Physically, it is generally accepted that supercritical matter is homogeneous, with no appreciable difference between liquid and gas-like States. However, recent studies have shown that the introduction of the Frenkel line as a dynamic crossover line in the supercritical state can separate the supercritical state into two States with different properties. Separating the combined oscillatory and diffusive motion below the line from the pure diffusive motion above the line and extending to arbitrary high pressures and temperatures on the phase diagram expands people's understanding of the phase diagram of supercritical fluids^[7].

Van der Waals equation is a classical equation describing gas-liquid phase transition and critical phenomena:

The segment of {\left(\frac{\partial P}{\partial V}\right)}_T > 0 on the van der Waals curve does not correspond to the actual physical state, but there is a segment of liquid phase line below the coexistence point of the two phases and a segment of gas phase line above the coexistence point, which correspond to the state of superheating and supercooling, respectively. Two metastable regions of superheated liquid and supercooled gas exist between the metastable boundary line and the coexistence line. Supercooled gases are more common in nature. Water vapor aloft tends to be in a supercooled state and does not condense into clouds. As the jet flies through the clear sky, the gas and particles it ejects provide the conditions for condensation, thus pulling out the aircraft cloud band. We can use a similar phenomenon to create a cloud chamber to detect the track of charged particles^[8].

Relative to the supercooled gas, the crystalline phase is above its equilibrium melting temperature and reaches a superheated state, which is a non-equilibrium state that is very difficult to reach. In the supercooled state, the slowing down of the kinetics helps to prolong the lifetime of the metastable liquid, reducing the possibility of crystallization of the liquid upon cooling, however, in the heated state, the speeding up of the crystal vibration kinetics has the opposite effect, making it difficult for the crystal to reach the superheated state above the melting temperature. Willart et al. Realized the superheating of anhydrous α-phase trehalose by isothermal solid-state vitrification, and explored the physical mechanism of the transition from crystalline phase to metastable phase when the liquid is above the melting point (Fig. 4)^[9].

Clouds without rain are caused by the existence of supercritical water. When the temperature of clouds is close to zero, there may be ice crystals in them, and water vapor will not condense into liquid to form rainfall, so there is artificial rainfall.According to the physical characteristics of different clouds, the method is to select a specific time to spread dry ice, silver iodide, salt powder and other catalysts into the clouds by aircraft and rockets to make the clouds precipitate^[10]. China successfully conducted the earliest artificial rainfall experiment in 1958 (Fig. 5).

Compared with liquid water, the physical properties of supercritical water, such as density, dielectric constant, viscosity, diffusion coefficient and solubility, will change greatly^[12]. For example, the density of supercritical water is greatly affected by temperature and pressure. At 450 ℃ and 27 MPa, the density of supercritical water is 0.128 g/cm³. When the temperature and pressure increase to 500 ℃ and 100 MPa, the density increases to 0.540 g/cm³. The change of density has a great influence on the solubility, and the higher the density, the stronger the solubility. In addition, the dielectric constant of supercritical water is greatly reduced, which affects the solvent behavior and the degree of salt dissociation. The dissolved salt ions exist in the form of ion pairs, the solubility of inorganic substances is rapidly reduced, the inorganic substances dissolved in water will be rapidly separated out, and the water will change from a good solvent for inorganic salts to a good solvent for non-polar organic compounds, and can be mixed with oxygen, carbon dioxide and other gases in any proportion.

CO₂ is a good supercritical solvent because of its mild critical condition (T_c=31.1℃,P_c=7.38 MPa), low cost, non-toxicity and non-flammability, which can replace many conventional organic solvents used in traditional material processing. Supercritical CO₂ has the advantages of high diffusivity, zero surface tension, low viscosity and small molecular size, so it can easily insert and exfoliate layered materials, and the yield of graphene nanosheets with less than three layers can reach 93%^[13]. At the same time, its non-polar properties and weak van der Waals forces allow the formation of hydrophilic or lipophilic dispersions in the CO₂ continuous phase, and the phase behavior of the inverse micellar emulsion microenvironment can be manipulated by adjusting the physical properties of the supercritical CO₂ system^[14]. As an ideal external stimulus, supercritical CO₂ can also effectively induce the phase transition of layered materials because of its unique properties^[15].

Supercritical alcohol is mainly used in supercritical solvothermal synthesis, which is an advanced nano-preparation technology^[16]. Solid materials in supercritical solvents have the characteristics of small dielectric constant and homogeneous reaction, which can realize the rapid crystallization of nanocrystals. In addition, alcohol system has many advantages in the preparation of nanomaterials. Taking ethanol as an example, (1) the supersaturation of the product is high, which is conducive to the instantaneous nucleation of nanocrystals; (2) Ethanol has high compatibility with organic ligands, which is beneficial to the diffusion of organic ligands to the surface of growing crystals; (3) Supercritical ethanol has high viscosity, which can inhibit the growth and agglomeration of nanocrystals and obtain nanoscale products^[17,18].

