore the possible origin of its non-photochemical gases, Glein [42] analyzed chemical and isotopic data from the Cassini–Huygens mission. This study suggests hydrothermal and cryovolcanic processes and the rocky core as the possible origin of methane, nitrogen, and noble gases (Note S7). Referring to mass balance and chemical equilibrium calculations, Glein suggested that the geochemical feasibility of this theory is supported by observations, experiments, and theory [42].
This study explored Titan’s noble gas origins in terms of CI carbonaceous meteorite contents, Henry’s law constants for noble gases in relevant condensed phases, and the stability of clathrate hydrates. Exploring the geochemistry of noble gases opened an avenue to the origin of N2 and CH4. It was suggested that the thermal decomposition of NH3 in a hot rocky core could generate atmospheric nitrogen, and more than 95% of Titan’s nitrogen could still be trapped in its interior. Additionally, methane possibly originates endogenously from CO2 through serpentinization, i.e., through geochemical reactions between liquid water and anhydrous rock, which could be so significant that methane escape and irreversible photolysis are overlooked. The study concluded that the geophysical evolution of Titan’s interior may have led to volatile processing that formed its atmosphere [42].
These results may support the findings of Glein et al. [83], who argued that Titan’s methane likely did not come from inside the moon (via processes such as serpentinization) but was present from when Titan formed. In this study, they estimated the D/H ratio from serpentinization by employing equilibrium isotopic fractionation models in the CH4–H2O–H2 system, assuming that the bulk D/H ratio is the same as that of water in the plume of “Enceladus”. They concluded that hydrothermal systems capable of producing methane might not have existed due to Titan’s too-cold interior, or if they did, they would have been sufficiently oxidized, making carbon mostly in the form of CO2, not CH4, while leading to hydrothermal oxidation of ammonia (NH3), which could be the origin of nitrogen (N2) in its atmosphere [83]. Notably, researchers may mention uncertainties, as both studies suggest the need for more evidence, data, and studies.
Referring to the abundance of organic materials in Titan’s atmosphere and surface, Glein and Shock [38] also explored a geochemical model of nonideal solutions on Titan and proposed “cryogenic fluvial geochemistry” as a novel type of geochemistry [38] by employing a thermodynamic model to explore the solubilities of gases and solids in liquid hydrocarbons at cryogenic temperatures. The van Laar model employed experimental phase equilibrium data for a chemical system of methane (CH4)–ethane (C2H6)–propane (C3H8)–nitrogen (N2)–acetylene (C2H2) at temperatures ranging from 90 K to 110 K. The authors argued that their model offers a reasonable balance between accuracy and simplicity and the possibility of expanding to more components while maintaining thermodynamic consistency. They reported that the quantity of atmospheric CH4 governs the equilibrium composition of surface liquids, as shown in Figure 14, which plots liquid-phase mole fractions of CH4, C2H6, and N2 versus pressure at 95 K using experimental vapor-liquid equilibrium (VLE) data. Figure 15 summarizes key physical and thermodynamic properties of Titan-relevant compounds (e.g., critical volumes, fugacities, and Gibbs energies), providing context for cryogenic solution behavior and solubility trends. Together, these figures illustrate how Titan’s hydrocarbon lakes exhibit non-ideal solution chemistry and how acetylene’s high solubility could lead to sedimentary deposits and chemical erosion. The C2H2 geochemical features of Titan’s liquid hydrocarbons may be intermediate to those of calcite or gypsum in Earth’s surface water [38].
To estimate the potential chemical composition of Titan’s lakes, Cordier et al. [36] employed data from the gas chromatography mass spectrometer on the Huygens probe, along with photochemical models that simulate chemical reactions in Titan’s atmosphere on the basis of height–temperature profiles measured by the Huygens Atmospheric Structure Instrument. By assuming thermodynamic equilibrium between the atmosphere and the lakes and nonideal solutions, they argued for Titan’s lake compositions and approximate quantities, as represented in Table 4 [36].
Titan’s hypothesized ocean properties are also explored by Dubouloz et al. [84] on the basis of atmosphere composition and surface temperature suggested by data from Voyager 1 analyzed by Lellouch et al. [85]. Assuming that the ocean is a nonideal solution in thermodynamic balance with the atmosphere, Dubouloz et al. [84] suggested that ocean properties can significantly vary on the basis of atmospheric parameters. This study also examined how solids form at the bottom of the ocean. The thickness of these solid deposits does not change much with different surface temperatures or atmospheric conditions. Additionally, Titan’s ocean could act as a major reservoir for carbon monoxide (CO), potentially storing 0.25 to 11 times more CO than the amount present in the atmosphere, depending on the ocean’s composition (more ethane at lower temperatures or more methane at higher temperatures). Assuming a surface pressure of 1.5 atm and considering the presence of argon in the atmosphere, this study’s major reports of surface temperatures of 92.5 K and 101 K are summarized in Table 5 [84].
Modeling studies in the Saturn system share the goals of constraining fluid compositions and reaction pathways but differ in the targeted environments and thermodynamic treatments. These findings consistently indicate strongly alkaline conditions and carbonate buffering for Enceladus and ethane-rich, multicomponent liquids for Titan’s lakes, whereas uncertainties persist in low-temperature kinetic data, cryogenic activity coefficients, and plume fractionation effects. Addressing these gaps will improve system-wide comparisons and strengthen links between modeled geochemical states and observations from past and future missions [35,36,78].

3.4. Uranus

Uranus (Figure S10) is the seventh planet from the Sun and the solar system’s third-largest and coldest planet, as it emits much less heat than Jupiter, Saturn, and Neptune. This ice giant has a thick atmosphere mostly made of hydrogen (H2: 82.5 ± 3.3%), helium (He: 15.2% ± 3.3%), and traces of methane (CH4: 2.3%) [86] and includes flowing icy materials, including H2O, CH4, and NH3 above this small rocky core in quantities greater than those of Jupiter and Saturn, making it unlikely to harbor life with current known principles on Earth. Surrounded by 13 faint rings and 27 identified moons, Uranus is among the least known planets in our solar system and was only explored once in 1986 by NASA’s Voyager 2 spacecraft since its discovery by William Herschel in 1781. Uranus’s unique tilt of access makes it the only planet in our solar system whose poles face the sun, while it rotates east to west like Venus does [87,88,89].
The Uranus system presents sparse in situ constraints, yet geochemical modeling remains relevant for hypothesized cryovolcanic processes and potential subsurface liquids on mid-sized moons. Voyager-era morphology and subsequent analyses of cryovolcanic resurfacing motivate the cautious use of geochemical scenarios under limited data regimes [90]. Recent re-analyses of energetic particle observations indicate localized sources between Miranda and Ariel that may be consistent with material release from active moons, renewing interest in ocean-world signatures and the role of internal plasma sources [91]. Geochemical models that can parametrize endmember compositions and brine behavior should be used in future missions to confirm subsurface reservoirs.
The findings from Diab et al. [24]’s geochemical modeling could explain celestial icy objects that have harbored deep oceans throughout their evolution, particularly those less than 1000 km in size. These bodies can retain only low temperatures within their rocky mantles. Notable examples include Pluto, Charon, and other dwarf planets, as well as the medium-sized moons orbiting Saturn and Uranus. Details of this study are discussed in Section 3.1 (Ceres).
Various studies have focused on evidence and mechanisms for effusive volcanism. Fagents [90] reviewed studies on previous effusive cryovolcanism to explore morphologies on the moons of Jupiter (Europa and Ganymede), Saturn (Enceladus, Dione, Tethys, and Iapetus), and Uranus (Miranda and Ariel).
The objectives of Uranus-related studies have focused primarily on the plausibility of cryovolcanism and subsurface activity, with methodological differences driven by data scarcity. While geomorphological assessments caution against over-interpreting cryovolcanic resurfacing, emerging particle evidence motivates renewed consideration of internal sources that geochemical modeling can formalize as test cases for future exploration [90,91].

3.5. Neptune

Neptune (Figure S11) is the eighth and farthest planet from the Sun, which makes it our solar system’s sole planet that cannot be seen with the naked eye. This ice giant is characterized by darkness, coldness, and supersonic winds (up to 580 m/s) and is the densest of our solar system’s giant planets [92]. Hydrogen (H2: 80.0% ± 3.2%), helium (He: 19.0% ± 3.2%) and methane (CH4: 1.5% ± 0.5%) are major constituents of Neptune’s atmosphere [93]. Neptune’s radius and mass are about 3.83 and 17.15 times greater than those of Earth, respectively, but the temperature, pressure, and materials make it unsuitable for adaptation of the currently known life forms.
Neptune’s ring system comprises at least five primary rings (Galle, Leverrier, Lassell, Arago, and Adams) and four distinct ring arcs (Liberté, Egalité, Fraternité, and Courage), all of which are believed to be relatively new and transient. Contrary to Newtonian motion predictions, these dust arcs in the outermost Adams ring persist in clusters. The gravitational impact of Galatea, a moon within this ring, is currently considered the stabilizing factor for these formations [92].
Neptune was the first planet discovered through mathematical predictions in 1846. Among Neptune’s 15 moons, Triton (Figure S8) is the largest and was discovered shortly after Neptune. Unique in its retrograde orbit, Triton may have once been a separate celestial body captured by Neptune. Despite its harsh surface temperature of approximately −235 °C, Voyager 2 reported that geysers were erupting icy materials. The same mission also revealed Triton’s thin and mysteriously warming atmosphere [92].
In the Neptune system, Triton provides the primary context for geochemical modeling because it offers direct observational anchors for volatile cycling and cryovolcanism. To date, nitrogen (N2) gas plumes have been identified on Triton, which seems to be a KBO captured by Neptune [94,95]. Seasonal cycles of N2, CH4, and CO ices, combined with evidence for active surface–atmosphere interactions, motivate simulations that quantify polar cold traps, volatile redistribution, and surface composition, which can be coupled with geochemical scenarios for interactions between volatiles and water-ice substrates [96,97]. On the basis of these observations, a low-temperature equilibrium framework that couples geophysics and geochemistry quantifies how exsolution of CO, N2, CH4, and H2 in aqueous cracks can overcome negative buoyancy and propagate fractures to the surface, using HKF/CHNOSZ thermodynamics extended into the subzero regime to predict vapor fractions, molalities, and crack-length thresholds across plausible internal structures; this directly informs hypotheses for Triton’s plumes and volatile inventories [26].
At the planetary scale, atmosphere-to-interior inference developed for Neptune-class exoplanets provides a transferable strategy for Neptune: quench-chemistry constraints on redox pairs (CO2–CH4, CO–CO2, N2–NH3) can be used to back out pressure–temperature regimes and interior–atmosphere exchange, linking volatile enrichments to origin scenarios [98]. Complementary geochemical models of miscible, metal-rich envelopes show how equilibrium/quenched compositions and oxidation states reconcile depleted CO and NH3 with strong C and O enrichments, adapting the same quench-chemistry and mixing frameworks to Neptune’s atmosphere—and to moon–planet interactions that source and recycle N2, CH4, and CO across the Triton–Neptune system—enables testable predictions for volatile partitioning and interior redox conditions [99].

