Suppression of Star Formation in Galaxy Pairs

We investigate the suppression of star formation in galaxy pairs based on the isolated galaxy pair sample derived from the Sloan Digital Sky Survey survey. By comparing the star formation rate between late-type galaxies (LTGs) in galaxy pairs and those in the isolated environment, we detect a signal of the suppression of star formation in galaxy pairs at d p < 100 kpc and 200 kpc < d p < 350 kpc. The occurrence of the suppression of star formation in these LTGs requires their companion galaxies to have an early-type morphology (n s > 2.5). The suppression of star formation in wide galaxy pairs at 200 kpc < d p < 350 kpc mainly occurs in massive LTGs, while in close galaxy pairs at d p < 100 kpc, it only appears in LTGs with a massive companion ( logM⋆>11.0 ), nearly independent of their own stellar mass. Based on these findings, we infer that the suppression of star formation in wide galaxy pairs is actually a result of galaxy conformity, while in close galaxy pairs, it stems from the influence of hot circumgalactic medium surrounding companion galaxies.


INTRODUCTION
In the framework of hierarchical galaxy formation, galaxy merging plays a key role in the mass assembly and structure formation of galaxies.Before the final coalesce, two galaxies are gravitationally bound in the form of galaxy pairs for 1 − 2Gyrs (Kitzbichler & White 2008;Lotz et al. 2008).In galaxy pairs, the evolution of member galaxies is inevitably influenced by their companions, resulting in a variety of peculiar properties compared to the galaxies in isolated environments.Nevertheless, how galaxy interaction affects the physical properties of galaxy pairs, especially the star formation rate (SFR), is still not clear.
The enhancement of star formation is one of the most prominent features of galaxy pairs, which has been known since the 1970s (Larson & Tinsley 1978).Numer-ous studies have shown that the SFRs of galaxy pairs are significantly higher than those of isolated galaxies at given stellar masses (Barton et al. 2000;Ellison et al. 2008).Such an enhancement of star formation shows a strong correlation with the projected separation between pair members, where the enhancement is detectable up to d p ∼ 150 kpc and reaches the maximum at d p < 30 kpc (Li et al. 2008;Patton et al. 2013;Feng et al. 2019).Furthermore, enhanced star formation is usually correlated with other peculiarities, such as asymmetric photometric morphology (Patton et al. 2016;Pan et al. 2019), disturbed gas kinematics (Bloom et al. 2018;Feng et al. 2020), and diluted gas phase metallicity (Kewley et al. 2006;Scudder et al. 2012).In combination with the result of the numerical simulation, the enhanced star formation of galaxy pairs can be explained by the scenario of tidal-induced gas inflow, where the tidal perturbation from companion galaxies is the key to the nature of star formation (Di Matteo et al. 2008;Torrey et al. 2012).

Feng et al.
However, more detailed studies later pointed out that not all galaxy pairs exhibit enhanced star formation, which depends on the types of companion galaxies.They found that galaxies with late-type (star-forming) companions usually display increased SFRs, while it is not for galaxies with early-type (passive) companions for given stellar mass intervals (Park & Choi 2009;Xu et al. 2010;Yuan et al. 2012).This indicates that there may be other physical mechanisms regulating the star formation of paired galaxies.
In terms of the star formation properties of paired galaxies with early-type (passive) companions, previous studies have not reached an agreement.Some studies suggested that the star formation of these galaxies is comparable to that of isolated galaxies (Cao et al. 2016;He et al. 2022), while some other studies considered that their star formation is suppressed compared to that of isolated galaxies (Moon et al. 2019;Brown et al. 2023).The divergence in observations led to controversy about the physical mechanisms that regulate the star formation of galaxy pairs, including whether the effects of the hot gaseous halo are not negligible (Hwang & Park 2015;Zuo et al. 2018;Lisenfeld et al. 2019;Moon et al. 2019).
In addition to the properties of companion galaxies, some other factors, such as the projected separation and mass ratio, are also related to the star formation suppression in galaxy pairs (Davies et al. 2015;Ellison et al. 2022;Das et al. 2023).Behind these dependencies lie associated physical mechanisms, which are crucial for understanding the star formation properties during galaxy mergers.Clarifying whether star formation suppression exists in galaxy pairs and its dependency relationships are of great significance for further understanding the process of galaxy merging.
In this paper, we investigate the star formation of galaxy pairs based on a large galaxy pair sample, in particular focusing on the suppression of star formation.By analyzing the dependence of the occurrence of star formation suppression, we are attempting to gain insight into the physical process that occurs in paired galaxies.This paper is constructed as follows.We first introduce the galaxy pair sample (Section 2), then analyze the dependence of star formation suppression on the properties of galaxy pairs (Section 3).In Section 4, we discuss the physical origin of star formation suppression.Finally, we give a summary in Section 5. Throughout this paper, we use the standard cosmological model with H 0 = 70km s −1 Mpc −1 , Ω Λ = 0.7, Ω m = 0.3.