The hydrothermal synthesis method has significant advantages in the preparation of mixed-valence composite oxides and the study of their valence States. On the one hand, the relatively low temperature will not affect the oxygen content in the crystal, thus eliminating the generation of oxygen defects to the greatest extent. On the other hand, thermodynamic metastable States, which are unstable at high temperatures, can be prepared.

For example, in the hydrothermal synthesis of manganite perovskite, the oxidation state of Mn is easy to show + 3 and + 4 mixed valence, and the mixed valence manganite is a special compound in the strongly correlated electron system.It not only has the strong correlation effect between d electrons, but also has the charge transfer effect between d electrons and p electrons of surrounding oxygen, thus showing rich electrical and magnetic properties. Therefore, the precise regulation of Mn oxidation state becomes a key issue. In the early stage, our research group used the characteristics of subcritical/supercritical hydrothermal disproportionation reaction and the formation of equilibrium defect crystals and stable special valence States in hydrothermal system.In the LaMnO₃ perovskite oxide system, a series of precise valence state regulation is realized: (1) the molar ratio of the reactant manganese source, namely potassium permanganate and manganese sulfate, is strictly controlled to be 3:7, and the final product with the average valence state of + 3.5 is prepared; (2) The co-doping of univalent potassium ion and bivalent calcium ion at A-site was realized by the disproportionation reaction of a single valence manganese source, so that the manganese ion at B-site showed a triple mixed valence of Mn³⁺, Mn⁴⁺ and Mn⁵⁺, and the single crystal was proved to possess an atomic scale pn junction of [Mn³⁺-O-Mn⁴⁺-O-Mn⁵⁺] type.

The equation for the hydrothermal melting-disproportionation reaction in this process is

The reaction occurring in this sub-/supercritical hydrothermal environment is quite complex, it will not be a simple disproportionation reaction, it may be a collection of several reactions, but the disproportionation reaction is undoubtedly the core of this chemical reaction. The electrical test results of the synthesized L {{{{\mathrm{a}}_{1-}}_x}_{-}}_y Ca_xK_yMnO₃(LCKMO) single crystal sample are the same as the I-V rectification characteristics of the rectifying molecule obtained by Aviram and Ratner's theoretical calculation in 1974. The coexistence of Mn³⁺ and Mn⁵⁺ in the sample provides a structural model for the exciton theory of high temperature superconductivity proposed by Bardeen in 1973. At room temperature, when a certain electric field is applied, the current density of the LCKMO single crystal sample can reach a 10⁴A/cm², and electrons generate net attraction through interaction with excitons to form electron pairs, thereby realizing the transformation from fermions to bosons, and finally realizing Bose-Einstein condensation under the promotion of a certain electric field to achieve room temperature superconductivity.

In ABO₃ type perovskite oxides, due to the partial substitution of A-site ions by low-valence ions, the valence of B-site ions increases to meet the charge balance, resulting in the coexistence of two valence States. For example, in the Ca_xMnO₃ of L {{\mathrm{a}}_{1-}}_x, Mn presents + 3 and + 4 valence, and the electron exchange between different valence ions brings rich electromagnetic effects. The electronic structure of LCKMO single crystal will be more complex due to the simultaneous doping of three different valence elements in its A-site. Taking single crystal La_0.66Ca_0.29K_0.05MnO₃ as an example, we use synchrotron radiation to find that the Mn K-edge X-ray near-edge absorption spectrum (XANES) is significantly different from that of the known two-fold mixed-valence perovskite manganites, and the absorption peak at 6564.3 eV is consistent with the reported Mn⁵⁺.NIR absorption spectrum at 625 nm corresponding to ³T₁ ( {\mathrm{t}}_2^2) →¹A₁ ( {\mathrm{t}}_2^2) excitation,Several peak positions such as 1143 nm, 1169 nm and 1224 nm in the laser-induced emission spectrum all prove the exact existence of Mn⁵⁺, which is strictly corresponding to the doping of K⁺ ions (Fig. 6). The triplet valence state of Mn was also confirmed by X-ray photoelectron spectroscopy in the films prepared by pulsed laser deposition.