3.6. Pluto

Once designated our solar system’s ninth planet, Pluto (Figure S13) is now recognized as a notable dwarf planet within the Kuiper Belt, a toroidal region of numerous icy bodies beyond Neptune’s orbit [100]. With an average radius and mass of approximately 18.63% and 0.22% of Earth, respectively, Pluto’s diameter and mass are approximately 68.37% and 17.81% of Earth’s moon [76], and its complex geography boasts a glacier and towering mountains, with an intriguing occurrence of red snow. Pluto’s highest mountains are 2 to 3 km tall and made of water ice, occasionally coated by frozen gases such as methane. Notably, Pluto plains seem to consist of frozen nitrogen and lack craters but show casing structures, indicating the vertical circulation of materials due to convection [100]. Nitrogen (N2: 99%), methane (CH4: 0.5%), and carbon monoxide (CO: 0.05%) are major constituents of Pluto’s thin and tenuous atmosphere [101], which extends as it moves toward the Sun and shrinks as it moves away from it, such as comets [100]. NASA’s New Horizons was the only spacecraft to visit Pluto in July 2015. Pluto’s surface is inhospitably cold (228–238 °C) to support life as we know it. Five moons are identified for Pluto: Charon, Nix, Hydra, Kerberos, and Styx. A diameter of approximately half of Pluto’s diameter makes Charon the largest satellite relative to the planet in the solar system. Pluto and Charon are usually described as double planets [100].
New Horizons observations revealed geologically young terrains and active volatile processes on Pluto, including nitrogen–ice convection and glacial transport, underscoring the value of geochemical modeling for evaluating volatile partitioning, phase behavior, and possible cryovolcanic materials in water–ice crusts [102]. Models that integrate volatile chemistry with thermophysical evolution can help explain spatial heterogeneity and guide hypotheses for subsurface reservoirs, even in the absence of direct sampling.
The term “cryovolcanism” refers to the explosive ejection of cold material from the interiors of icy celestial bodies and is supposedly observed on Enceladus, Europa, Triton, and Ceres. A geochemical equilibrium model [26] was used to explore the behavior of volatiles within cryovolcanic environments, specifically in KBO. Their model accounts for the standard aspects of geochemical modeling (e.g., reactions between various components, phases, and changes in those relationships with varying physical conditions), bridges geophysics and geochemistry, and, while many geochemical models neglect low-temperature regimes, it extends the reach of geochemical modeling into an underexplored context: the realm of subzero-temperature liquid water.
Neveu et al. [26] first examined the possibility of subsurface liquid spatial and temporal distributions, strategies to counter the negative buoyancy of water enclosed in ice, and the volatile composition of KBO. Then, they introduced a geochemical equilibrium model for volatile exsolution rising through cracks. They demonstrated how volatiles such as carbon monoxide (CO), nitrogen (N2), methane (CH4), and hydrogen (H2) generated through hydrothermal activity could play pivotal roles in rising fluids. Additionally, they hypothesized that other potential volatile contributors to this process, such as nitrogen (N2), methane (CH4), and hydrogen (H2), further expand the boundaries of geochemical modeling. Neveu et al. [26] noted that the internal structure of celestial bodies could influence the propagation of cracks. For example, a hydrated core was favorable for explosive cryovolcanism, whereas an undifferentiated crust was not (Figure 16). Additional factors potentially leading to fluid ascent, such as the freezing of fluid in cracks, are briefly discussed. Neveu et al. [26] also offered a chart of the applicability of pressure–temperature–density geochemical models in icy dwarf planets [26,103,104,105,106] by adapting Holten et al. [107]’s work (Note S8). Finally, they provided estimations for the New Horizons exploration of the Pluto–Charon system.
The studies reviewed for Pluto share objectives centered on volatile dynamics and resurfacing, typically using thermophysical and geologic mapping constrained by spacecraft data. Differences appear in the proposed cryovolcanic mechanisms and volatile inventory assumptions. The collective findings point to active volatile transport and plausible cryovolcanic signatures while highlighting the need for geochemical constraints on nitrogen-rich ices and their interactions with water-ice substrates [102].

3.7. Synthesis of Reviewed Studies

While the preceding subsections provide detailed descriptions of individual studies, several common themes and limitations emerge across this body of work. Thermodynamic modeling dominates the literature, yet gaps persist in kinetic modeling and in handling extreme pressure–temperature conditions relevant to icy moons and dwarf planets. Consensus exists on the likelihood of subsurface oceans in Europa and Enceladus, but uncertainties remain regarding their chemical composition and habitability potential. These patterns, along with research gaps such as the scarcity of studies on Triton, Charon, and other KBO, are synthesized and discussed in detail in Section 4.

4. Discussion

Building on the descriptive mapping of individual studies in Section 3, this section provides a critical synthesis of the reviewed literature, highlighting overarching themes, methodological limitations, and research gaps. It evaluates the implications of these findings for planetary science and outlines directions for future work.
The extreme expenses and significant timeframes required for space exploration missions, particularly those directed at celestial bodies beyond Mars, present a significant barrier to advancing the understanding of the solar system’s distant regions. From the Asteroid Belt to the Kuiper Belt, the challenge of exploring these remote worlds has driven the development of novel solutions. The present study aimed to explore novel solutions for employing geochemical modeling techniques to leverage existing data and resources, which could be called advanced data mining solutions in space, an alternative to direct sampling and analysis, and delves into current practical solutions for studying celestial bodies from a distance to reduce the need for costly and time-consuming missions. In addition to delving into geochemical modeling methods, models, and the most studied celestial objects, highlighting key trends and areas of focus, this study provides insights into works that employ methods such as remote sensing, spectrometry, and geochemical modeling methods and tools to explore the chemical composition, processes, and evolutionary histories of distant celestial bodies in our solar system to offer a feasible alternative to direct exploration, allowing researchers to investigate ambitious questions, from the potential for harboring alien life to the broader habitability of other worlds, despite the constraints of accessibility and cost.

4.1. Challenges and Solutions

Identifying studies that focus on the geochemical modeling of celestial bodies farther from Mars presents unique challenges. The limited scope of space exploration, the rarity of physical samples, and the prohibitive costs of missions to these distant objects limit the availability of data for researchers. This challenge is compounded by the interdisciplinary nature of the field, which integrates geology, chemistry, astrophysics, and planetary science, and the nuanced terminologies employed in these studies. Moreover, the topic of geochemical modeling for celestial bodies beyond Mars remains a niche within planetary science, leading to a relatively small volume of directly relevant research.
Despite these inherent challenges, the tools and strategies discussed in this study, including advanced search techniques, bibliographic screening methods, and keyword-specific database alerts, provide effective solutions for identifying limited but valuable studies on the geochemical modeling of distant celestial bodies. Bibliometric analysis, in particular, has proven essential in identifying key studies, authors, and trends within the field. While bibliometric and AI-based tools can assist in identifying relevant studies, the role of human expertise in interpreting complex concepts and establishing meaningful connections between diverse data sources remains indispensable. Researchers must balance computational tools with a deep understanding of geochemical processes to achieve meaningful insights.

4.2. Prominent Researchers, Journals, and Country Contributions

The reviewed studies suggest that the concept of geochemical modeling beyond Mars is dominated by contributions from a few key researchers, including Castillo-Rogez and McCord [66], who have significantly advanced our understanding of moons such as Ceres and Europa. Their work, along with that of other scholars, has been published predominantly in major planetary science journals. Icarus remains the most frequently cited journal, accounting for approximately 30% of the reviewed publications. Other prominent journals include Planetary and Space Science (25%) and Geochimica et Cosmochimica Acta (15%). The majority of these studies have been conducted by researchers based in the United States, followed closely by European countries, particularly Germany and the UK.

4.3. Most Studied Celestial Bodies

On the basis of the reviewed works focused on geochemical modeling, the most studied object is Europa, a moon of Jupiter, suggesting that it is a favored target for researchers. Enceladus, Titan, Ganymede, and Callisto also feature prominently [108], together accounting for more than 60% of the research. The focus on these moons is driven largely by their potential to support life, particularly in subsurface oceans where conditions may be favorable for microbial life [109]. Studies have extensively modeled the geochemistry of these environments to evaluate their habitability and potential biosignatures.

4.4. Most Popular Methods and Models

Various methods are utilized for geochemical modeling, such as remote sensing, spectrometry, spectroscopic data analysis, thermodynamic modeling, data analysis, and data mining. The reviewed studies suggest that among the methods used in geochemical modeling, spectrometry and thermodynamic simulations are the most common. Spectrometry, employed in 45% of the reviewed studies, has become a critical tool because of its extensive use in space missions such as Cassini and Galileo, which provide valuable data on the surface and subsurface compositions of moons such as Europa and Titan [110]. Galileo’s (near-infrared mapping spectrometer) NIMS observations further indicated hydrated salt minerals consistent with flash-frozen brines, strengthening the spectroscopic case for endogenic ocean-derived surface materials on Europa [111]. For Europa, exospheric detections of alkalis tied to regolith inventories provide complementary constraints on surface composition and the potential linkage to subsurface brines [112]. The thermodynamic models observed in 35% of the reviewed works offer robust predictions for chemical processes under varying temperature and pressure conditions. In particular, high-pressure water–ammonia phase equilibria up to ~300 MPa are directly applicable to ocean-world interiors, where dissolved NH3 both depresses melting points and alters eutectics and activity coefficients relevant to brine evolution [113]. The equilibrium models observed in 25% of the studies are also frequently applied to understand chemical stability, particularly in environments such as Titan’s surface lakes and Ganymede’s icy crust.
FREZCHEM is the most widely mentioned geochemical model in reviewed works, particularly for simulating processes in icy environments. The PHREEQC and CHNOSZ models are also commonly employed, particularly in studies analyzing interactions between water and rock, such as the formation of minerals and organic compounds in the subsurface oceans of Enceladus and Europa.

