Dust is formed out of stellar material and is constantly affected by different mechanisms occurring in the ISM. Dust grains behave differently under these mechanisms depending on their sizes, and therefore the dust grain size distribution also evolves as part of the dust evolution itself. Following how the grain size distribution evolves is a difficult computing task that is just recently being overtaking. Smoothed particle hydrodynamic (SPH) simulations of a single galaxy as well as cosmological simulations are producing the first predictions of the evolution of the dust grain size distribution. We compare for the first time the evolution of the dust grain size distribution predicted by the SPH simulations with the results provided by the observations. We analyse how the radial distribution of the small to large grain mass ratio (D(S)/D(L)) changes over the whole discs in three galaxies: M 101, NGC 628 and M 33. We find good agreement between the observed radial distribution of D(S)/D(L) and what is obtained from the SPH simulations of a single galaxy. The central parts of NGC 628, at high metallicity and with a high molecular gas fraction, are mainly affected not only by accretion but also by coagulation of dust grains. The centre of M 33, having lower metallicity and lower molecular gas fraction, presents an increase of D(S)/D(L), showing that shattering is very effective in creating a large fraction of small grains. Observational results provided by our galaxies confirm the general relations predicted by the cosmological simulations based on the two grain size approximation. However, we present evidence that the simulations could be overestimating the amount of large grains in high massive galaxies.

Recent progress in submillimetre surveys by single-dish telescopes allows us to further challenge the consistency between cosmological simulations and observations. In particular, we compare our simulations that include dust formation and destruction with the recent SCUBA-2 surveys (`STUDIES') by putting emphases on basic observational properties of dust emission such as dust temperature, size of infrared (IR)-emitting region, IR luminosity function and IRX--$\beta$ relation. After confirming that our models reproduce the local galaxy properties, we examine the STUDIES sample at $z\approx 1-4$, finding that the simulation reproduces the aforementioned quantities except for the $z\gtrsim 2$ IR luminosity function at the massive end ($\sim 10^{13}$ L$_{\odot}$). This means that the current simulation correctly reproduces the overall scaling between the size and luminosity (or star formation rate) of dusty region, but lacks extreme starburst phenomena at $z\gtrsim 2$. We also discuss extinction curves and possible AGN contribution.

To study the dust evolution in the cosmological structure formation history, we perform a smoothed particle hydrodynamic simulation with a dust enrichment model in a cosmological volume. We adopt the dust evolution model that represents the grain size distribution by two sizes and takes into account stellar dust production and interstellar dust processing. We examine the dust mass function and the scaling properties of dust in terms of the characteristics of galaxies. The simulation broadly reproduces the observed dust mass functions at redshift $z = 0$, except that it overproduces the massive end at dust mass $M_\mathrm{d} \gtrsim 10^{8}$ ${\rm M}_\odot$. This overabundance is due to overproducing massive gas/metal-rich systems, but we also note that the relation between stellar mass and gas-phase metallicity is reproduced fairly well by our recipe. The relation between dust-to-gas ratio and metallicity shows a good agreement with the observed one at $z=0$, which indicates successful implementation of dust evolution in our cosmological simulation. Star formation consumes not only gas but also dust, causing a decreasing trend of the dust-to-stellar mass ratio at the high-mass end of galaxies. We also examine the redshift evolution up to $z \sim~ 5$, and find that the galaxies have on average the highest dust mass at $z = 1-2$. For the grain size distribution, we find that galaxies with metallicity $\sim 0.3~ Z_\odot$ tend to have the highest small-to-large grain abundance ratio; consequently, the extinction curves in those galaxies have the steepest ultraviolet slopes.