In the powder diffraction (XRD) analysis of LCKMO single crystal sample, it is found that the three angles with 2θ of 7.48 °, 11.5 ° and 15.1 ° correspond to the periodic superstructure with unit cell parameters of a = 3.8864 8864 Å by 3, 2 and 1.5 times, respectively, that is, the modulated structure. Fig. 7 is a high-resolution transmission electron microscope (HRTEM) image of the sample along the [001] direction. It can be clearly seen that the periodic structure spacing is three times the side length of the unit cell. In the corresponding selected area electron diffraction (SAED) (fig. 7), the distance between the strong bright spots corresponds to the (001) crystal plane spacing, while the two relatively weak diffraction spots correspond to three times the modulation structure. At the same time, a twofold modulation structure is also observed in the [010] direction.

Quantum mechanics clarifies the universal law of microscopic particle motion, and the transport behavior of complex atoms and solids under quantum mechanics is an excellent model for condensed matter chemistry. When the interaction between electrons is not negligible, the wave functions of electrons overlap, and the transport of charge and energy in the crystal field is a very extensive non-equilibrium statistical problem, which has a wide range of practical significance for condensed matter chemical reactions under subcritical/supercritical conditions, and can understand the properties and laws of the interaction between carriers and crystals.

When a homogeneous semiconductor or a heterogeneous semiconductor forms a contact between a donor and an acceptor, a so-called pn-junction is formed. Conventional pn junctions are generally obtained by different types of doping in a continuously growing crystal. There are work function and contact potential difference between them. The work function W is usually defined as the difference between the surface vacuum rest electron energy E_0 and the Fermi energy E_{\mathrm{F}} when the internal energy band of the material concerned is flat:

The work function of a semiconductor can be expressed as:

Where \chi = E_0 - E_{\mathrm{C}} is called electron affinity. The value of \chi is relatively fixed for a given material, but the Fermi level E_{\mathrm{F}}, which is W to the value of its work function, varies with the type and amount of doping. When two crystals with different work functions are contacted, the two crystals will flow to form a potential difference until the Fermi level reaches the same level, and the contact potential difference compensates the difference between the two Fermi levels.

This results in the IV characteristic when the p-type and n-type materials are in contact, that is, the rectification characteristic under forward and reverse voltages, which can be seen to be related to the potential barrier near the interface.

The potential difference corresponding to the equilibrium barrier, which is the contact potential difference above, is often referred to as the self-built potential or built-in potential, and is denoted by V_{\mathrm{D}}, so that e V_{\mathrm{D}} is the height of the equilibrium barrier:

E_i is called the intrinsic Fermi level, E_{\mathrm{F}\mathrm{n}} 、 E_{\mathrm{P}\mathrm{n}} is the actual Fermi level position on both sides of n and p, and n_i is the carrier concentration.

In an arbitrary equilibrium junction, the electron concentration n_{\mathrm{p}}^0 in equilibrium is expressed by the n_{\mathrm{n}}^0 of the n-region as:

That is, between two points in equilibrium, the Boltzmann relation is preserved.

This establishes the unidirectional conductivity of the pn junction, and the barrier height is reduced from e V_{\mathrm{D}} to e ( V_{\mathrm{D}} - V ) by considering the application of a forward voltage V. Applying a reverse voltage V, the barrier height will become e ( V_{\mathrm{D}} + V ).

Considering the electric field and electric field distribution in the space charge region at this time, assuming the catastrophe model, using the condition that the total amount of charge on both sides of the pn junction is equal, we have:

Where N^{\mathrm{*}} is the introduced reduced concentration and E_{\mathrm{M}} is the electric field maximum.

In V_{\mathrm{D}} - V, it can be expressed as:

Thus, d is obtained as

The E_{\mathrm{M}} can thus also be expressed as:

These results can be applied to p-n junctions with different dielectric constants. In the general case of an applied voltage V, the distribution of the potential difference between the two sides is related to the dielectric constants of the two sides. The above equation (X) can be rewritten as:

The reduced dielectric constant {\mathrm{\epsilon }}^{\mathrm{*}} is introduced as

Under this potential distribution, the electrons and holes in the barrier region of the traditional semiconductor will diffuse more than drift, and the electrons and holes will be injected in opposite directions from their respective sub-regions, and these injection processes show that they diffuse with the depth of the gradient: the injected carriers diffuse and recombine at the same time, forming an exponential decay distribution. The forward and reverse current characteristic curves of typical Si-based pn junctions can be obtained by considering the thermal motion of charges without any assumption on the recombination mechanism.

Now we combine the triple valence atomic-level pn junction model to examine the extreme case: the barrier is thin enough that the carriers flowing through the barrier region undergo transport without scattering; At the same time, the barrier is strong enough to neglect the scattering and attenuation caused by normal thermal equilibrium, and only electrons with high energy at the edge of the barrier in some semiconductors can use quantum properties to cross the barrier.