4.5. Key Materials and Alien Life Exploration

In scientific works, the most studied materials in geochemical modeling beyond Mars include water ice, sulfates, and carbonates, which are critical for understanding the geochemistry of moons such as Europa and Enceladus. On Europa, surface radiolysis drives a sulfur cycle that produces sulfuric acid within the ice, supplying oxidants that may be transported downward and interact with putative ocean brines, with important consequences for habitability [72]. Laboratory studies of frozen Na–Mg–sulfate–chloride brines reproduce spectral and chemical patterns expected for ocean-derived materials, providing a mechanistic bridge between surface detections and subsurface ocean compositions on Europa [114]. For Enceladus, equilibrium assessments indicate alkaline Na–Cl–CO3 fluids consistent with Cassini plume-grain chemistry, reinforcing carbonate buffering and potential energy sources for chemosynthetic pathways [108]. These materials are of interest because of their potential role in supporting life, as water ice is a key component of subsurface oceans, whereas sulfates and carbonates are linked to chemical processes that could sustain microbial ecosystems. The evaluation of these materials and their potential for extraterrestrial life remains a central theme in geochemical modeling, with over 50% of reviewed studies focusing on the habitability of these moons.

4.6. Relationship Between Distance and Study Volume

The relationship between the distance of celestial bodies from Earth and the volume of research is evident in this review. Naturally, objects that are closer to Earth, such as those orbiting Jupiter and Saturn, have been the primary focus of geochemical modeling, whereas more distant objects, such as Neptune’s Triton and Pluto’s Charon, are less studied. This discrepancy is largely due to the limited number of space missions that have reached these farthest celestial bodies. Future missions, such as the Europa Clipper and JUICE missions, are expected to increase the availability of data for more distant moons, potentially expanding the scope of geochemical modeling in the outer solar system. These missions will also provide critical observational constraints that can be coupled with improved thermodynamic and kinetic models, enabling more accurate predictions of ocean chemistry, volatile cycling, and habitability potential across icy worlds.

5. Concluding Remarks

In the foreseeable future, geochemical modeling offers a reasonable substitute for the limited number of direct explorations to understand the composition and evolution of celestial bodies beyond Mars. The present study is an initiative to unify the concept of geochemical modeling in space and offers novel contributions by providing a comprehensive review of relevant scientific works to date, highlighting the key methods and models used in this field, as well as the primary objects of study, to remediate the challenges posed by limited data availability. In particular, the scarcity of studies on Triton, Pluto, and other KBO represents a significant opportunity for advancing modeling approaches that compensate for observational gaps. The development of robust models for these distant bodies will not only guide future exploration priorities but also expand our understanding of cryovolcanism and volatile dynamics in the outer solar system.
Despite the complexities associated with modeling distant bodies, the insights gained from these studies significantly contribute to our understanding of the solar system’s outer reaches and the potential for life beyond Earth. The research landscape in this area is demanding, but with the continued development of tools and strategies, geochemical modeling will remain a critical tool for future discoveries in planetary science.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/encyclopedia6020038/s1. References [115,116,117,118,119,120,121,122,123] are cited in the Supplementary Materials.

Author Contributions

R.M.S.: conceptualization, funding acquisition, supervision, writing—review and editing. A.Y.: conceptualization, methodology, investigation, writing—original draft. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Natural Sciences and Engineering Research Council of Canada (Discovery Grant 401497).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Acknowledgments