DATA
The galaxy pair sample is obtained from the SDSS DR7 main galaxy sample (Abazajian et al. 2009;Strauss et al. 2002), where all galaxies are brighter than m r = 17.77.We supplement a significant number of spectral redshifts from other spectral surveys, such as LAMOST (Luo et al. 2015) and GAMA (Baldry et al. 2018), to minimize the spectral incompleteness induced by the fiber collision effect (see more details in Shen et al. 2016 andFeng et al. 2019).After supplementation, 713, 366 galaxies have spectral redshifts and are named the spectroscopic sample).
We select galaxy pairs from the spectroscopic sample, ensuring that two member galaxies in each pair satisfy the following criteria: (1) the difference of Petrosian magnitude in SDSS-r band is |∆m r | < 2.5; (2) the projected separation between them fulfills 10 kpc < d p < 500 kpc; (3) the difference in line-of-sight velocity satisfies |∆v| < 1000 km s −1 .
To ensure that our galaxy pairs are isolated pairs rather than part of galaxy groups/clusters, We require that each member galaxy of a galaxy pair has no other neighbor galaxies except for the other member galaxy.In practice, for a pair member with an apparent magnitude of m r , there should be no other galaxy brighter than m r + 2.5 within a range of d p < 500 kpc and |∆v| < 1000 km s −1 in the spectroscopic sample.Additionally, we also require that there are no other neighbor galaxies around each pair member in the photometric sample in which galaxies have no spectral redshift measurement.The photometric sample contains two parts: galaxies in the SDSS main galaxy sample that is missed by the spectral observations, and galaxies with 17.77 < m r < 20.5.For a pair member with a redshift of z and magnitude of m r , if it has at least one neighbor galaxy brighter than m r + 2.5 within a range of d p < 500 kpc and |z − z phot | < δz phot in the photometric sample, then we consider that this galaxy pair is not isolated.The z phot and δz phot are the photometric redshift and its uncertainty of galaxies in the photometric sample, which are taken from Beck et al. (2016).In this step, we obtain 13, 321 isolated galaxy pairs.
For each pair member, we cross-match the stellar mass and global SFR with the MPA-JHU catalog 1 (Brinchmann et al. 2004;Kauffmann et al. 2003), and adopt the Sersic index from Simard et al. (2011) to represent galaxy morphology.Finally, we have 11, 265 galaxy pairs with the measurements of stellar mass (log M ⋆ ), star formation rate (log SFR), and Sersic index (n s ) for both two pair members.According to the morphology of pair members, we separate galaxy pairs into three categories: (1) 3, 275 S-S pairs, comprising two late-type galaxies (LTG, n s < 2.5); (2) 4, 918 S-E pairs, comprising one early-type galaxy (ETG, n s ≥ 2.5) and one late-type galaxy; (3) 3, 072 E-E pairs, comprising two early-type galaxies.
We also identify isolated galaxies from the spectroscopic sample for our following analysis.We define a galaxy with magnitude m r and redshift z as an isolated galaxy if it satisfies the following two criteria: (1) in the spectroscopic sample, there are no neighbor galaxies brighter than m r + 2 located within a range of d p < 500 kpc and ∆v < 1000 km s −1 from it.(2) in the photometric sample, there are no galaxies brighter than m r + 2 located within a range of d p < 500 kpc and |z − z phot | < δz phot from it.With these two criteria, we obtain 12345 isolated galaxies.