Obviously, the atomic-level pn junction model has the above characteristics and can be accepted as a sufficient condition. The thickness d of the barrier layer in the triple valence lattice is far less than the mean free path l, which belongs to the atomic-level thin barrier. At the same time, the hybrid energy band of the e_g-t_2g orbit in the crystal establishes a large microscopic energy difference barrier which is strong enough to meet the condition of quantum tunneling of carriers.

Assuming that the transport is along the x-direction, under the condition of applied voltage V, the pn junction barrier height is e ( V_{\mathrm{D}} - V ). Only those electrons that move towards the x-direction and the associated kinetic energy exceeds the barrier height, i.e., satisfy the following condition:

To get over the barrier. From the above equation, the lower limit of the k_x of electrons that can cross the barrier can be obtained:

The current j : of pn junction can be obtained by calculating the current of electrons satisfying the condition.

Where \frac{ћk_x}{m} gives v_x, and if the energy is based on the bottom of the conduction band, we can use the Boltzmann distribution to give the distribution function f:

The integral can be obtained:

The A^{\mathrm{*}} is constant:

When the V = 0 is used, the current in the left and right directions is equal, and the current is further obtained as:

Define the average thermal motion velocity \stackrel{-}{v} , \stackrel{-}{v} = \mathrm{(8} k_{\mathrm{B}} {T/\pi m)}^{1/2}.

Here, the diffusion velocity is replaced by the thermal motion velocity. Under similar conditions, the tunneling current density of the atomic pn junction is much larger than that of the traditional pn junction.

At the same time, according to quantum mechanics, the probability of an electron moving towards the barrier with energy E penetrating the barrier is:

It can be seen that the tunneling probability depends on the thickness and height of the barrier.

If the pn barrier height is U_0 and the thickness is d. Described by a simple one-dimensional model:

The characteristics of triple valence atomic level pn junction come from the breakdown mechanism in the non-degenerate pn junction with high doping concentration, which is caused by the tunnel effect. Moreover, for this abrupt junction, the built-in electric field difference reaches a maximum.

N^{\mathrm{*}} is the reduced concentration. This small number of carriers, which can only tunnel under strong electric field conditions, can cause significant ionization collisions, such as avalanche breakdown, which occurs when the carriers are higher than the {\mathrm{\epsilon }}_{\mathrm{g}}, and impact ionization occurs by interband Auger recombination (excitation of electrons from the valence band to the conduction band). This induced carrier performance leads to a sharp increase in the junction current. This constructs the triple valence atomic-level pn junction ideal IV characteristic.

Moreover, this triple valence crystal is an ideal periodic structure of atomic-level pn junction, which is equivalent to a superlattice structure with numerous independent quantum wells perfectly stacked. Carriers in conventional semiconductors can move freely in three-dimensional space, but first of all, the thickness of each layer of the independent quantum well structure of triple valence crystal materials is atomic level.The width of the quantum well is within the mean free path of the carrier, which is comparable to the electron de Broglie wavelength, and the quantum confinement effect changes the carrier wave function so that its conduction and valence bands are split into subbands. More attention should be paid to the triple valence crystal, that is, the superlattice composed of repeated quantum wells. The thin barrier between adjacent quantum wells makes the wave functions in adjacent quantum wells couple together, and the discrete energy States in each quantum well are expanded to form several delocalized zonules. The electron moves along the alternating direction of the material, which causes a unique oscillation mode, the resonant state. At this time, if the electron energy is below the Fermi level, different electrons can combine together, and this new type of Cooper pair provides a solid foundation for the realization of Bose-Einstein condensation.

Here, we comprehensively review the superior tunneling performance of triple valence atomic-scale pn-junctions. The generation of such transport properties stems first from creating reasonable donor-acceptor models and tunneling barriers with appropriate heights. In conventional 3D semiconductor materials such as Si, SiGe, Ⅲ-Ⅴ, etc., controlling the sharp doping profile at the sub-nanometer scale is not tenable. The blurred interface will produce the edge tail phenomenon of the energy band, which will seriously reduce the quantum tunneling performance. At the same time, the construction of heterojunction depends on molecular beam epitaxy or chemical vapor deposition, which is limited by the choice of materials and the requirement of solving the lattice matching problem, and the artificial tailoring of band structure engineering is limited by various complex factors. The unique structure of the triple valence atomic-level pn junction makes it atomically thick in terms of scale, and there are no factors such as quantum mechanical reflection of the interface and scattering of optical phonons near the interface. Because of these unique characteristics, the triple valence atomic-level pn junction is "mutated" on a unit cell scale, and the donor and acceptor are constructed to prevent the formation of low concentration in the tetravalent intermediate valence state. The thickness of the buffer layer is thin enough, the doping on both sides of the barrier is equivalent to the heavy doping of a traditional semiconductor, and the width of the barrier is larger than that of an Esaki diode. This realizes the novel quantum IV property. Benchmarked against conventional Si technology, triple valence atomic-scale pn junctions have the ability to shrink device size and can face the severe challenge of Moore's law. At the same time, due to the natural periodic structure of the crystal to construct an "ideal" superlattice, it is possible to use the periodic structure of the crystal to induce Bose Einstein condensation by periodically regulating microscopic particles, which will become an important research direction of condensed matter chemistry in the future.