To investigate the evolution of dust in a cosmological volume, we perform hydrodynamic simulations, in which the enrichment of metals and dust is treated self-consistently with star formation and stellar feedback. We consider dust evolution driven by dust production in stellar ejecta, dust destruction by sputtering, grain growth by accretion and coagulation, and grain disruption by shattering, and treat small and large grains separately to trace the grain size distribution. After confirming that our model nicely reproduces the observed relation between dust-to-gas ratio and metallicity for nearby galaxies, we concentrate on the dust abundance over the cosmological volume in this paper. The comoving dust mass density has a peak at redshift $z\sim 1$--2, coincident with the observationally suggested dustiest epoch in the Universe. {In the local Universe}, roughly 10 per cent of the dust is contained in the intergalactic medium (IGM), where only 1/3--1/4 of the dust survives against dust destruction by sputtering. We also show that the dust mass function is roughly reproduced at $\lesssim 10^8$ M$_\odot$, while the massive end still has a discrepancy, which indicates {the necessity of stronger feedback in massive galaxies}. %%The relation showed that accretion is essential for dusty galaxies. In addition, our model broadly reproduces the observed radial profile of dust surface density in the circum-galactic medium (CGM). While our model satisfies the observational constraints for the dust extinction {on cosmological scales}, it predicts that the dust in the CGM and IGM is dominated by large ($> 0.03~\mu$m) grains, which is in tension with the steep reddening curves {observed} in the CGM.

There are two major theoretical issues for the star formation law (the relation between the surface densities of molecular gas and star formation rate on a galaxy scale): (i) At low metallicity, it is not obvious that star-forming regions are rich in H$_2$ because the H$_2$ formation rate depends on the dust abundance; and (ii) whether or not CO really traces H$_2$ is uncertain, especially at low metallicity. To clarify these issues, we use a hydrodynamic simulation of an isolated disc galaxy with a spatial resolution of a few tens parsecs. The evolution of dust abundance and grain size distribution is treated consistently with the metal enrichment and the physical state of the interstellar medium. We compute the H$_2$ and CO abundances using a subgrid post-processing model based on the dust abundance and the dissociating radiation field calculated in the simulation. We find that when the metallicity is $\lesssim 0.4$ Z$_\odot$ ($t<1$ Gyr), H$_2$ is not a good tracer of star formation rate because H$_2$-rich regions are limited to dense compact regions. At $Z\gtrsim 0.8$ Z$_\odot$, a tight star formation law is established for both H$_2$ and CO. At old ($t \sim 10$ Gyr) ages, we also find that adopting the so-called MRN grain size distribution with an appropriate dust-to-metal ratio over the entire disc gives reasonable estimates for the H$_2$ and CO abundances. For CO, improving the spatial resolution of the simulation is important while the H$_2$ abundance is not sensitive to sub-resolution structures at $Z\gtrsim 0.4$ Z$_\odot$.

To understand the evolution of extinction curve, we calculate the dust evolution in a galaxy using smoothed particle hydrodynamics simulations incorporating stellar dust production, dust destruction in supernova shocks, grain growth by accretion and coagulation, and grain disruption by shattering. The dust species are separated into carbonaceous dust and silicate. The evolution of grain size distribution is considered by dividing grain population into large and small gains, which allows us to estimate extinction curves. We examine the dependence of extinction curves on the position, gas density, and metallicity in the galaxy, and find that extinction curves are flat at $t \lesssim 0.3$ Gyr because stellar dust production dominates the total dust abundance. The 2175 \AA\ bump and far-ultraviolet (FUV) rise become prominent after dust growth by accretion. At $t \gtrsim 3$ Gyr, shattering works efficiently in the outer disc and low density regions, so extinction curves show a very strong 2175 \AA\ bump and steep FUV rise. The extinction curves at $t\gtrsim 3$ Gyr are consistent with the Milky Way extinction curve, which implies that we successfully included the necessary dust processes in the model. The outer disc component caused by stellar feedback has an extinction curves with a weaker 2175 \AA\ bump and flatter FUV slope. The strong contribution of carbonaceous dust tends to underproduce the FUV rise in the Small Magellanic Cloud extinction curve, which supports selective loss of small carbonaceous dust in the galaxy. The snapshot at young ages also explain the extinction curves in high-redshift quasars.

We perform smoothed particle hydrodynamics (SPH) simulations of an isolated galaxy with a new treatment for dust formation and destruction. To this aim, we treat dust and metal production self-consistently with star formation and supernova feedback. For dust, we consider a simplified model of grain size distribution by representing the entire range of grain sizes with large and small grains. We include dust production in stellar ejecta, dust destruction by supernova (SN) shocks, grain growth by accretion and coagulation, and grain disruption by shattering. We find that the assumption of fixed dust-to-metal mass ratio becomes no longer valid when the galaxy is older than 0.2 Gyr, at which point the grain growth by accretion starts to contribute to the nonlinear rise of dust-to-gas ratio. As expected in our previous one-zone model, shattering triggers grain growth by accretion since it increases the total surface area of grains. Coagulation becomes significant when the galaxy age is greater than $\sim$ 1 Gyr: at this epoch the abundance of small grains becomes high enough to raise the coagulation rate of small grains. We further compare the radial profiles of dust-to-gas ratio $(\mathcal{D})$ and dust-to-metal ratio $(\mathcal{D}/Z)$ (i.e., depletion) at various ages with observational data. We find that our simulations broadly reproduce the radial gradients of dust-to-gas ratio and depletion. In the early epoch ($\lesssim 0.3$ Gyr), the radial gradient of $\mathcal{D}$ follows the metallicity gradient with $\mathcal{D}/Z$ determined by the dust condensation efficiency in stellar ejecta, while the $\mathcal{D}$ gradient is steeper than the $Z$ gradient at the later epochs because of grain growth by accretion. The framework developed in this paper is applicable to any SPH-based galaxy evolution simulations including cosmological ones.

Recent ALMA observations of high-redshift normal galaxies have been providing a great opportunity to clarify the general origin of dust in the Universe, not biased to very bright special objects even at $z>6$. To clarify what constraint we can get for the dust enrichment in normal galaxies detected by ALMA, we use a theoretical model that includes major processes driving dust evolution in a galaxy; that is, dust condensation in stellar ejecta, dust growth by the accretion of gas-phase metals, and supernova destruction. Using the dust emission fluxes detected in two normal galaxies at $z>6$ by ALMA as a constraint, we can get the range of the time-scales (or efficiencies) of the above mentioned processes. We find that if we assume extremely high condensation efficiency in stellar ejecta ($f_{\mathrm{in}} \ga 0.5$), rapid dust enrichment by stellar sources in the early phase may be enough to explain the observed ALMA flux, unless dust destruction by supernovae in those galaxies is stronger than that in nearby galaxies. If we assume a condensation efficiency expected from theoretical calculations ($f_{\mathrm{in}} \la 0.1$), strong dust growth (even stronger than assumed for nearby galaxies if they are metal-poor galaxies) is required. These results indicate that the normal galaxies detected by ALMA at $z>6$ are biased to objects (i) with high dust condensation efficiency in stellar ejecta, (ii) with strong dust growth in very dense molecular clouds, or (iii) with efficient dust growth because of fast metal enrichment up to solar metallicity. A measurement of metallicity is crucial to distinguish among these possibilities.

Extinction curves, especially those in the Milky Way (MW), the Large Magellanic Cloud (LMC), and the Small Magellanic Cloud (SMC), have provided us with a clue to the dust properties in the nearby Universe. We examine whether or not these extinction curves can be explained by well known dust evolution processes. We treat the dust production in stellar ejecta, destruction in supernova shocks, dust growth by accretion and coagulation, and dust disruption by shattering. To make a survey of the large parameter space possible, we simplify the treatment of the grain size distribution evolution by adopting the `two-size approximation', in which we divide the grain population into small ($\lesssim 0.03~\mu$m) and large ($\gtrsim 0.03~\mu$m) grains. It is confirmed that the MW extinction curve can be reproduced in reasonable ranges for the time-scale of the above processes with a silicate-graphite mixture. This indicates that the MW extinction curve is a natural consequence of the dust evolution through the above processes. We also find that the same models fail to reproduce the SMC/LMC extinction curves. Nevertheless, this failure can be remedied by giving higher supernova destruction rates for small carbonaceous dust and considering amorphous carbon for carbonaceous dust; these modification fall in fact in line with previous studies. Therefore, we conclude that the current dust evolution scenario composed of the aforementioned processes is successful in explaining the extinction curves. All the extinction curves favor efficient interstellar processing of dust, especially, strong grain growth by accretion and coagulation.