Special thanks to Mitra Kaviani for invaluable advice and expertise in programming and implementing advanced data analysis techniques.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Albarède, F. Geochemistry: An Introduction, 5th ed.; Cambridge University Press: Cambridge, UK, 2007. [Google Scholar]
  2. Kragh, H. From geochemistry to cosmochemistry: The origin of a scientific discipline, 1915–1955. In Chemical Sciences in the 20th Century: Bridging Boundaries; Reinhold, C., Ed.; John Wiley & Sons: Hoboken, NJ, USA, 2008; pp. 160–192. [Google Scholar]
  3. McSween, H.Y.; Richardson, S.M.; Uhle, M.E.; Richardson, S.M. Geochemistry Pathways and Processes, 2nd ed.; Columbia University: Cambridge, UK, 2003. [Google Scholar]
  4. McSween, H.Y.J.; Huss, G.R. Cosmochemistry; Cambridge University Press: Cambridge, UK, 2010. [Google Scholar]
  5. White, W.M. Geochemistry, 2nd ed.; Wiley-Blackwell: Hoboken, NJ, USA, 2020. [Google Scholar]
  6. Zhu, C. Geochemical Modelling in Environmental and Geological Studies. In Encyclopedia of Sustainability Science and Technology; Springer: New York, NY, USA, 2012; pp. 4094–4104. [Google Scholar] [CrossRef]
  7. Bethke, C. Geochemical Reaction Modelling: Concepts and Applications; Oxford University Press: Oxford, UK, 1996. [Google Scholar]
  8. Khalidy, R.; Santos, R.M. Assessment of geochemical modelling applications and research hot spots—A year in review. Environ. Geochem. Health 2021, 43, 3351–3374. [Google Scholar] [CrossRef] [PubMed]
  9. Zhu, C.; Anderson, G. Geochemical Modelling. Zhu Laboratory, Indiana University Bloomington. 2023. Available online: https://hydrogeochem.earth.indiana.edu/software/index.html (accessed on 27 January 2026).
  10. Johnson, P.V.; Hodyss, R.; Vu, T.H.; Choukroun, M. Insights into Europa’s ocean composition derived from its surface expression. Icarus 2019, 321, 857–865. [Google Scholar] [CrossRef]
  11. IAU. Pluto and the Solar System; International Astronomical Union (IAU): Paris, France, 2006; Available online: https://web.archive.org/web/20230217184727/https://www.iau.org/public/themes/pluto/ (accessed on 27 January 2026).
  12. Beinahegut. The Solar System. Wikimedia Commons. 29 August 2019. Available online: https://commons.wikimedia.org/wiki/File:The_Solar_System.svg (accessed on 27 January 2026).
  13. Williams, D.R. Planetary Fact Sheet. NASA Goddard Space Flight Center. 2023. Available online: https://web.archive.org/web/20230308001412/https://nssdc.gsfc.nasa.gov/planetary/factsheet/ (accessed on 27 January 2026).
  14. Williams, D.R. Planetary Fact Sheet Notes. NASA Goddard Space Flight Center. 2021. Available online: https://web.archive.org/web/20221206024506/https://nssdc.gsfc.nasa.gov/planetary/factsheet/planetfact_notes.html (accessed on 27 January 2026).
  15. NASA. Jupiter Moons—Overview. NASA Solar System Exploration. 2023. Available online: https://web.archive.org/web/20230325204005/https://solarsystem.nasa.gov/moons/jupiter-moons/overview/ (accessed on 27 January 2026).
  16. Sheppard, S.S.; Tholen, D.J.; Alexandersen, M.; Trujillo, C.A. New Jupiter and Saturn Satellites Reveal New Moon Dynamical Families. Res. Notes AAS 2023, 7, 100. [Google Scholar] [CrossRef]
  17. Bigliardi, B.; Filippelli, S. Factors affecting the growth of academic oriented spin-offs. In Innovation Strategies in the Food Industry: Tools for Implementation, 2nd ed.; Elsevier: Amsterdam, The Netherlands, 2021. [Google Scholar] [CrossRef]
  18. Vogel, R.; Güttel, W.H. The dynamic capability view in strategic management: A bibliometric review. Int. J. Manag. Rev. 2013, 15, 426–446. [Google Scholar] [CrossRef]
  19. Aria, M.; Cuccurullo, C. bibliometrix: An R-tool for comprehensive science mapping analysis. J. Informetr. 2017, 11, 959–975. [Google Scholar] [CrossRef]
  20. Garfield, E. KeyWords PIus: ISI’s Breakthrough Retrieval Method. Part 1. Expanding Your Searching Power on Current Contents on Diskette. Curr. Contents 1990, 32, 295–299. Available online: https://garfield.library.upenn.edu/essays/v13p295y1990.pdf (accessed on 27 January 2026).
  21. Garfield, E.; Sher, I.H. Key Words Plus [TM]-Algorithmic Derivative Indexing. J. Am. Soc. Inf. Sci. 1993, 44, 298–299. [Google Scholar] [CrossRef]
  22. Zhang, J.; Yu, Q.; Zheng, F.; Long, C.; Lu, Z.; Duan, Z. Comparing keywords plus of WOS and author keywords: A case study of patient adherence research. J. Assoc. Inf. Sci. Technol. 2016, 67, 967–972. [Google Scholar] [CrossRef]
  23. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ 2021, 372, n71. [Google Scholar] [CrossRef]
  24. Diab, J.; Daswani, M.M.; Castillo-Rogez, J. Bulk composition and thermal evolution constrain the formation of organics in Ceres’ subsurface ocean via geochemical modelling. Icarus 2023, 391, 115339. [Google Scholar] [CrossRef]
  25. Castillo-Rogez, J.C.; Hesse, M.A.; Formisano, M.; Sizemore, H.; Bland, M.; Ermakov, A.I.; Fu, R.R. Conditions for the Long-Term Preservation of a Deep Brine Reservoir in Ceres. Geophys. Res. Lett. 2019, 46, 1963–1972. [Google Scholar] [CrossRef]
  26. Neveu, M.; Desch, S.J.; Shock, E.L.; Glein, C.R. Prerequisites for explosive cryovolcanism on dwarf planet-class Kuiper belt objects. Icarus 2015, 246, 48–64. [Google Scholar] [CrossRef]
  27. Marion, G.M.; Kargel, J.S.; Catling, D.C.; Jakubowski, S.D. Effects of pressure on aqueous chemical equilibria at subzero temperatures with applications to Europa. Geochim. Cosmochim. Acta 2005, 69, 259–274. [Google Scholar] [CrossRef]
  28. McCollom, T.M. Methanogenesis as a potential source of chemical energy for primary biomass production by autotrophic organisms in hydrothermal systems on Europa. J. Geophys. Res. Planets 1999, 104, 30729–30742. [Google Scholar] [CrossRef]
  29. Moore, J.C. Models of Radar Absorption in Europan Ice. Icarus 2000, 147, 292–300. [Google Scholar] [CrossRef]
  30. Marion, G.M.; Mironenko, M.V.; Roberts, M.W. FREZCHEM: A geochemical model for cold aqueous solutions. Comput. Geosci. 2010, 36, 10–15. [Google Scholar] [CrossRef]
  31. Marion, G.M. A molal-based model for strong acid chemistry at low temperatures. Geochim. Cosmochim. Acta 2002, 66, 2499–2516. [Google Scholar] [CrossRef]
  32. McCord, T.B.; Teeter, G.; Hansen, G.B.; Sieger, M.T.; Orlando, T.M. Brines exposed to Europa surface conditions. J. Geophys. Res. E Planets 2002, 107, 4-1–4-6. [Google Scholar] [CrossRef]
  33. Orlando, T.M.; McCord, T.B.; Grieves, G.A. The chemical nature of Europa surface material and the relation to a subsurface ocean. Icarus 2005, 177, 528–533. [Google Scholar] [CrossRef]
  34. Daswani, M.M.; Vance, S.D.; Mayne, M.J.; Glein, C.R. A Metamorphic Origin for Europa’s Ocean. Geophys. Res. Lett. 2021, 48, e2021GL094143. [Google Scholar] [CrossRef]
  35. Glein, C.R.; Waite, J.H. The Carbonate Geochemistry of Enceladus’ Ocean. Geophys. Res. Lett. 2020, 47, e2019GL085885. [Google Scholar] [CrossRef]
  36. Cordier, D.; Mousis, O.; Lunine, J.I.; Lavvas, P.; Vuitton, V. An estimate of the chemical composition of Titan’s lakes. Astrophys. J. 2009, 707, L128–L131. [Google Scholar] [CrossRef]
  37. Postberg, F.; Kempf, S.; Schmidt, J.; Brilliantov, N.; Beinsen, A.; Abel, B.; Buck, U.; Srama, R. Sodium salts in E-ring ice grains from an ocean below the surface of Enceladus. Nature 2009, 459, 1098–1101. [Google Scholar] [CrossRef] [PubMed]
  38. Glein, C.R.; Shock, E.L. A geochemical model of non-ideal solutions in the methane-ethane-propane-nitrogen-acetylene system on Titan. Geochim. Cosmochim. Acta 2013, 115, 217–240. [Google Scholar] [CrossRef]
  39. Robinson, K.J.; Gould, I.R.; Hartnett, H.E.; Williams, L.B.; Shock, E.L. Hydrothermal Experiments with Protonated Benzylamines Provide Predictions of Temperature-Dependent Deamination Rates for Geochemical Modelling. ACS Earth Space Chem. 2021, 5, 1997–2012. [Google Scholar] [CrossRef]
  40. Fukushi, K.; Imai, E.; Sekine, Y.; Kitajima, T.; Gankhurel, B.; Davaasuren, D.; Hasebe, N. In situ formation of monohydrocalcite in alkaline saline lakes of the valley of gobi lakes: Prediction for mg, ca, and total dissolved carbonate concentrations in enceladus’ ocean and alkaline-carbonate ocean worlds. Minerals 2020, 10, 669. [Google Scholar] [CrossRef]
  41. Zolotov, M.Y. Aqueous fluid composition in CI chondritic materials: Chemical equilibrium assessments in closed systems. Icarus 2012, 220, 713–729. [Google Scholar] [CrossRef]
  42. Glein, C.R. Noble gases, nitrogen, and methane from the deep interior to the atmosphere of Titan. Icarus 2015, 250, 570–586. [Google Scholar] [CrossRef]
  43. Affholder, A.; Guyot, F.; Sauterey, B.; Ferrière, R.; Mazevet, S. Bayesian Analysis of Enceladus’s Plume Data to Assess Methanogenesis. Nat. Astron. 2021, 5, 724–729. [Google Scholar] [CrossRef]
  44. Barge, L.M.; White, L.M. Experimentally Testing Hydrothermal Vent Origin of Life on Enceladus and Other Icy/Ocean Worlds. Astrobiology 2017, 17, 820–833. [Google Scholar] [CrossRef] [PubMed]
  45. Bassez, M.-P. Follow the High Subcritical Water. Geosciences 2019, 9, 249. [Google Scholar] [CrossRef]
  46. Borghini, G.; Rampone, E. Postcumulus Processes in Oceanic-Type Olivine-Rich Cumulates: The Role of Trapped Melt Crystallization versus Melt/Rock Interaction. Contrib. Mineral. Petrol. 2007, 154, 619–633. [Google Scholar] [CrossRef]
  47. Duca, Z.A.; Speller, N.C.; Craft, K.L.; Cable, M.L.; Stockton, A.M. Quantitative and compositional analysis of trace amino acids in icy moon analogues using microcapillary electrophoresis laser-induced fluorescence detection. ACS Earth Space Chem. 2022, 6, 379–392. [Google Scholar] [CrossRef]
  48. Fifer, L.M.; Catling, D.C.; Toner, J.D. Chemical Fractionation Modeling of Plumes Indicates a Gas-Rich, Moderately Alkaline Enceladus Ocean. Planet. Sci. J. 2022, 3, 191. [Google Scholar] [CrossRef]
  49. Fones, E.M.; Templeton, A.S.; Mogk, D.W.; Boyd, E.S. Transformation of low-molecular-weight organic acids by microbial endoliths in subsurface mafic and ultramafic igneous rock. Environ. Microbiol. 2022, 24, 4137–4152. [Google Scholar] [CrossRef] [PubMed]