In the industrial production process, supercritical fluid extraction technology is used to separate and refine aromatic homologues, and to extract paraffin and coal tar from coal. Commonly used supercritical fluid extractants include water, CO₂, alcohols and light hydrocarbons. For example, toluene is used as extractant to deeply crack coal under supercritical conditions (P_c=10 MPa,T_c=673.15~713.15 K) to convert it into liquid fuel, and the conversion efficiency is as high as one third. Supercritical hydrothermal conditions can also achieve efficient conversion of biomass. Supercritical fluids are well suited to enhance the chemical conversion of biological feedstocks to useful liquid and gaseous fuels^[27]. Reactions and separations in certain supercritical fluids have advantages over traditional biochemical processing methods.

In the food industry, supercritical fluid extraction technology has been established to extract useful substances or remove harmful substances from natural products. For example, in 1970, Zosel first used supercritical fluid technology to extract caffeine from coffee beans. By 1978, Hag performed the first commercial supercritical fluid extraction in Germany for the extraction of caffeine. Two years later, Carlton and United Breweries in Australia developed a process for extracting hops using liquid carbon dioxide^[28]. Both applications have been commercially successful and have generated many changes and improvements that have also been developed on an industrial scale (Fig. 12)^[29].

Currently, supercritical fluids have been widely used to design new drug particles. It can improve the solubility and dissolution rate of drugs with poor water solubility, such as crystalline drugs^[31]. Stanipharm's patented StaniTab^® process is a bottom-up process based on supercritical CO₂ for the manufacture of poorly soluble drug nanoparticles and auxiliary substances commonly found in tablets. In addition, drug micronization and drug polymer particles are mainly carried out by supercritical solution rapid expansion or supercritical antisolvent processes^[32]. The use of supercritical CO₂ as an antisolvent for the production of antibiotic nanoparticles and polymeric nanoparticles for cancer therapy are a few examples^[33][34]. Critical Pharmaceuticals has developed CriticalMix^TM as a one-step solvent-free process that uses supercritical carbon dioxide to facilitate the mixing of therapeutic and biodegradable polymers to produce drug-loaded controlled release formulations. Based on this technology, a long-acting formulation of human growth hormone, CP016, has been developed and has established preclinical proof-of-concept in pharmacokinetic and pharmacodynamic studies in nonhuman primates.

In recent years, supercritical fluid technology has not only been limited to the extraction process, but also the gaseous molecular reaction under supercritical conditions has attracted much attention. Gaseous reactions under supercritical conditions can break the limitations of thermodynamics and kinetics. At the same time, fluids under supercritical conditions not only have good dissolution and mass transfer characteristics, but also can be used as non-toxic solvents and reactants. Therefore, the relevant researchers applied the CO₂ under supercritical conditions to the synthesis of dimethyl carbonate, a green chemical product, to realize the green synthesis of chemical products. For example, Wang et al proposed that under double supercritical conditions (the synthesis temperature is higher than the supercritical temperature of each reactant, and the partial pressure of each reactant component is greater than its supercritical partial pressure), Y_xFe_1-xO_δ is used as the catalyst, and CO₂ and methanol are used as the reaction medium to synthesize dimethyl carbonate, so as to improve the diffusion rate and contact opportunity of the reaction molecules, thereby improving its yield^[35]. Compared with the traditional phosgene method for synthesizing the dimethyl carbonate, the synthesis process under the supercritical CO₂ condition not only uses CO₂ as a carbon source to reduce the greenhouse effect, but also avoids the emission of a large number of toxic and harmful gases, so that the whole process flow is in a green and harmless environment.