RESULT
In this study, we do not simply use the star-forming galaxies as the study sample because that may bring about selection effects when discussing the star-forming properties.Instead, we examine the global SFRs of LTGs, which by definition (n < 2.5) are not directly related to their star formation properties.As shown in the left panel of Figure 1, most of the LTGs (blue solid contours) are star-forming galaxies located in the star-forming main sequence (black dotted lines, adopted from the result of Salim et al. 2007).In contrast, ETGs (red dashed contours) fall below the star-forming main sequence.
In the following, we only consider the S-S pairs and the S-E pairs.For convenience, we use 'target galaxy' to denote the pair member whose SFR we aim to study.The other pair member is denoted as a 'companion galaxy'.In particular, for S-E pairs, the target galaxy is the late-type pair member, and the companion galaxy is the early-type pair member.In S-S pairs, two late-type pair members take turns to be the target galaxy and companion galaxy once each.
For each LTG in galaxy pairs, we match 5 isolated galaxies as control galaxies2 .The control galaxies should have the similar stellar mass (|∆ log M ⋆ | < 0.2), redshift (|∆z| < 0.02) and Sersic index (|∆n s | < 0.2) as the corresponding paired galaxies.Moreover, we require that control galaxies have morphologies similar to paired galaxies because the SFR is strongly correlated with morphology as shown in the left panel of Figure 1.This requirement would prevent us from underestimating the suppression of star formation by preventing false matching of isolated ETGs to the LTGs in galaxy pairs.We take the median SFR of the control galaxies to represent the characteristic SFR of the isolated galaxies (SFR 0 ).Then, we define the variation of SFR (either enhancement or suppression) where log SFR pair is the star formation rate of the paired galaxies.In the following sections, we use ∆ log SFR to explore whether and where star formation of paired galaxies is suppressed.

Projected Separation
The right panel of Figure 1 displays the variation of SFR as a function of the projected separation of the paired members.The gray dashed line shows the median value of ∆ log SFR at the given d p intervals for all galaxy pairs.The error bars represent the uncertainty of those median values estimated by the bootstrap method.The general trend of ∆ log SFR for all galaxy pairs (black dashed lines) is nearly the same as in previous studies (e.g.Li et al. 2008;Patton et al. 2013;Feng et al. 2019).At d p < 150 kpc, the SFR increases significantly as the projected separation decreases, indicating the influence of galaxy interaction.While at d p > 150 kpc, the SFR of galaxy pairs is almost identical to that of isolated galaxies.
However, when we investigate the behavior of ∆ log SFR separately for the S-S pairs (blue triangles) and S-E pairs (red circles), we find that the projected separation, below which galaxy pairs exhibit different star formation properties than isolated galaxies, has reached 350 kpc.This value is more than twice as large as previous results (e.g.∼ 150 kpc in Patton et al. 2016;Feng et al. 2019Feng et al. , 2020)).Furthermore, we notice that galaxy interaction not only triggers the enhancement of star formation but can also suppress star formation.
When companion galaxies are ETGs (in S-E pairs), ∆ log SFR of target galaxies have negative values at d p < 350 kpc, indicating suppressed star formation.The suppression starts from d p ∼ 350 kpc and becomes stronger as d p decreases.The strongest suppression of star formation occurs at d p ∼ 200 kpc.At d p < 200 kpc, the ∆ log SFR begins to increase as d p decreases.For very close galaxy pairs with d p < 30 kpc, the SFR of S-E pairs is nearly the same as that of isolated galaxies.In comparison, the star formation of LTGs with late-type companions (in S-S pairs) is dominated by the enhancement at d p < 150 kpc.At d p > 150 kpc, the SFR of S-S pairs are nearly comparable to isolated galaxies.There is no evidence that star formation has been suppressed in the ∆ log SFR-d p relation of S-S pairs, except for one data point at d p ∼ 450 kpc which may be attributed to the peculiarities of a few individual galaxies.Left: SFR as a function of stellar mass for LTGs (ns < 2.5, blue contours) and ETGs (ns ≥ 2.5, red contours).The black dotted line represents the star-forming main sequence of Salim et al. (2007).Right: The variation of SFR between LTGs in galaxy pairs and isolated galaxies as a function of projected separation for S+S pairs (blue triangles), S+E pairs (red circles), and all pairs (gray solid line).