  50. Fox-Powell, M.G.; Cousins, C.R. Partitioning of Crystalline and Amorphous Phases during Freezing of Simulated Enceladus Ocean Fluids. J. Geophys. Res. Planets 2021, 126, e2020JE006628. [Google Scholar] [CrossRef]
  51. Higgins, P.M.; Glein, C.R.; Cockell, C.S. Instantaneous Habitable Windows in the Parameter Space of Enceladus’ Ocean. J. Geophys. Res. Planets 2021, 126, e2021JE006951. [Google Scholar] [CrossRef]
  52. Lang, S.Q.; Brazelton, W.J. Habitability of the marine serpentinite subsurface: A case study of the Lost City hydrothermal field. Philos. Trans. R. Soc. A 2020, 378, 20180429. [Google Scholar] [CrossRef]
  53. Lunine, J.I.; Hörst, S.M. Organic Chemistry on the Surface of Titan. Rend. Lincei 2011, 22, 183–189. [Google Scholar] [CrossRef]
  54. Marion, G.M.; Kargel, J.S.; Catling, D.C.; Lunine, J.I. Modeling Ammonia–Ammonium Aqueous Chemistries in the Solar System’s Icy Bodies. Icarus 2012, 220, 932–946. [Google Scholar] [CrossRef]
  55. Marusiak, A.G.; Vance, S.; Panning, M.P.; Běhounková, M.; Byrne, P.K.; Choblet, G.; Daswani, M.M.; Hughson, K.; Journaux, B.; Lobo, A.H.; et al. Exploration of Icy Ocean Worlds Using Geophysical Approaches. Planet. Sci. J. 2021, 2, 150. [Google Scholar] [CrossRef]
  56. Muñoz-Iglesias, V.; Prieto-Ballesteros, O. Low-Temperature High-Pressure Chemistry of Ammonia and Methanol Aqueous Solutions in the Presence of Different Carbon Sources: Application to Icy Bodies. ACS Earth Space Chem. 2022, 6, 1482–1494. [Google Scholar] [CrossRef]
  57. Neveu, M.; Anbar, A.D.; Davila, A.F.; Glavin, D.P.; MacKenzie, S.M.; Phillips-Lander, C.M.; Sherwood, B.; Takano, Y.; Williams, P.; Yano, H. Returning Samples from Enceladus for Life Detection. Front. Astron. Space Sci. 2020, 7, 26. [Google Scholar] [CrossRef]
  58. Sanloup, C.; Fei, Y. Closure of the Fe–S–Si Liquid Miscibility Gap at High Pressure. Phys. Earth Planet. Inter. 2004, 147, 57–65. [Google Scholar] [CrossRef]
  59. Seckbach, J.; Chela-Flores, J. Astrobiology: From Extremophiles in the Solar System to Extraterrestrial Civilizations. In Astronomy and Civilization; Tymieniecka, A.-T., Grandpierre, A., Eds.; Analecta Husserliana, Vol. CVII; Springer: Dordrecht, The Netherlands, 2011; pp. 237–246. [Google Scholar] [CrossRef]
  60. Tajeddine, R.; Hedman, M.M.; Soderlund, K.M.; Thomas, P.C.; Helfenstein, P.; Hedman, M.M.; Burns, J.A.; Schenk, P.M. True Polar Wander of Enceladus from Topographic Data. Icarus 2017, 295, 46–60. [Google Scholar] [CrossRef]
  61. Theiling, B.P. The Effect of Europa and Enceladus Analog Seawater Composition on Isotopic Measurements of Volatile CO2. Icarus 2021, 358, 114216. [Google Scholar] [CrossRef]
  62. Yu, X.; Chen, H.; Zong, Q.; Wang, J.; Shi, W.; Zou, H.; Zou, J.; Zhong, W.; Chen, Z.; Shao, S.; et al. Mitigating Deep Dielectric Charging Effects in Space. IEEE Trans. Nucl. Sci. 2017, 64, 2822–2828. [Google Scholar] [CrossRef]
  63. Davis, P.; Carney, S.; Boeck, M.; Dunford, B. Ceres in Depth. NASA Solar System Exploration. 23 May 2023. Available online: https://web.archive.org/web/20230523214232/https://solarsystem.nasa.gov/planets/dwarf-planets/ceres/in-depth/ (accessed on 27 January 2026).
  64. Castillo-Rogez, J.C.; Neveu, M.; Scully, J.E.C.; House, C.H.; Quick, L.C.; Bouquet, A.; Miller, K.; Bland, M.; De Sanctis, M.C.; Ermakov, A.I.; et al. Ceres: Astrobiological Target and Possible Ocean World. Astrobiology 2020, 20, 269–291. [Google Scholar] [CrossRef]
  65. De Sanctis, M.C.; Mitri, G.; Castillo-Rogez, J.C.; House, C.H.; Marchi, S.; Raymond, C.A.; Sekine, Y. Relict Ocean Worlds: Ceres. Space Sci. Rev. 2020, 216, 60. [Google Scholar] [CrossRef]
  66. Castillo-Rogez, J.C.; McCord, T.B. Ceres’ evolution and present state constrained by shape data. Icarus 2010, 205, 443–459. [Google Scholar] [CrossRef]
  67. NASA. Jupiter in Depth. NASA Solar System Exploration. 2023. Available online: https://solarsystem.nasa.gov/planets/jupiter/in-depth/ (accessed on 27 January 2026).
  68. NASA. Jupiter—Overview. NASA Solar System Exploration. 2023. Available online: https://solarsystem.nasa.gov/planets/jupiter/overview/ (accessed on 27 January 2026).
  69. Davis, P.; Carney, S.; Dunford, B.; Boeck, M. Ganymede–Overview. NASA Solar System Exploration. 22 March 2023. Available online: https://web.archive.org/web/20230326132540/https://solarsystem.nasa.gov/moons/jupiter-moons/ganymede/overview/ (accessed on 27 January 2026).
  70. NASA. Io—Overview. NASA Solar System Exploration. 2023. Available online: https://web.archive.org/web/20230325170120/https://solarsystem.nasa.gov/moons/jupiter-moons/io/overview/ (accessed on 27 January 2026).
  71. Davis, P.; Carney, S.; Dunford, B.; Boeck, M. Europa—Overview. NASA Solar System Exploration. 22 March 2023. Available online: https://web.archive.org/web/20230326040431/https://solarsystem.nasa.gov/moons/jupiter-moons/europa/overview/ (accessed on 27 January 2026).
  72. Carlson, R.W.; Johnson, R.E.; Anderson, M.S. Sulfuric acid on Europa and the radiolytic sulfur cycle. Science 1999, 286, 97–99. [Google Scholar] [CrossRef]
  73. Kargel, J.S.; Kaye, J.Z.; Head, J.W.; Marion, G.M.; Sassen, R.; Crowley, J.K.; Ballesteros, O.P.; Grant, S.A.; Hogenboom, D.L. Europa’s Crust and Ocean: Origin, Composition, and the Prospects for Life. Icarus 2000, 148, 226–265. [Google Scholar] [CrossRef]
  74. Vance, S.D.; Craft, K.L.; Shock, E.; Schmidt, B.E.; Lunine, J.; Hand, K.P.; McKinnon, W.B.; Spiers, E.M.; Chivers, C.; Lawrence, J.D.; et al. Investigating Europa’s Habitability with the Europa Clipper. Space Sci. Rev. 2023, 219, 81. [Google Scholar] [CrossRef]
  75. Lyu, Z.; Shao, N.; Akinyemi, T.; Whitman, W.B. Methanogenesis. Curr. Biol. CB 2018, 28, R727–R732. [Google Scholar] [CrossRef]
  76. Williams, D.R. Saturn Fact Sheet. NASA. 2023. Available online: https://nssdc.gsfc.nasa.gov/planetary/factsheet/saturnfact.html (accessed on 27 January 2026).
  77. Davis, P.; Carney, S.; Dunford, B.; Boeck, M. Saturn Moons. NASA Solar System Exploration. 2023. Available online: https://web.archive.org/web/20230320131033/https://solarsystem.nasa.gov/moons/saturn-moons/overview/?page=0&per_page=40&order=name+asc&search=&placeholder=Enter+moon+name&condition_1=38%3Aparent_id&condition_2=moon%3Abody_type%3Ailike (accessed on 27 January 2026).
  78. Glein, C.R.; Baross, J.A.; Waite, J.H. The pH of Enceladus’ ocean. Geochim. Cosmochim. Acta 2015, 162, 202–219. [Google Scholar] [CrossRef]
  79. Hao, J.; Glein, C.R.; Huang, F.; Yee, N.; Catling, D.C.; Postberg, F.; Hillier, J.K.; Hazen, R.M. Abundant phosphorus expected for possible life in Enceladus’s ocean. Proc. Natl. Acad. Sci. USA 2022, 119, e2201388119. [Google Scholar] [CrossRef] [PubMed]
  80. Fukushi, K.; Matsumiya, H. Control of Water Chemistry in Alkaline Lakes: Solubility of Monohydrocalcite and Amorphous Magnesium Carbonate in CaCl2-MgCl2-Na2CO3 Solutions. ACS Earth Space Chem. 2018, 2, 735–744. [Google Scholar] [CrossRef]
  81. Nissen, S.B.; Taubner, R.; Leitner, J.J.; Firneis, M.G. On the possibility of Serpentinization on Enceladus. EPSC Abstr. 2015, 10, 644. Available online: https://meetingorganizer.copernicus.org/EPSC2015/EPSC2015-644.pdf (accessed on 27 January 2026).
  82. Wolery, T.J. EQ3/6 A Software Package for Geochemical Modelling; Computer Software; U.S. Department of Energy Office of Scientific and Technical Information (USDOE): Washington, DC, USA, 2010. [CrossRef]
  83. Glein, C.R.; Desch, S.J.; Shock, E.L. The absence of endogenic methane on Titan and its implications for the origin of atmospheric nitrogen. Icarus 2009, 204, 637–644. [Google Scholar] [CrossRef]
  84. Dubouloz, N.; Raulin, F.; Lellouch, E.; Gautier, D. Titan’s hypothesized ocean properties: The influence of surface temperature and atmospheric composition uncertainties. Icarus 1989, 82, 81–96. [Google Scholar] [CrossRef]
  85. Lellouch, E.; Coustenis, A.; Gautier, D.; Raulin, F.; Dubouloz, N.; Frère, C. Titan’s atmosphere and hypothesized ocean: A reanalysis of the Voyager 1 radio-occultation and IRIS 7.7-μm data. Icarus 1989, 79, 328–349. [Google Scholar] [CrossRef]
  86. Williams, D.R. Uranus Fact Sheet. NASA. 2024. Available online: https://web.archive.org/web/20240923103204/https://nssdc.gsfc.nasa.gov/planetary/factsheet/uranusfact.html (accessed on 27 January 2026).
  87. Ladd, D.; Lentz, M.; Simon, A.A.; Kim, K.; McElligott, J.; Lepsch, A.E. Exploring Planet Uranus. NASA Scientific Visualization Studio. 15 May 2024. Available online: https://web.archive.org/web/20241001140906/https://svs.gsfc.nasa.gov/14580 (accessed on 27 January 2026).
  88. NASA. All About Uranus. NASA Space Place. 2024. Available online: https://web.archive.org/web/20241001133624/https://spaceplace.nasa.gov/all-about-uranus/en/ (accessed on 27 January 2026).
  89. NASA. Uranus. NASA Science. 2024. Available online: https://web.archive.org/web/20241001135248/https://science.nasa.gov/uranus/ (accessed on 27 January 2026).
  90. Fagents, S.A. Considerations for effusive cryovolcanism on Europa: The post-Galileo perspective. J. Geophys. Res. Planets 2003, 108, 5139. [Google Scholar] [CrossRef]
  91. Cohen, I.J.; Turner, D.L.; Kollmann, P.; Clark, G.B.; Hill, M.E.; Regoli, L.H.; Gershman, D.J. A Localized and Surprising Source of Energetic Ions in the Uranian Magnetosphere between Miranda and Ariel. Geophys. Res. Lett. 2023, 50, e2022GL101998. [Google Scholar] [CrossRef]
  92. Davis, P.; Carney, S.; Boeck, M.; Dunford, B. Neptune in Depth. NASA Solar System Exploration. 19 May 2023. Available online: https://solarsystem.nasa.gov/planets/neptune/in-depth/ (accessed on 27 January 2026).
  93. Williams, D.R. Neptune Fact Sheet. NASA Goddard Space Flight Center. 2023. Available online: https://nssdc.gsfc.nasa.gov/planetary/factsheet/neptunefact.html (accessed on 27 January 2026).