In addition, due to the advantages of nuclear power in economic benefits and environmental protection, it has become the most important clean energy in China. Since Qinshan Nuclear Power Station, the first nuclear power station in mainland China, was put into commercial operation in 1994, and Taishan EPR and Sanmen AP1000 nuclear power units were connected to the grid in 2018, China's nuclear power unit equipment has developed to the third generation. As the only water-cooled reactor in the fourth generation reactor selected by the international nuclear energy system, the thermal efficiency of supercritical water-cooled reactor is about 1/3 higher than that of traditional light water reactor, which can greatly improve the utilization rate of nuclear fuel^[36,37].

The destruction of highly concentrated, toxic, persistent, recalcitrant organic compounds present in industrial and municipal wastewater remains a major challenge to overcome. Supercritical water oxidation (SCWO) has been developed into a new type of organic hazardous waste disposal technology. Under supercritical conditions, organic compounds and oxygen are completely miscible in water, and complete oxidation can occur in a very short reaction time to convert biomass, organic waste, and coal into H₂O, N₂, CO₂, and other small molecules, thereby reducing the production of air pollutants. In addition, supercritical CO₂ with excellent mass transfer capacity can effectively remove fouling and achieve environmental cleanliness. Black liquor with high alkalinity and pungent odor is produced in the process of papermaking and pulping, and the alkali in the black liquor is recycled by the synergistic gasification of the black liquor and wheat straw in supercritical water^[39]. In addition, there is co-oxidation between sludge in supercritical water and active alcohols (methanol, ethanol and isopropanol), which achieves excellent decontamination effect. One of the reasons is that alcohols provide active HO₂· and OH ·. Another reason is that the addition of alcohols not only prevents the formation of recalcitrant products (non-nitrogen and nitrogen aromatic compounds), but also promotes the production of reactive products (non-nitrogen open chain compounds)^[40]. Supercritical water oxidation process was used to treat hospital wastewater containing drugs and other toxic chemicals as well as various pathogenic microorganisms, and the removal rates of nine drugs were between 72% and 99.9%^[41].

The origin of life is one of humanity's greatest mysteries. Life's requirements for the environment are not as strict as we think, perhaps in some special environments, such as supercritical fluids, life can still exist; Or in the special solvent of supercritical fluid, the chemical evolution stage before the birth of life is easier, thus providing a good medium for the origin of life. Current astrophysical observations of the universe also show that supercritical fluids are widespread in the universe, which may have implications for our exploration of extraterrestrial life.

So far, human beings have found that matter in many parts of the universe exists in a supercritical state. For example, the atmosphere of Venus is composed of 96.5% CO₂ and 3.5% N₂, but its surface temperature is as high as 462 ℃, which is much higher than the critical temperature of CO₂, and the surface pressure is 93 atm, which means that the CO₂ on the surface of Venus is in a supercritical state. Using in situ Raman spectroscopy, Sun et al. Found supercritical CO₂ in the form of bubbles in hydrothermal vents in the southern Okinawa Trough. In particular, the peak value of N₂ in supercritical CO₂ is much higher than that in seawater and sluicing fluid, indicating that supercritical CO₂ can enrich N₂ from the surrounding environment. Researchers believe that it is a favorable substance that may have triggered the origin of life on Earth. This suggests that in the early history of the Earth, supercritical CO₂ promoted the synthesis, preconcentration and preservation of amino acids and other organic substances essential for the origin of life^[42].

On Earth, we have found supercritical fluids under natural conditions. Macromolecular hydrocarbons were also synthesized from some inorganic small molecules by simulating the reaction characteristics of amino acid condensation in the submarine hydrothermal system^[43][44]. On the one hand, these studies explain the important link of the origin of life, on the other hand, they also illustrate the important role of supercritical water environment in the synthesis of life substances. The study of hydrothermal biochemistry under supercritical conditions is of great significance in the study of the origin and evolution of life. Our research group has realized the transformation of C₁ in the hydrothermal system, that is, the evolution from inorganic small molecules such as CO and CO₂ and their generated formaldehyde molecules to organic small molecules and biomolecules, which is the basis of the hydrothermal life origin model (Fig. 14)^[45]. In recent years, with the exploration of the underwater world, the study of the extreme environment ecosphere in the deep lithosphere and the early environment of the earth, people expect to find more signs of early life in extreme environments such as supercritical fluids, and to describe the origin and evolution of life on the earth more clearly.