Stellar Mass of Pair Members
Many studies demonstrated that the stellar mass of galaxy pairs has a strong impact on the enhancement of star formation (e.g.Scudder et al. 2012), which may also affect the suppression of star formation.To better understand the behavior of star formation suppression, we investigate the relationship between the suppression and the stellar mass of pair members.Particularly, we focus on the galaxy pairs with d p < 400 kpc, at which the significant suppression of star formation mainly occurs according to the right panel of Figure 1.
Figure 2 displays the ∆ log SFR as a function of the stellar mass of the pair members.The upper panels show the result of S-E pairs, and the lower panels show the result of S-S pairs.Each column represents the different projected separation range, which is labeled in the upper left corner of each panel.In each panel, the x-axis represents the stellar mass of the target galaxies, and the y-axis represents the stellar mass of the companion galaxies.The ∆ log SFR of target galaxies are displayed by color-coded circles.The red color represents the suppression of star formation (∆ log SFR < 0), while the blue color represents the enhancement of star formation (∆ log SFR > 0).To display the general trend, the values of ∆ log SFR are smoothed by the LOESS algorithm (Cappellari et al. 2013).For convenience, we use black dashed lines to label ∆ log SFR = −0.2 to better illustrate the dependency of the star formation suppression on the stellar mass.
These figures reinforce the crucial role of the morphology of companion galaxies and projected separation revealed by Figure 1.Firstly, the early-type morphology is more conducive to inducing the suppression of star formation.We find that the significant signals of star for-mation suppression (∆ log SFR < −0.2) only appeared within the S-E pairs.When comparing the top panels to the bottom panels, it is shown that the S-E pairs are more likely to display negative ∆ log SFR than the S-S pairs with given stellar mass and projected separation.Second, the suppression of star formation is related to the projected separation.At 200 < d p < 300 kpc, the S-E galaxy pairs exhibit a remarkably significant suppression of star formation, as demonstrated in Figure 1.Additionally, we observed that close galaxy pairs with d p < 100 kpc can also display strong suppression of star formation.This phenomenon is only evident when we analyze galaxy pairs based on stellar mass separately.Hence, the stellar mass of member galaxies is another key factor in the star formation suppression observed in galaxy pairs.
The dependence of the star formation suppression on the stellar mass of member galaxies varies at different projected separations.For close S-E galaxy pairs with d p < 100 kpc, the suppression of star formation primarily depends on the stellar mass of the companion galaxy.There is a characteristic stellar mass of companion galaxies, approximately log M ⋆ ∼ 11.0, above which the star formation in the target galaxy shows significant suppression.Below this value, the SFR of the target galaxy is enhanced, consistent with previous findings in studies on galaxy pairs (Ellison et al. 2008;Li et al. 2008;Patton et al. 2016;Feng et al. 2019).For close galaxy pairs, both stellar formation suppression and enhancement may occur.Specifically in our galaxy pair sample, the S-E pairs with log M ⋆ > 11.0 companions display negative ∆ log SFR, while the S-E pairs with log M ⋆ < 11.0 companions display positive ∆ log SFR.Due to the relatively low number of S-E galaxy pairs .The variation of SFRs as a function of the stellar mass of galaxy pair members.The x-axis shows the mass of the target galaxies, and the y-axis displays the companion galaxies.The color coding represents the variation of SFRs for target galaxies and the black dashed contours represent the ∆ log SFR = −0.2.The top rows show the result when companion galaxies are early-type galaxies, whereas the bottom rows show the result when companion galaxies are late-type galaxies.The projected separation range is labeled in the upper left corner of each panel.
with massive companion galaxies, the signal of star formation suppression is obscured by the signal of enhancement when calculating the average ∆ log SFR of all S-E galaxy pairs with d p < 100 kpc.Therefore, we cannot observe the star formation suppression of close galaxy pairs in Figure 1.At 200 kpc < d p < 300 kpc, the star formation suppression is primarily determined by the stellar mass of the target galaxy rather than the companion galaxy.According to the results of the top third panel in Figure 2, the significant star formation suppression mainly occurs in S-E pairs where the stellar mass of the target galaxy is greater than log M ⋆ ∼ 11.0.The close galaxy pairs and wide galaxy pairs exhibit completely different dependencies, suggesting that the star formation suppression phenomenon in these two types of galaxy pairs may have different physical origins.We speculate that the star formation suppression in close galaxy pairs may originate from the interaction with the hot gas surrounding the companion galaxy (see more discussion in Section 3.3 and Section 4.2.1), while the suppression phenomenon in wide galaxy pairs may be a manifestation of galaxy conformity (see more discussion in Section 4.3).