  94. Kirk, R.L.; Soderblom, L.A.; Brown, R.H.; Kieffer, S.W.; Kargel, J.S. Triton’s plumes: Discovery, characteristics, and models. In Neptune and Triton; Cruikshank, D.P., Matthews, M.S., Schumann, A.M., Eds.; University of Arizona Press: Tucson, AZ, USA, 1995; Volume 1, pp. 949–989. Available online: https://www.usgs.gov/publications/tritons-plumes-discovery-characteristics-and-models (accessed on 27 January 2026).
  95. Smith, B.A.; Soderblom, L.A.; Banfield, D.; Barnet, C.; Basilevksy, A.T.; Beebe, R.F.; Bollinger, K.; Boyce, J.M.; Brahic, A.; Briggs, G.A.; et al. Voyager 2 at Neptune: Imaging Science Results. Science 1989, 246, 1422–1449. [Google Scholar] [CrossRef]
  96. Bertrand, T.; Lellouch, E.; Holler, B.J.; Young, L.A.; Schmitt, B.; Marques Oliveira, J.; Sicardy, B.; Forget, F.; Grundy, W.M.; Merlin, F.; et al. Volatile Transport Modeling on Triton with New Observational Constraints. Icarus 2022, 373, 114764. [Google Scholar] [CrossRef]
  97. Soderblom, L.A.; Kieffer, S.W.; Becker, T.L.; Brown, R.H.; Cook, A.F.; Hansen, C.J.; Johnson, T.V.; Kirk, R.L.; Shoemaker, E.M. Triton’s Geyser-Like Plumes: Discovery and Basic Characterization. Science 1990, 250, 410–415. [Google Scholar] [CrossRef]
  98. Lichtenberg, T.; Shorttle, O.; Teske, J.; Kempton, E.M.R. Constraining exoplanet interiors using observations of their atmospheres. Science 2025, 390, eads3360. [Google Scholar] [CrossRef]
  99. Glein, C.R.; Yu, X.; Luu, C.N. Deciphering Sub-Neptune Atmospheres: New Insights from Geochemical Models of TOI-270 d. Astrophys. J. 2025, 985, 187. [Google Scholar] [CrossRef]
  100. Davis, P.; Carney, S. Pluto—Overview. NASA Solar System Exploration. 19 May 2023. Available online: https://solarsystem.nasa.gov/planets/dwarf-planets/pluto/ (accessed on 27 January 2026).
  101. Williams, D.R. Pluto Fact Sheet. NASA Goddard Space Flight Center. 2023. Available online: https://nssdc.gsfc.nasa.gov/planetary/factsheet/plutofact.html (accessed on 27 January 2026).
  102. Moore, J.M.; McKinnon, W.B.; Spencer, J.R.; Howard, A.D.; Schenk, P.M.; Beyer, R.A.; Nimmo, F.; Singer, K.N.; Umurhan, O.M.; White, O.L.; et al. The Geology of Pluto and Charon Through the Eyes of New Horizons. Science 2016, 351, 1284–1293. [Google Scholar] [CrossRef]
  103. Hare, D.E.; Sorensen, C.M. The density of supercooled water. II. Bulk samples cooled to the homogeneous nucleation limit. J. Chem. Phys. 1987, 87, 4840–4845. [Google Scholar] [CrossRef]
  104. Mishima, O. Volume of supercooled water under pressure and the liquid-liquid critical point. J. Chem. Phys. 2010, 133, 144503. [Google Scholar] [CrossRef]
  105. International Association for the Properties of Water and Steam. Revised Release on the IAPWS Formulation 1995 for the Thermodynamic Properties of Ordinary Water Substance for General and Scientific Use (2018). IAPWS R6-95(2018). Available online: https://iapws.org/documents/release/IAPWS-95 (accessed on 27 January 2026).
  106. Wagner, W.; Pruß, A. The IAPWS Formulation 1995 for the Thermodynamic Properties of Ordinary Water Substance for General and Scientific Use. J. Phys. Chem. Ref. Data 2002, 31, 387–535. [Google Scholar] [CrossRef]
  107. Holten, V.; Bertrand, C.E.; Anisimov, M.A.; Sengers, J.V. Thermodynamics of supercooled water. J. Chem. Phys. 2012, 136, 094507. [Google Scholar] [CrossRef]
  108. Zolotov, M.Y. An oceanic composition on early and today’s Enceladus. Geophys. Res. Lett. 2007, 34, 23203. [Google Scholar] [CrossRef]
  109. McCord, T.B.; Hansen, G.B.; Fanale, F.P.; Carlson, R.W.; Matson, D.L.; Johnson, T.V.; Smythe, W.D.; Crowley, J.K.; Martin, P.D.; Ocampo, A.; et al. Salts on Europa’s surface detected by Galileo’s near infrared map-ping spectrometer. The NIMS Team. Science 1998, 280, 1242–1245. [Google Scholar] [CrossRef] [PubMed]
  110. McCord, T.B.; Hansen, G.B.; Clark, R.N.; Martin, P.D.; Hibbitts, C.A.; Fanale, F.P.; Granahan, J.C.; Segura, M.; Matson, D.L.; Johnson, T.V.; et al. Non-water-ice constituents in the surface material of the icy Galilean satellites from the Galileo near-infrared mapping spectrometer investigation. J. Geophys. Res. Planets 1998, 103, 8603–8626. [Google Scholar] [CrossRef]
  111. McCord, T.B.; Hansen, G.B.; Matson, D.L.; Johnson, T.V.; Crowley, J.K.; Fanale, F.P.; Carlson, R.W.; Smythe, W.D.; Martin, P.D.; Hibbitts, C.A.; et al. Hydrated salt minerals on Europa’s Surface from the Galileo near-infrared mapping spectrometer (NIMS) investigation. J. Geophys. Res. Planets 1999, 104, 11827–11851. [Google Scholar] [CrossRef]
  112. Cipriani, F.; Leblanc, F.; Witasse, O.; Johnson, R.E. Exospheric signatures of alkali abundances in Europa’s regolith. Geophys. Res. Lett. 2009, 36, 12202. [Google Scholar] [CrossRef]
  113. Leliwa-Kopystyński, J.; Maruyama, M.; Nakajima, T. The Water–Ammonia Phase Diagram up to 300 MPa: Application to Icy Satellites. Icarus 2002, 159, 518–528. [Google Scholar] [CrossRef]
  114. Vu, T.H.; Hodyss, R.; Choukroun, M.; Johnson, P.V. Chemistry of frozen sodium–magnesium–sulfate–chloride brines: Implications for surface expression of Europa’s ocean composition. Astrophys. J. Lett. 2016, 816, L26. [Google Scholar] [CrossRef]
  115. NASA; JPL-Caltech; UCLA; MPS; DLR; IDA. Ceres in Color. NASA Solar System Exploration. 2016. Available online: https://web.archive.org/web/20230523170143/https://solarsystem.nasa.gov/resources/622/ceres-in-color/ (accessed on 27 January 2026).
  116. Andreoli, C.; Braun, H.; Villard, R.; Simon, A.; Wong, M.H.; Garner, R. Hubble Captures Crisp New Portrait of Jupiter’s Storms. NASA. 2020. Available online: https://www.nasa.gov/feature/goddard/2020/hubble-captures-crisp-new-portrait-of-jupiters-storms (accessed on 27 January 2026).
  117. NASA. Jupiter’s Moons: Family Portrait. NASA Solar System Exploration. 2007. Available online: https://web.archive.org/web/20230325143537/https://solarsystem.nasa.gov/resources/2660/jupiters-moons-family-portrait/?category=moons%2Fjupiter-moons_callisto (accessed on 27 January 2026).
  118. Andreoli, C.; Villard, R.; Simon, A.; Wong, M.H.; Garner, R. Hubble Sees Summertime on Saturn. NASA. 2020. Available online: https://www.nasa.gov/feature/goddard/2020/hubble-sees-summertime-on-saturn (accessed on 27 January 2026).
  119. Sromovsky, L. PIA17306: Keck Telescope Views of Uranus. NASA. 2004. Available online: https://web.archive.org/web/20240927015204/https://photojournal.jpl.nasa.gov/catalog/PIA17306 (accessed on 27 January 2026).
  120. NASA; JPL. Neptune Full Disk View. NASA Solar System Exploration. 1998. Available online: https://web.archive.org/web/20230523031947/https://solarsystem.nasa.gov/resources/611/neptune-full-disk-view/?category=planets_neptune (accessed on 27 January 2026).
  121. NASA; ESA; Showalter, M. (SETI Institute). Neptune’s Moon Hippocamp. NASA Solar System Exploration. 2019. Available online: https://web.archive.org/web/20230523031938/https://solarsystem.nasa.gov/resources/2289/neptunes-moon-hippocamp/?category=planets_neptune (accessed on 27 January 2026).
  122. NASA; JPL; USGS. Montage of Neptune and Triton. NASA Solar System Exploration. 1998. Available online: https://web.archive.org/web/20230523033313/https://solarsystem.nasa.gov/resources/221/montage-of-neptune-and-triton/?category=planets_neptune (accessed on 27 January 2026).
  123. NASA; JHUAPL; SwRI. Pluto and Charon: Strikingly Different Worlds. NASA Solar System Exploration. 2015. Available online: https://web.archive.org/web/20230523034715/https://solarsystem.nasa.gov/resources/697/pluto-and-charon-strikingly-different-worlds/ (accessed on 27 January 2026).
Figure 1. The solar system: the Sun, planets, dwarf planets, and moons are at scale, and their distance scale is at the bottom. The largest moons are shown by the proximity of their orbits. Reprinted from [12].
Figure 1. The solar system: the Sun, planets, dwarf planets, and moons are at scale, and their distance scale is at the bottom. The largest moons are shown by the proximity of their orbits. Reprinted from [12].
Encyclopedia 06 00038 g001
Figure 2. PRISMA 2020 workflow diagram for study selection; screening and eligibility used reinforced computational thematic filtering (CBAT), as described in Section 2.3.
Figure 2. PRISMA 2020 workflow diagram for study selection; screening and eligibility used reinforced computational thematic filtering (CBAT), as described in Section 2.3.
Encyclopedia 06 00038 g002
Figure 3. The three-field graph (Sankey diagram) of studies on the geochemical modeling of space objects beyond Mars in the solar system (left), their main keywords (middle) and keywords plus (right). Note that it is just a preview of the original Sankey diagram to indicate its size and complexity. An 8K resolution version of this diagram, for visualization of all details, is available in the Supplementary Materials.
Figure 3. The three-field graph (Sankey diagram) of studies on the geochemical modeling of space objects beyond Mars in the solar system (left), their main keywords (middle) and keywords plus (right). Note that it is just a preview of the original Sankey diagram to indicate its size and complexity. An 8K resolution version of this diagram, for visualization of all details, is available in the Supplementary Materials.
Encyclopedia 06 00038 g003
Figure 4. Sankey diagram of studies [8,10,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62] on the geochemical modeling of space objects beyond Mars in the solar system (left) and their connection to the target keywords and most common keywords (right). Note that custom-developed data analysis tools are utilized to enhance the identification of the most relevant publications for target keywords, and these keywords are not essentially author keywords of these publications. Additionally, in this graph, the geochemical model represents the “geochemical model*”, referring to possible variations (e.g., ‘modeling’ vs. ‘modelling’).