In gas giant planets such as Jupiter, Saturn, exoplanets, and brown dwarfs, molecular hydrogen exists in a supercritical state (critical temperature and critical pressure of ∼ 33 K and 1.3 MPa) (Fig. 15)^[46,47]. The fundamental question of how to define the boundary between the interior and the exterior (atmosphere) of a planet is related to the supercritical nature of molecular hydrogen. Unlike the sharply bounded terrestrial planets such as Earth and Venus, the boundaries of the gas giant planets are conditional because the supercritical state can be considered physically homogeneous and smooth. Using the Earth's atmospheric pressure as a benchmark to describe the boundary of gas giant planets, the calculated radius of Jupiter is about 70 000 km, and the radius of Saturn is about 57 000 km, which is roughly consistent with the measured optical size^[48]. But recently it was found that the supercritical state is not physically homogeneous, but exists in two States with different physical properties^[49,50]. In the phase diagram, the dynamic transition is separated by the Frenkel line, which means that supercritical molecular hydrogen exists in two different physical States. This is the case for Jupiter and Saturn. The physical boundary of supercritical hydrogen exists in a gas giant at the Frenkel line, with a qualitative change in the character of the particle motion, accompanied by a qualitative change in the particle dynamics and a large change in all the relevant main physical properties from gas-like to liquid-like. This is exactly the same change that occurs at the liquid-gas planetary atmosphere boundary for smaller planets such as Earth or Venus, except that there is no first-order phase transition.

Supercritical condensation has also been found in other celestial bodies, and most of the original clay on Mars was formed when the primary crust of Mars reacted with the supercritical atmosphere of concentrated steam or water and carbon dioxide degassed during the cooling of the magma ocean^[51⇓~53]. Nitrogen is also found on Venus in a supercritical state, which is further fixed by volcanism, lightning, or volcanic lightning, and then accumulates on the surface of Venus as nitrate or nitrite^[46]. In addition to Venus, supercritical molecular hydrogen exists on the giant planets Jupiter and Saturn, as well as brown dwarfs and some exoplanets^[54].

Since the liquefaction of helium was first realized by Onnes, a Dutch physicist, in 1908, important macroscopic quantum phenomena in condensed matter physics, such as superconductivity and superfluidity, have been observed successively. Electrons in superconductors are weakly coupled to form Cooper pairs mediated by phonons, and the directed motion shows "zero resistance" ”;⁴He. Bose-Einstein condensation occurs below 2.2 K and enters the superfluid phase, and the "zero resistance" motion can be completely without viscosity. Its isotopic ³He realizes fermion condensation at 2.6 mK, which is driven by the spin fluctuation between the paired atoms. The novel state of matter at low temperature contains rich scientific connotation and attractive application prospects. With the continuous discovery of new superconducting systems, Bardeen et al. Have put forward a sufficient microscopic explanation for superconductivity with the "BCS theory". However, the exploration of superconductors with higher transition temperature is still the unremitting pursuit of the field of physics and chemistry^[55].

Subsequently, Bardeen et al. Proposed a model of superconductivity based on the exciton mechanism in 1973. A semiconductor is covered by an extremely thin metal layer, and electrons near the Fermi surface of the metal tunnel into the band gap of the semiconductor, forming a net exchange attraction between electrons and excitons. Because the frequency spectrum of excitons is higher than that of phonons, superconductors with high T_c may be realized in the exciton model. Pb/PbTe and amorphous Si/Si systems have been used to verify the exciton model, but they have not been confirmed experimentally because of the difficulty in achieving ideal energy level matching or chemical bonding^[56]. Therefore, based on the unique advantages of chemical synthesis in the control of atomic structure and electronic structure, the rational design and construction of microscopic functional building blocks will be an effective way to prepare functional solids.

For example, we first obtained the triplet valence state of manganese in LCKMO system by disproportionation under mild hydrothermal conditions, which presents ideal IV rectification characteristics and room temperature supercurrent phenomenon, the former is attributed to the transition metal ions Mn³⁺ ( {t_{2\mathrm{g}}}^3 {{\mathrm{e}}_{\mathrm{g}}}^1), Mn⁴⁺ ( {t_{2\mathrm{g}}}^3 {{\mathrm{e}}_{\mathrm{g}}}^0) in the octahedron,The d electrons of the Mn⁵⁺ ( {t_{2\mathrm{g}}}^2 {{\mathrm{e}}_{\mathrm{g}}}^0) decrease in a gradient manner, and the electron-rich Mn³⁺ and the electron-deficient Mn⁵⁺ are bridged on both sides of the Mn⁴⁺ to act as a donor and an acceptor of electrons, respectively, to form a unique atomic-scale pn junction, which is quickly turned on when the voltage overcomes the atomic junction barrier in series; Under the action of an electric field, the electrons and holes near the Mn³⁺ and the Mn⁵⁺ form an exciton like localization, and the electron-hole pairs are mutually correlated and constrained in a certain space, resulting in a net attractive effect. When two electrons with opposite spin directions exchange momentum with the exciton respectively (Fig. 16), the exciton-mediated electron pair can migrate directionally in the lattice network without hindrance, and the maximum current density observed in the experiment is about 5×10⁴A/cm². Triplet valence manganites are very similar to Bardeen's exciton superconducting model in both material structure and experimental phenomena. A small amount of K element doping leads to the distribution of the corresponding electron-rich Mn³⁺ in the semiconductor manganite matrix. Under the induction of an electric field, a physical state transition from a semiconductor state to a superconducting-like state is realized at room temperature^[57].