Sersic Index of Companion Galaxies
Compared to LTGs, ETGs are evolved galaxies that have experienced more galaxy merging processes.Because galaxy merging is conducive to the formation of hot gaseous halos, the correlation between the morphology of companion galaxies and star formation suppression at d p < 100 kpc may imply that the hot gaseous halos of companion galaxies play an important role in the suppression of star formation.To test this hypothesis, we further examine how the suppression depends on the Sersic index of the companion galaxies which is an indicator of their merging history.
We divide ETGs and LTGs in companion galaxies into two subcategories according to the Sersic index (n C ) respectively and display the results in Figure 3.In this figure, the Sersic index of target galaxies (n T ) is smaller than 2.5 consistent with Figure 2, and the range of the Sersic index of companion galaxies is labeled in the upper left corner of each penal.
These four plots clearly demonstrate the close correlation between the morphology of the companion galaxy and the suppression of star formation in the target galaxy.The higher the Sersic index of the companion galaxy, the stronger the suppression of star formation  in the target galaxy.When n C is less than 1.5, the star formation in all target galaxies is enhanced.When 1.5 < n C < 2.5, the trend of enhanced star formation in the target galaxies, whose companion galaxies satisfy log M ⋆ > 11.0, is significantly weakened.When n C is greater than 2.5, the target galaxies with companion galaxies of stellar mass greater than log M ⋆ ∼ 11.0 begin to exhibit negative values in the ∆ log SFR.For the galaxy pairs with the highest n C values (n C > 4.0), the suppression of star formation is the strongest.Among them, for the target galaxies with companion galaxies having log M ⋆ > 11.0, the value of ∆ log SFR has already dropped below −0.2dex.Based on this phenomenon, we conclude that the more evolved companion galaxies are more likely to trigger the suppression of star formation in target galaxies.
We also analyzed the correlation between the Sersic index of the companion galaxies and the suppression of star formation when the Sersic index of the target galaxies is less than 1.5.The results remained unchanged.Therefore, the correlation observed in this section indeed originates from the morphology of the companion galaxies rather than the morphology of the target galaxies.