Figure 4. Sankey diagram of studies [8,10,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62] on the geochemical modeling of space objects beyond Mars in the solar system (left) and their connection to the target keywords and most common keywords (right). Note that custom-developed data analysis tools are utilized to enhance the identification of the most relevant publications for target keywords, and these keywords are not essentially author keywords of these publications. Additionally, in this graph, the geochemical model represents the “geochemical model*”, referring to possible variations (e.g., ‘modeling’ vs. ‘modelling’).
Encyclopedia 06 00038 g004
Figure 5. Decision-style schematic of Ceres’ core evolution linking initial formation time to heat budget, hydration, permeability, cooling efficiency, and present-day structure. Early formation (<5 Myr) with strong 26Al heating promotes hydrothermal activity, silicate hydration, and reduced core permeability, which can limit cooling and lead to dehydration/melting and metallic core formation. Late formation (>5 Myr) yields dry silicates initially; hydration, if any, occurs later along cooling-induced cracks. The dashed oval is meant to indicate the cross-sectional plane in the third dimension. The question mark (?) indicated an uncertainty based on the original figure’s description. Data from Castillo-Rogez & McCord [66].
Figure 5. Decision-style schematic of Ceres’ core evolution linking initial formation time to heat budget, hydration, permeability, cooling efficiency, and present-day structure. Early formation (<5 Myr) with strong 26Al heating promotes hydrothermal activity, silicate hydration, and reduced core permeability, which can limit cooling and lead to dehydration/melting and metallic core formation. Late formation (>5 Myr) yields dry silicates initially; hydration, if any, occurs later along cooling-induced cracks. The dashed oval is meant to indicate the cross-sectional plane in the third dimension. The question mark (?) indicated an uncertainty based on the original figure’s description. Data from Castillo-Rogez & McCord [66].
Encyclopedia 06 00038 g005
Figure 6. Computed fluid composition throughout mixing hydrothermal fluid with reduced seawater (left). Only methanogenesis-related components are presented. Estimated fluid composition throughout mixing hydrothermal fluid with oxidized seawater (right). Only the methanogenesis-related components and sulfate reduction are presented. In both figures, the top axis represents the equivalent mixing ratios of seawater to hydrothermal fluid at each temperature and C O 2 t o t a l includes C O 2 a q and H C O 3 . S O 4 t o t a l includes S O 4 2 , H S O 4 , and various aqueous sulfate complexes (e.g., M g S O 4 aq and N a S O 4 ). Reprinted with permission from ref. [28]. Copyright 1999 John Wiley and Sons.
Figure 6. Computed fluid composition throughout mixing hydrothermal fluid with reduced seawater (left). Only methanogenesis-related components are presented. Estimated fluid composition throughout mixing hydrothermal fluid with oxidized seawater (right). Only the methanogenesis-related components and sulfate reduction are presented. In both figures, the top axis represents the equivalent mixing ratios of seawater to hydrothermal fluid at each temperature and C O 2 t o t a l includes C O 2 a q and H C O 3 . S O 4 t o t a l includes S O 4 2 , H S O 4 , and various aqueous sulfate complexes (e.g., M g S O 4 aq and N a S O 4 ). Reprinted with permission from ref. [28]. Copyright 1999 John Wiley and Sons.
Encyclopedia 06 00038 g006
Figure 7. pH versus temperature for two hypothetical Europa brines (MgSO4–H2SO4–H2O and Na2SO4–MgSO4–H2SO4–H2O). Data from Marion [31].
Figure 7. pH versus temperature for two hypothetical Europa brines (MgSO4–H2SO4–H2O and Na2SO4–MgSO4–H2SO4–H2O). Data from Marion [31].
Encyclopedia 06 00038 g007
Figure 8. Europa’s ice-shell brine system: (top x-axis) ionic molalities (Na+, Mg2+, Ca2+, Cl, SO42−, aH2O) vs. ice-layer depth; (bottom x-axis) temperature (K) and pressure (bar) vs. depth. Recomposed and re-plotted by the authors. Data from Marion et al. [27].
Figure 8. Europa’s ice-shell brine system: (top x-axis) ionic molalities (Na+, Mg2+, Ca2+, Cl, SO42−, aH2O) vs. ice-layer depth; (bottom x-axis) temperature (K) and pressure (bar) vs. depth. Recomposed and re-plotted by the authors. Data from Marion et al. [27].
Encyclopedia 06 00038 g008
Figure 9. Progression of ionic concentrations and precipitated minerals during thermodynamic freezing for five initial solutions: Oxidized Salty, Saturated, Equimolar, and the experimentally verified Pathways A and B. Bars show Na+, Cl, Mg2+, SO42−, and H+ concentrations at the initial solution (init) and subsequent stages; mineral phases identified at each stage are annotated above the bars. Plotted by the authors. Data from Johnson et al. [10].
Figure 9. Progression of ionic concentrations and precipitated minerals during thermodynamic freezing for five initial solutions: Oxidized Salty, Saturated, Equimolar, and the experimentally verified Pathways A and B. Bars show Na+, Cl, Mg2+, SO42−, and H+ concentrations at the initial solution (init) and subsequent stages; mineral phases identified at each stage are annotated above the bars. Plotted by the authors. Data from Johnson et al. [10].
Encyclopedia 06 00038 g009
Figure 10. Europa’s theoretical ocean from the seafloor to the surface. (a) Precipitation of minerals and exsolving gases with reducing water depth. (b) The sum of substances distributed among all species and dissolved in water. The dissolved components shown here are the sum of those components distributed among all species in the solution, e.g., ∑C is the total carbon in aqueous HCO3, CH4, CO2, and organics. Minimum represented concentrations:10−5 mol/kg. (c) pH, and (d) redox potential for retained-to-extracted ratio (R/E) = 0 models of models made of CI carbonaceous chondrites (EM-CI), CM chondrites (EM-CM), and Monte Carlo (MC)-Scale. Solid and dashed horizontal lines show the pressure at the base of a current 5 and 30 km ice shell, respectively. Reprinted from [34].
Figure 10. Europa’s theoretical ocean from the seafloor to the surface. (a) Precipitation of minerals and exsolving gases with reducing water depth. (b) The sum of substances distributed among all species and dissolved in water. The dissolved components shown here are the sum of those components distributed among all species in the solution, e.g., ∑C is the total carbon in aqueous HCO3, CH4, CO2, and organics. Minimum represented concentrations:10−5 mol/kg. (c) pH, and (d) redox potential for retained-to-extracted ratio (R/E) = 0 models of models made of CI carbonaceous chondrites (EM-CI), CM chondrites (EM-CM), and Monte Carlo (MC)-Scale. Solid and dashed horizontal lines show the pressure at the base of a current 5 and 30 km ice shell, respectively. Reprinted from [34].
Encyclopedia 06 00038 g010
Figure 11. Enceladus’ thermodynamically favored forms of dissolved phosphorus as a function of pH and equilibrium oxidation state, represented by the activity of dissolved hydrogen or the fugacity of hydrogen gas. At 0 °C and 70 bar (1 bar for reference in dashed lines). Reprinted from [79].
Figure 11. Enceladus’ thermodynamically favored forms of dissolved phosphorus as a function of pH and equilibrium oxidation state, represented by the activity of dissolved hydrogen or the fugacity of hydrogen gas. At 0 °C and 70 bar (1 bar for reference in dashed lines). Reprinted from [79].
Encyclopedia 06 00038 g011
Figure 12. Logarithmic relationship of carbonate (CO32−) and calcium (Ca2+, black) or magnesium (Mg2+, red) activities in Olgoy, Boon Tsagaan, and Orog Lakes, compared to alkaline lakes in earth. The experimentally estimated solubility of monohydrocalcite (MHC, black line) and amorphous magnesium carbonate (AMC, red line) illustrate how mineral solubility controls the water chemistry [40,80]. Reprinted from [40].
Figure 12. Logarithmic relationship of carbonate (CO32−) and calcium (Ca2+, black) or magnesium (Mg2+, red) activities in Olgoy, Boon Tsagaan, and Orog Lakes, compared to alkaline lakes in earth. The experimentally estimated solubility of monohydrocalcite (MHC, black line) and amorphous magnesium carbonate (AMC, red line) illustrate how mineral solubility controls the water chemistry [40,80]. Reprinted from [40].
Encyclopedia 06 00038 g012
Figure 13. Geochemical modeling with the EQ3/6 software package to explore Enceladus’ core composition: an example of chemical evolution and expected minerals as a function of the reaction path at 25 bars, 0 °C, pH = 9, and 1.0 mol fayalite, assuming the chemical composition of Enceladus’ subsurface sea and plumes are similar. Blue: Carbonates, Green: Silicates, Yellow: Oxides, Red: Serpentines, Black: Graphite (C), and Hydrogen. Reprinted from [81].
Figure 13. Geochemical modeling with the EQ3/6 software package to explore Enceladus’ core composition: an example of chemical evolution and expected minerals as a function of the reaction path at 25 bars, 0 °C, pH = 9, and 1.0 mol fayalite, assuming the chemical composition of Enceladus’ subsurface sea and plumes are similar. Blue: Carbonates, Green: Silicates, Yellow: Oxides, Red: Serpentines, Black: Graphite (C), and Hydrogen. Reprinted from [81].
Encyclopedia 06 00038 g013
Figure 14. CH4, C2H6, and N2 liquid-phase mole fractions plotted against total pressure for the CH4–C2H6–N2 system at 95 K. Values in parentheses in the source denote vapor-phase mole fractions and are not plotted here. Pressure uncertainty is ±0.01 bar. Composition uncertainty is ±0.003 mole fraction. Data from Glein & Shock [38].
Figure 14. CH4, C2H6, and N2 liquid-phase mole fractions plotted against total pressure for the CH4–C2H6–N2 system at 95 K. Values in parentheses in the source denote vapor-phase mole fractions and are not plotted here. Pressure uncertainty is ±0.01 bar. Composition uncertainty is ±0.003 mole fraction. Data from Glein & Shock [38].
Encyclopedia 06 00038 g014
Figure 15. Small-multiple bar charts summarizing physical and thermodynamic properties of Titan-relevant compounds at Pref = 1.467 bar and Tref = 90.6941 K. Panels show: critical volume (cm3 mol−1), liquid molar volume at Pref, Tref (cm3 mol−1), enthalpy of vaporization at Tref (J mol−1), liquid fugacity at Pref, Tref (bar; log scale), and ΔG°f for the gas and liquid phases (kJ mol−1). Data from Glein & Shock [38].