Alternatively, the role of dissipation in physics can be used to generate new States of matter, such as: manipulating qubits, engineering decoherence-free subspaces, generating entangled quantum States, and refining quantum features^{[58,59][60][61][62][63]}. It has been shown that supercritical non-equilibrium Bose-Einstein condensation can be achieved by combining dissipation with evaporative cooling in chemical reactions. This process produces a long-lived nonequilibrium state, creates and grows a supercritical Bose condensate, exhibits quasi-static behavior typical of a prethermal state, and can be used to perform experiments on highly condensed parts at high temperatures, expanding the hierarchy of condensed matter chemistry^[64].

Normal cells are filled with many types of biocoacervates, such as cytoplasmic stress granules and P granules. Proteins that form biological condensed matter (amino acid polymerization) usually have the function of multiple copies of protein interaction domains, which in turn recruit other proteins to form biological condensed structures. These membraneless organelles can undergo liquid-liquid phase separation through chemical interactions between molecules to form separate condensed phases. The condensed phase has a high concentration of specific molecules, which can isolate the molecules from other biochemical reactions and facilitate the specific biochemical reactions. These condensed substances can be reversibly disassembled and assembled to maintain chemical equilibrium in the body.

Biomolecular condensates, generated through phase transitions, can form interfaces with the cytoplasm, other condensates, and surrounding cellular structures. The capillary force or interfacial force caused by the interface formed by the multi-level structure plays an important organizational role in non-living soft matter systems, and the analysis of this kind of interface from the perspective of condensed matter chemistry is an excellent entrance to understand the structure, organization, and function of living cells^[65].

Chen et al. Used supercritical fluid technology to design a stable, biocompatible and dual-triggered indocyanine green encapsulated silk fibroin nanomaterial (Fig. 17). The material has excellent photothermal stability in the acidic environment of tumor, and can destroy tumor cells only under photoinduced hyperthermia. Supercritical fluid technology has great potential in the preparation of materials for sustained delivery of cancer therapy.

The biophysical properties of many tumor cell biocondensates have changed, and the protein domains can not be closely arranged into a specific structure, which leads to the disorder of the condensed structure and the change of its function^[67]. These properties are relevant to gene regulation and cell signaling. Tumors are characterized by abnormal cell proliferation, which is driven by abnormal activity of tumor-related genes caused by genomic instability, gene expression disorders, or protein degradation. This suggests that the modification of biocondensates makes tumor cells have various tumor characteristics and promote tumor development, so regulating the assembly and characteristics of biocondensates can contribute to tumor regression. From the perspective of the multi-level structure of condensed matter chemistry, the regulatory mechanism of biological condensed matter is deeply understood, and its phase behavior and biochemical characteristics are explored.As well as the methods of regulating biological condensed matter, will help us break through the traditional molecular biology research on tumors, deepen the understanding and treatment of tumors from a new perspective, and achieve tumor regression.

In the evolution of the universe, from the formation of light elements to the production of heavy elements, and then to the formation of some molecules, and finally to the development of biological macromolecules and life phenomena under certain conditions, the movement and structure of matter are becoming more and more complex. Since 1822, when Baron de la Tour heated a mixture of liquid and vapor to a point where there was no clear distinction between them (the critical point), Thomas Andrews called it the supercritical state and explained most of its properties, there has been no qualitative leap in the understanding of the supercritical state in 200 years. Condensed matter chemistry has jumped out of the scale of "elementary particles" in traditional chemistry, focusing on the condensed state of matter with specific composition, multi-level structure, properties and functions, which provides a direction and basis for the re-understanding of chemical reactions. The problem of condensed state transition driven by chemical reaction under subcritical/supercritical conditions provides a useful window for how matter moves towards higher order and organization. In addition, when the system enters subcritical/supercritical conditions, it may fall into a "chaotic" state, such as genome instability and gene expression disorder in organisms. In this paper, we focus on the chemical reactions under subcritical/supercritical conditions, summarize the formation of atomic-scale pn junctions and the essence of their quantum IV properties, extend them to the multi-level structure of condensed matter chemistry, and sort out the frontier scientific issues and potential applications contained therein, with a view to promoting the in-depth exploration of condensed matter chemistry under subcritical/supercritical conditions.