Morphology of Target Galaxies
The morphology of target galaxies can also affect star formation properties of themselves during the merging process according to the previous studies (Barnes & Hernquist 1996;Di Matteo et al. 2008).In this section, we investigate the role of the target's morphology in the suppression of star formation.We display the results in Figure 4, where all companion galaxies satisfy n C > 2.53 .The range of the Sersic index of target galaxies (n T ) is labeled in the upper left corner of each panel.
Comparing these two panels, we find that the Sersic index of target galaxies indeed influences the properties of star formation.Firstly, for galaxy pairs with companion galaxies log M ⋆ > 11.0, the larger the Sersic index of the target galaxy, the stronger the suppression of star formation.Target galaxies with n T < 1.5 exhibit only very weak signals of star formation suppression, while those with n T > 1.5 have ∆ log SFR < −0.2.Secondly, for galaxy pairs with companion galaxies log M ⋆ < 11.0, the larger the Sersic index of the target galaxy, the weaker the enhancement of star formation.We suggest that the lower Sersic index of target galaxies is conducive to enhanced star formation, which can weaken the phenomenon of star formation suppression (see more discussion in Section 4.2.2).

Comparison to Previous Studies
In general, our results do not conflict with previous studies.Because of the different criteria for selecting galaxy pairs and different methods for analyzing star formation, our study unveils some new phenomena.
Our galaxy pair sample is selected from the SDSS survey and spans a wide range of stellar mass and mass ratio between pair members, which enables us to explore the dependence on stellar mass in detail.If we focus on the close major merger pairs with log M ⋆ > 10.0 (similar to the galaxy pairs in Xu et al. 2010), the average SFR of the S-E pairs is also comparable to the isolated galaxies, in agreement with previous studies (Xu et al. 2010;Yuan et al. 2012;Cao et al. 2016).As we discussed in Section 3.2, that is actually the combination of the enhancement in target galaxies with log M ⋆ < 11.0 companions and suppression in target galaxies with log M ⋆ > 11.0 companions.
In our study, we analyzed the SFR of LTG rather than star-forming galaxies.This avoids underestimating star formation suppression due to the omission of galaxies whose star formation has already been suppressed.In some previous studies, the suppression of star formation unveiled by the SFR of star-forming galaxies is very limited, but the suppression of star formation can still be reflected by estimating the fraction of quiescent galaxies (Moon et al. 2019).Therefore, these previous results on the suppression of star formation are generally consistent with our study.
At last, we should mention that most paired galaxies showing suppressed star formation are still star-forming galaxies rather than quiescent galaxies.On one hand, the results of ∆ log SFR in Figure 3.2 indicate that the degree of star formation suppression does not exceed 0.5dex.On the other hand, the distribution of sSFR for galaxies exhibiting star formation suppression also supports this point.We analyzed the distribution of sSFR for target galaxies in S-E galaxy pairs with d p < 100 kpc, and found that 75% of target galaxies with ∆ log SFR < −0.2 still have sSFR greater than −11.5, indicating that they are still star-forming galaxies.

Physical Origin of Star Formation Suppression in Close Galaxy Pairs
The complicated dependence mentioned in the above sections shows that we cannot explain the star formation suppression in galaxy pairs with a single mechanism.In the following, we discuss two possible physical mechanisms, each of which can explain a part of the observed phenomena, but we do not rule out the contribution of other mechanisms.

Galaxies
The hot circum-galactic medium (CGM) around companion galaxies is one of the explanations for the star formation suppression in target galaxies (Park & Choi 2009;Moon et al. 2019).Theoretical studies suggest that massive passive galaxies are surrounded by hot gaseous halos, which can shut off the cold gas supply (Dekel & Birnboim 2006).Because the hot halo can extend to the virial radius, it is possible that neighbor galaxies will also be affected by the hot gas.Numerical simulations found that the existence of hot gaseous halos is indeed able to induce the star formation of merging galaxies lower than the isolated galaxies (Moster et al. 2011;Karman et al. 2015;Hwang & Park 2015).
In observation, the dependence of hot CGMs on the stellar mass and morphology is highly similar to that of star formation suppression.First, the hot gas content in CGMs is correlated with the morphology of their host galaxies.Both individual detection and statistical studies through the stacking method find that diffuse X-ray emission around ETGs is more luminous than that of LTGs (Anderson et al. 2013;Li et al. 2017), indicating that ETGs prefer to hold hotter gaseous halos.In contrast, CGM around LTGs contains more cold gas (Lan 2020).Second, the hot gas content shows a positive correlation with the stellar mass of host galaxies (Anderson et al. 2013;Goulding et al. 2016;Forbes et al. 2017).There is also a characteristic stellar mass around log M ⋆ ∼ 11.0, below which the signal of hot gas in CGMs is almost undetectable (Greco et al. 2015;Comparat et al. 2022).This similarity indicates the suppression of star formation in target galaxies may be induced by the hot gas around companion galaxies.
Based on the above discussion, we speculate that during close encounters of galaxy pairs, besides the dynamical effects of cold gas, the CGM properties of the companion galaxy can also significantly influence the star formation of galaxies.If the CGM of the companion galaxy contains a large amount of hot gas, the inflow of cold gas in the target galaxy will be greatly weakened, leading to the suppression of star formation.

Morphology of Target Galaxies
The structure of galaxies can influence the properties of star formation during the galaxy merging process.Numerical simulations predicted that massive bulges in disk galaxies can stabilize the gas distribution and suppress the starburst during galaxy encounters by preventing tidal-induced gas inflow (Barnes & Hernquist 1996;Di Matteo et al. 2008).Observational studies confirm that the SFR of paired galaxies is indeed strongly correlated with the bulge-to-total ratio, where only galaxies with a low bulge-to-total ratio will exhibit enhancement of star formation (He et al. 2022).In our study, we find that the target galaxy with a lower Sersic index, which approximately equals a smaller bulge-to-total ratio, more easily exhibits enhancement of star formation instead of suppression, consistent with previous findings.
Combined with the above results, we suggest that the enhancement of star formation and the suppression of star formation is dominated by two independent pro-cesses that exist simultaneously.The morphology of target galaxies regulates the enhancement of star formation by stabilizing the existing gas within the disks of target galaxies which can affect the inflow of cold gas, while the morphology and stellar mass of companion galaxies regulate the suppression of star formation by shutting off the cold gas supply through the effect of hot gaseous halos.For n s < 1.5 galaxies with early-type companions, the enhanced star formation caused by their own morphology will offset the suppression of star formation caused by their companion galaxies, resulting in reduced suppression of star formation in those galaxies.

Galactic Conformity
Galactic conformity, first proposed by Weinmann et al. (2006), refers to the phenomenon where the star formation properties of satellite galaxies within a galaxy group tend to be similar to those of the central galaxy.If the central galaxy is an early-type galaxy, satellite galaxies often are quiescent as well.This phenomenon is quite similar to the dependency observed in our findings regarding star formation suppression.In many studies on galactic conformity (e.g.Weinmann et al. 2006;Phillips et al. 2014), the galaxy group samples typically included a large number of galaxy groups with only two member galaxies.Therefore, galactic conformity also exists in galaxy pairs.Because the galaxy interactions between pair members typically do not extend beyond d p ∼ 150 kpc (e.g.Patton et al. 2016;Feng et al. 2020), we infer that the observed star formation suppression in wide galaxy pairs with 200 kpc < d p < 350 kpc may indeed be a manifestation of galactic quenching.
It is generally believed that galactic conformity arises from the properties of host dark matter halos of galaxy groups, such as the assembly history of dark halos.Therefore, the premise for two galaxies to exhibit galactic conformity is that they are located within the same dark matter halo, in other words, they are gravitationally bound.Based on the relationship between stellar mass and dark halo mass proposed by Velander et al. (2014) and the assumption of a spherical dark matter halo, a galaxy with log M ⋆ ∼ 11.0 has a virial radius of 300 kpc.Other galaxies within d p < 300 kpc would fall into the host dark halo of this galaxy, thus altering their star formation properties.
Specifically regarding the results in Figure 2, at d p < 300 kpc, galaxy pairs containing galaxies with log M ⋆ > 11.0 are gravitationally bound, and the member galaxies exhibit galactic conformity due to the influence of the properties of the host halo, thereby showing star formation suppression.However, when d p > 300kpc, except for a few galaxy pairs with very massive pair members (e.g.log M ⋆ > 11.5), two member galaxies are not located within the same dark halo, and therefore, star formation suppression does not naturally occur.At d p < 200 kpc, the interactions between galaxy pairs can enhance star formation (Patton et al. 2016;Feng et al. 2020).As a result, the degree of star formation suppression is weakened at 100 kpc < d p < 200 kpc.

SUMMARY
In this paper, we investigate the suppression of star formation in galaxy pairs based on the isolated galaxy pair sample derived from the SDSS main galaxy sample.We select 13, 321 isolated galaxy pairs with the criteria of |∆m r | < 2.5, d p < 500 Mpc and |∆v| < 1000 km s −1 .For a pair member with a magnitude of m r , there are no other neighbor galaxies brighter than m r +2.5 within d p < 500 Mpc and |∆v| < 1000 km s −1 .
By comparing the SFR of late-type pair members (n s < 2.5, denoted as target galaxies) to isolated galaxies, we find the following results about the star formation suppression.
1.The suppression of star formation depends on the projected separation between pair members.The significant signal of the suppressed star formation mainly happens in galaxy pairs at d p < 100 kpc and 200 kpc < d p < 350 kpc.
2. The occurrence of star formation suppression requires the companion galaxy's Sersic index satisfying n s > 2.5, and the degree of suppression increases as the Sersic index of the companion galaxy increases.For target galaxies with a companion galaxy Sersic index less than 2.5, there is no evidence of star formation suppression.
3. In close galaxy pairs (d p < 100 kpc), the suppression of star formation further requires the companion galaxy to have a large stellar mass (log M ⋆ > 11.0).For target galaxies with a companion galaxy log M ⋆ < 11.0, the star formation is dominated by the enhancement.Besides, the morphology of target galaxies can also influence the suppression of star formation.The star formation of target galaxies with 1.5 < n s < 2.5 is more easily suppressed than those with n s < 1.5.
4. In wide galaxy pairs (200 kpc < d p < 300 kpc), the suppression of star formation requires the target galaxy to have a large stellar mass (log M ⋆ > 11.0).
According to the dependency of star formation suppression, we discuss the possible origin of this phenomenon.We suggest that the suppression of star formation in close galaxy pairs might be the combined effect of the hot circum-galactic medium around companion galaxies and the structure of target galaxies.In wide galaxy pairs, star formation suppression primarily arises from galaxy conformity.To further investigate the physical mechanisms underlying this phenomenon, multi-wavelength studies and numerical simulations are necessary.

Feng
Figure1.Left: SFR as a function of stellar mass for LTGs (ns < 2.5, blue contours) and ETGs (ns ≥ 2.5, red contours).The black dotted line represents the star-forming main sequence ofSalim et al. (2007).Right: The variation of SFR between LTGs in galaxy pairs and isolated galaxies as a function of projected separation for S+S pairs (blue triangles), S+E pairs (red circles), and all pairs (gray solid line).
Figure2.The variation of SFRs as a function of the stellar mass of galaxy pair members.The x-axis shows the mass of the target galaxies, and the y-axis displays the companion galaxies.The color coding represents the variation of SFRs for target galaxies and the black dashed contours represent the ∆ log SFR = −0.2.The top rows show the result when companion galaxies are early-type galaxies, whereas the bottom rows show the result when companion galaxies are late-type galaxies.The projected separation range is labeled in the upper left corner of each panel.

Figure 3 .
Figure 3. Similar with Figure 2. All pairs are dp < 100 kpc, and the target galaxies satisfy nT < 2.5.The four panels represent four Sersic index intervals of companion galaxies which are labeled in the top left corner.

Figure 4 .
Figure 4. Similar with Figure 2. All pairs are dp < 100 kpc, and the companion galaxies satisfy nC > 2.5.The two panels represent two Sersic index intervals of target galaxies which are labeled in the top left corner.