Figure 15. Small-multiple bar charts summarizing physical and thermodynamic properties of Titan-relevant compounds at Pref = 1.467 bar and Tref = 90.6941 K. Panels show: critical volume (cm3 mol−1), liquid molar volume at Pref, Tref (cm3 mol−1), enthalpy of vaporization at Tref (J mol−1), liquid fugacity at Pref, Tref (bar; log scale), and ΔG°f for the gas and liquid phases (kJ mol−1). Data from Glein & Shock [38].
Encyclopedia 06 00038 g015
Figure 16. Molality of key volatiles with canonical starting abundances (left panels) and volumic vapor fraction (right panels) as a function of depth for a Charon-like KBO in three scenarios: (a) no crust, dry core; (b) crust, dry core; (c) crust, hydrated core. Left panels: The degree to which a species exsolves at a given depth can be estimated by comparing its saturation molality (curve) to its starting concentration, indicated by an arrow at the top of each panel. Low discrepancies indicate that the species is mostly in solution, and high discrepancies indicate exsolution. In case (a), where the neutral buoyancy level (NBL) of water in ice is defined, it is indicated by a dashed line marked “NBL” beyond which calculations were not attempted since Pgas = 0. Right panels: xvap is plotted for the canonical run (thick line), as well as runs with half and twice the abundances (marked “/2” and “×2”, respectively). The limit for crack propagation in ice is indicated as a thin vertical line. Reprinted with permission from ref. [26]. Copyright 2015 Elsevier.
Figure 16. Molality of key volatiles with canonical starting abundances (left panels) and volumic vapor fraction (right panels) as a function of depth for a Charon-like KBO in three scenarios: (a) no crust, dry core; (b) crust, dry core; (c) crust, hydrated core. Left panels: The degree to which a species exsolves at a given depth can be estimated by comparing its saturation molality (curve) to its starting concentration, indicated by an arrow at the top of each panel. Low discrepancies indicate that the species is mostly in solution, and high discrepancies indicate exsolution. In case (a), where the neutral buoyancy level (NBL) of water in ice is defined, it is indicated by a dashed line marked “NBL” beyond which calculations were not attempted since Pgas = 0. Right panels: xvap is plotted for the canonical run (thick line), as well as runs with half and twice the abundances (marked “/2” and “×2”, respectively). The limit for crack propagation in ice is indicated as a thin vertical line. Reprinted with permission from ref. [26]. Copyright 2015 Elsevier.
Encyclopedia 06 00038 g016
Table 1. The solar system’s planets and the most famous dwarf planet. Reprinted from [13].
Table 1. The solar system’s planets and the most famous dwarf planet. Reprinted from [13].
MercuryVenus Earth MarsJupiter Saturn Uranus Neptune Pluto *
Mass (1024 kg)0.334.875.970.642189856886.81020.013
Diameter (km)487912,10412,7566792142,984120,53651,11849,5282376
Density (kg/m3)54295243551439341326687127016381850
Gravity (m/s2)3.78.99.83.723.198.7110.7
Length of Day (hr)4222.628022424.79.910.717.216.1153.3
Distance from Sun (106 km)57.9108.2149.6228778.51432286745155906.4
Orbital Period (days)88224.7365.2687433110,74730,58959,80090,560
Orbital Velocity (km/s)47.43529.824.113.19.76.85.44.7
Mean Temperature (C)16746415−65−110−140−195−200−225
Surface Pressure (bars)09210.01UNKN **UNKN **UNKN **UNKN **0.00001
Number of Moons001295 ***83 ***2715 ****5
Ring SystemNoNoNoNoYesYesYesYesNo
Magnetic FieldYesNoYesNoYesYesYesYesUNKN
* Once designated our solar system’s ninth planet, Pluto is now considered a notable dwarf planet within the Kuiper Belt. ** Unknown pressure and location of surfaces for Jupiter, Saturn, Uranus, and Neptune due to their atmospheric thickness [14]. *** Recent observational surveys show that Jupiter has 95 confirmed moons, while Saturn has approximately 145–146 confirmed moons, based on the most up-to-date peer-reviewed astronomical analyses [15,16]. **** S/2004N1 is a relatively newly identified Neptune moon (Table 2).
Table 2. The list of space objects of the solar system beyond Mars.
Table 2. The list of space objects of the solar system beyond Mars.
Celestial BodyMoons/Natural Satellites
PlanetsMarsPhobos, Deimos
JupiterIo, Europa, Ganymede, Callisto, Amalthea, Himalia, Elara, Pasiphae, Sinope, Lysithea, Carme, Ananke, Leda, Thebe, Adrastea, Metis, Callirrhoe, Themisto, Megaclite, Taygete, Chaldene, Harpalyke, Kalyke, Iocaste, Erinome, Isonoe, Praxidike, Autonoe, Thyone, Hermippe, Aitne, Eurydome, Euanthe, Euporie, Orthosie, Sponde, Kale, Pasithee, Hegemone, Mneme, Aoede, Thelxinoe, Arche, Kallichore, Helike, Carpo, Eukelade, Cyllene, Kore, Herse, S/2003J19, S/2003J12, S/2003J15, S/2003J16, S/2003J18, S/2010J1, S/2010J2, S/2011J1, S/2011J2, S/2000J11, S/2003J2, S/2003J3, S/2003J4, S/2003J5, S/2003J9, S/2003J10, S/2003J23
SaturnMimas, Enceladus, Tethys, Dione, Rhea, Titan, Hyperion, Iapetus, Phoebe, Janus, Epimetheus, Helene, Telesto, Calypso, Atlas, Prometheus, Pandora, Pan, Methone, Pallene, Polydeuces, Daphnis, Anthe, Aegaeon, Ymir, Paaliaq, Tarvos, Ijiraq, Suttungr, Kiviuq, Mundilfari, Albiorix, Skathi, Erriapus, Siarnaq, Thrymr, Narvi, Aegir, Bebhionn, Bergelmir, Bestla, Farbauti, Fenrir, Fornjot, Hati, Hyrrokkin, Kari, Loge, Skoll, Surtur, Jarnsaxa, Greip, Tarqeq, S/2004S7, S/2004S12, S/2004S13, S/2004S17, S/2006S1, S/2006S3, S/2007S2, S/2007S3
UranusMiranda, Puck, Caliban, Sycorax, Prospero, Setebos, Stephano, Trinculo, Francisco, Margaret, Ferdinand, Perdita, Mab, Cupid, Ariel, Umbriel, Titania, Oberon, Miranda, Cordelia, Ophelia, Bianca, Cressida, Desdemona, Juliet, Portia, Rosalind, Belinda
NeptuneNaiad, Thalassa, Despina, Galatea, Larissa, Hippocamp, Proteus, Triton, Nereid, Halimede, Sao, Laomedeia, Psamathe, Neso, S/2004N1
Dwarf planetsPlutoCharon, Nix, Hydra, Kerberos, Styx
Ceres-
Table 3. Examples of related keywords and phrases for identifying studies about geochemical modeling for celestial bodies.
Table 3. Examples of related keywords and phrases for identifying studies about geochemical modeling for celestial bodies.
CategoryKeywords/PhrasesImportance
Fundamental ConceptsGeochemistry, Astrochemistry, Thermodynamics, Chemical Equilibrium, Geochemical Kinetics, Reaction Kinetics, Geochemical Reaction Paths, Chemical Speciation, Mineral Saturation State, Ionic Strength, Solute TransportThe fundamental concepts of geochemical modeling are critical for understanding how chemical reactions occur in various environments, including those of celestial bodies.
Tools and ModelsGeochemical Modeling, Geochemical Software, Thermodynamic Databases, Geochemical Data Analysis, FREZCHEM, HKF Model, CHNOSZ, PHREEQC, EQ3/6, GWB, MINTEQ, PHASTThe tools, databases, and models used in geochemical modeling studies. These models and tools are commonly used to understand and predict geochemical processes, and can be applied to a variety of environments, including those of celestial bodies.
Geochemical ProcessesGeochemical Transport Processes, Mineral Formation, Mineral Dissolution and Precipitation, Weathering Processes, Redox Reactions, Acid–Base Equilibria, Water–Rock InteractionGeochemical processes and reactions that could be occurring in a geochemical environment, including those of celestial bodies.
Advanced ConceptsSorption Isotherms, Surface Complexation Models, Geochemical Mass Balance, Radiogenic Isotope Geochemistry, Stable Isotope Geochemistry, Trace Element Geochemistry, Geochemical Pathways, Geochemical CyclingThese terms represent more advanced or specific concepts often studied in detailed or advanced geochemical modeling studies. They can be useful for identifying research that delves into these specific areas of interest.
Application AreasExtraterrestrial Hydrology, Planetary Volcanism, Planetary Ices, Contaminant Transport Modeling, Acid Rock Drainage, Environmental Geochemistry, HydrogeochemistryWhile not naming specific celestial bodies, these terms highlight various contexts where geochemical modeling could be applied. They can help identify studies that focus on specific applications relevant to celestial bodies.
Table 4. Titan’s lake compositions and approximate quantity. Data from [36].
Table 4. Titan’s lake compositions and approximate quantity. Data from [36].
Ethane (C2H6)Propane (C3H8)Methane (CH4)Hydrogen Cyanide (HCN)Butene (C4H8)Butane (C4H10)Acetylene (C2H2)
~76–79%~7–8%~5–10%~2–3%~1%~1%~1%
Table 5. Titan’s atmosphere and hypothesized ocean compositions. Data from [84].
Table 5. Titan’s atmosphere and hypothesized ocean compositions. Data from [84].
Surface TemperatureAtmosphere CompositionOcean CompositionOcean/Atmosphere
CO
92.5 KArgon: 0%, Methane: 1.55%Ethane (C2H6) and heavier alkanes: >90%,
Methane (CH4): 7.3%, Nitrogen (N2): 1.8%
0.25
101 KArgon: 17%, Methane: 21.1%Ethane (C2H6) and higher alkanes: 5%, Methane (CH4): 83.4%; Nitrogen (N2): 6%, Argon (Ar): 5.6%11
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Yoosefdoost, A.; Santos, R.M. Geochemical Modeling from the Asteroid Belt to the Kuiper Belt: Systematic Review. Encyclopedia 2026, 6, 38. https://doi.org/10.3390/encyclopedia6020038

AMA Style

Yoosefdoost A, Santos RM. Geochemical Modeling from the Asteroid Belt to the Kuiper Belt: Systematic Review. Encyclopedia. 2026; 6(2):38. https://doi.org/10.3390/encyclopedia6020038

Chicago/Turabian Style

Yoosefdoost, Arash, and Rafael M. Santos. 2026. "Geochemical Modeling from the Asteroid Belt to the Kuiper Belt: Systematic Review" Encyclopedia 6, no. 2: 38. https://doi.org/10.3390/encyclopedia6020038

APA Style

Yoosefdoost, A., & Santos, R. M. (2026). Geochemical Modeling from the Asteroid Belt to the Kuiper Belt: Systematic Review. Encyclopedia, 6(2), 38. https://doi.org/10.3390/encyclopedia6020038

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop