Gaugino portal baryogenesis

We study baryogenesis via a gaugino portal, the supersymmetric counterpart to the widely studied kinetic mixing portal, to a hidden sector. CP and baryon number violating decays of a hidden sector gaugino into the visible sector can produce the observed baryon asymmetry of the Universe. The tiny portal coupling is crucial in producing late out-of-equilibrium decays, after washout processes that can erase the asymmetry have gone out of equilibrium. We study this mechanism within various scenarios, including freeze-in or freeze-out of the hidden gaugino, as well as extended frameworks where the hidden sector contains a weakly interacting massive particle (WIMP) dark matter candidate. This mechanism can produce the desired asymmetry over a wide range of mass scales, including for hidden gaugino masses as low as 10 GeV. We also discuss possible related signals with direct collider searches, at low energy experiments, and in dark matter direct and indirect detection.


Motivation
The preferred picture for the underlying theory of nature has changed in the past few years. R-parity preserving weak scale supersymmetry, which offered a natural solution to the hierarchy problem as well as a WIMP dark matter candidate, is now constrained by data from a wide variety of experiments, ranging from the Large Hadron Collider (LHC) to dark matter indirect and direct detection efforts. Despite its many appealing theoretical features, supersymmetry, if realized in nature -as a vestige of a theory of quantum gravity, or as a partial solution to the hierarchy problem -might therefore be neither R-parity preserving nor at the weak scale.
There has also been growing interest in the exploration of hidden or dark sectors, which can easily arise, for example, in string theories. Such sectors may communicate with our visible sector via one of the renormalizable portal interactions, giving rise to a rich array of phenomenological possibilities. (See [1] and references therein.) The most extensively studied of these is the kinetic mixing portal [2], where a gauged U(1) in a hidden sector has kinetic mixing with the Standard Model (SM) hypercharge U(1) Y [3], resulting in a mixing between the corresponding gauge bosons.

JHEP06(2019)096
In this paper, we explore how a hidden sector coupled to the Minimal Supersymmetric Standard Model (MSSM) via a kinetic mixing portal [4] may address one of the outstanding problems of particle physics and cosmology, the origin of the observed excess of matter over antimatter in our Universe. Supersymmetry can readily satisfy the Sakharov conditions [5] necessary for baryogenesis: baryon number violation (from R-parity violating (RPV) operators), C and CP violation (from soft terms after supersymmetry breaking), and out of equilibrium interactions (from late decays of supersymmetric particles). In a supersymmetric scenario, kinetic mixing can occur between U(1) gauge groups at the level of the supersymmetric field strengths, resulting in a gaugino mixing portal in addition to the more familiar gauge boson kinetic mixing portal. Moreover, there is a possibility for mass mixing between the gauginos. In this paper, we make use of late decays of the hidden sector gaugino into the visible sector via these mixings to produce the observed baryon asymmetry of the Universe.
Within the MSSM, the most challenging aspect of implementing baryogenesis via RPV decays of neutralinos [6][7][8][9][10][11][12][13] is to ensure that the decays occur sufficiently late as to avoid washout of the produced asymmetry from baryon number violating inverse decay and scattering processes without suppressing the production mechanism. In low scale baryogenesis mechanisms one must also be wary of inducing a too-large CP violation that would be visible in electric dipole moment (EDM) experiments, which are now strongly constrained by measurements. (For recent discussions of the implications of EDM bounds for supersymmetric theories, see [14,15].) In this paper, we demonstrate that the hidden sector implementation offers several novel features: the small portal mixing between the two sectors naturally provides the means to address the above problems, enabling low scale baryogenesis with rich phenomenology.
Furthermore, while R-parity violating supersymmetry has an unstable lightest supersymmetric particle (LSP) and therefore no dark matter candidate 1 (except possibly a longlived gravitino, see, e.g., [12]), the hidden/dark sector can contain additional particles that account for the dark matter. In this paper, we will study both a minimal hidden sector (where the only hidden sector particle relevant for baryogenesis is the hidden sector gaugino) and an extended hidden sector containing a WIMP dark matter particle, which also impacts the process of baryogenesis.

Framework
Our framework consists of the MSSM extended by two key ingredients: baryon number violation via an RPV coupling, and a kinetic mixing portal to a U(1) symmetry residing in a hidden sector.
For baryon number violation, we add to the MSSM superpotential the following RPV term

JHEP06(2019)096
We set other RPV terms, which break lepton number and can induce proton decay, to zero. The above RPV coupling λ ijk is constrained by various measurements, 2 depending on the flavor indices (i, j, k). In this paper, for simplicity we work with a single λ ijk without specifying the flavor indices.
Regarding the field content of the hidden sector, two variations are worth studying. One can consider a minimal setup where all of the relevant hidden sector gaugino phenomenology arises from the portal coupling to the visible sector. Alternatively, additional particles in the hidden sector may endow the hidden gaugino with additional interactions, which can affect baryogenesis. We now outline these two possibilities in turn, before returning to discuss their phenomenology.

Minimal hidden sector
Kinetic mixing between two abelian gauge field strength superfields has been studied in [4,16,17]; the relevant Lagrangian is where we use the notation W v and W h to denote the chiral gauge field strength superfields for the visible sector SM hypercharge U(1) Y and the hidden sector U(1) gauge symmetries respectively. Here, η is the kinetic mixing portal coupling between the two fields. Typically, η ∼ 10 −3 if the mixing is induced at one loop by chiral superfields charged under both gauge groups, but much smaller values are possible in, e.g., compactifications of heterotic and type II strings [4,[19][20][21][22][23][24]. The gauge kinetic terms can be made canonical by shifting the visible and hidden sector vector superfields V v and V h . (See [4,16,17] for details.) If the two gauginos have Majorana masses M v , M h , these shifts result in a small Dirac mass term ∼ ηM h between the two gauginos. Depending on the underlying model of supersymmetry breaking, an additional primordial Dirac mass term might be present, which may well be of the same order. While one could, in principle, proceed by eliminating the kinetic mixing via field redefinition, followed by diagonalizing the mass matrix, in this paper we work with a "simplified model" setup for the gaugino sector that captures the same effects: where we define to be the mixing between the hidden gaugino,B , and the MSSM bino, B, but do not imagine there is otherwise any kinetic mixing in the gaugino sector. 3 We expect ∼ η unless there are strong cancellations between the various contributions to the Dirac mass term between theB and theB. (See [16] for related discussion.) Upon diagonalization of the neutralino mass matrix, this mass mixing will induce O( ) couplings between the (primarily) hidden neutralino and the visible sector. We assume mB ∼ mB, as would be the case, e.g., with gravity mediation of supersymmetry breaking to both sectors.

JHEP06(2019)096
The MSSM binoB can mix with the other MSSM neutralinos, the wino and the Higgsinos. Such mixing is suppressed by O(M Z /µ), with µ the Higgsino mass term. If the hidden sector U(1) is spontaneously broken by a hidden Higgs vacuum expectation value (vev), one analogously also has mixing of the hidden sector gaugino with the hidden sector Higgsinos (for simplicity, we assume two Higgsinos for the sake of anomaly cancellation, as in the MSSM), which can also be suppressed by raising the hidden Higgsino mass term µ . In this paper, for simplicity, we assume µ M Z and µ M Z . This decouples the remaining neutralinos, and the only relevant gauginos for our purposes are the MSSM binõ B and the hidden sector gauginoB .
From now on, we will use the notationB ,B to denote the mass eigenstates, which contain admixtures of the gauge eigenstate of the opposite sector. Crucially, we also assume that the two mass terms carry a relative phase: Im(mB m * B ) = 0. This will be the source of CP violation necessary for baryogenesis. Finally, we assume thatB is the LSP, so it can only decay via the RPV coupling.
In addition to the hidden gauginoB , the hidden sector contains, at minimum, Higgs bosons h , HiggsinosH , and the dark gauge boson Z . These particles decay much faster than theB : for the mass hierarchy mH > m h > mB , the dark Higgsino decays as H → h B , the dark Higgs boson decays as h →B B via small but non-vanishingB −H mixing, while the Z boson undergoes 2-body decays into the SM via a kinetic mixing term. TheB interactions with both the Z and the h vanish as the HiggsinosH are decoupled. Therefore, the presence of these additional particles in the hidden sector is cosmologically irrelevant, and all relevant interactions of theB arise via its mixing with the visible sector binoB.

Extended hidden sector with WIMP dark matter
The presence of additional particles in the hidden sector is an attractive possibility from the point of view of dark matter, since the traditional dark matter candidate, the visible sector LSP, is unstable in the absence of R-parity. 4 While hidden sectors allow for a wide variety of possibilities in terms of additional field content, there are two important considerations to keep in mind: (1) additional field content should not introduce newB decay channels into the hidden sector, as these would dominate over the suppressed decay process generating the baryon asymmetry, and (2) additional field content should preserve the cancellation of anomalies related to the gauged U(1) . There are two straightforward ways to ensure anomaly cancellation: adding a U(1) singlet, or adding vector-like fields.
Adding a singlet opens up possible additional portal interactions between the hidden and visible fields. While such setups admit interesting possibilities, these are not directly related to the baryogenesis process of interest. Therefore, we consider instead a vector-like fermion X, with mass m X , charged under the U(1) with gauge coupling g D , which is stable under a Z 2 symmetry and therefore a good dark matter candidate. In particular, for weak scale m X > M Z , it can realize the correct abundance via the "WIMP miracle" through the freeze-out process XX → Z Z , where Z is the hidden sector gauge boson. Furthermore, while it does not introduce any newB decay channels, it does introduce additional Yukawa couplings for theB proportional to the hidden gauge coupling, involving its scalar superpartnerX. As we will see in the following sections, this can affect the cosmological history of the hidden gaugino and therefore the production of the baryon asymmetry, offering an interesting variation to the minimal setup discussed above.

Calculation of baryon asymmetry
The framework described in the previous section contains all the necessary ingredients for baryogenesis via the early Universe production and decay of the hidden gauginoB into SM fermions in the early Universe. The necessary ingredients are the Sakharov conditions [5] of baryon number violation, C and CP violation, and out of equilibrium interactions. Baryon number violation is made possible by the RPV interaction in eq. (2.1). The out of equilibrium condition is provided by lateB decays. Finally, CP violation arises from the interference of the tree and loop level decays shown in figure 1 via the non-trivial phase in the gaugino masses.
The baryon asymmetry generated with this setup can be calculated as: where YB = nB /s is the freeze-out abundance of theB , and (where ΓB is the total decay width of theB ) represents the fraction ofB decays that produce a baryon asymmetry. W I incorporates the washout or dilution of the asymmetry from subsequent processes. Note that, due to the long lifetime of theB , its production/freezeout and decays occur at different epochs, and the related quantities, YB and CP respectively, can be calculated independently. In the remainder of this section, we discuss the calculation of each of the above contributions.

Calculation of CP
Since the decay processes shown in figure 1 depend on as well as λ , the decay can have a long lifetime, easily satisfying the out of equilibrium condition.
Here, it is important to note that there exists another decay channel in addition to those shown in figure 1. As a consequence of the Nanopoulos-Weinberg theorem [25], in order for the generated CP asymmetry to be first order in the RPV coupling, theB must have an additional decay channel when the RPV coupling is set to zero. (See [9], also [13], for a detailed discussion.) This is realized for mB < mB , 5 for which theB has the additional decay channelB →Bff , which is independent of the RPV coupling λ , but notably depends on . Including this channel, the decay widths relevant for the calculation of the baryon asymmetry are [11,12,26] , and m 0 is the sfermion mass scale. φ = φB−φB is the relative phase between the two gaugino masses, and from here on we set it to its maximum value, Im[e 2iθ ] = 1, which maximizes the baryon asymmetry. There are in principle additional decay channels, such asB →B(h/Z), from the suppressed bino-Higgsino mixing. For sufficiently large values of |µ| > 8π g 1 m 0 mB 2 m Z , which we assume to be the case, this contribution is negligible compared to the channels above.
With the above decay rates, we have The various suppression factors in this expression are intuitive. The 4π in the first denominator represents the loop factor suppression in the tree-loop interference relative to the tree level decay. The second fraction represents the additional coupling in the interference term. Finally, the 1/m 2 0 factor in the third fraction can be understood from the presence of an additional sfermion propagator in the tree-loop interference diagram relative to the tree level process. Given the mass hierarchy m 0 > mB > mB, the above expression evaluates to ∼ < 10 −4 .
We now highlight some important aspects of theB decay. Suppose were not small. Since theB →Bff channel is independent of the RPV coupling λ , one cannot make theB long-lived simply by suppressing λ . This could be accomplished, instead, by JHEP06(2019)096 raising the sfermion mass scale m 0 relative to the gaugino masses (as occurs in splitsupersymmetry scenarios [27][28][29][30]), as implemented in the MSSM baryogenesis scenarios in [11,12]. However, this has the effect of also suppressing CP (see the final factor in eq. (3.6)), therefore suppressing the production of the baryon asymmetry. The hidden sector implementation is qualitatively different and advantageous in this regard: the tiny portal coupling suppresses all decay rates, thereby making theB long-lived, but without an accompanying suppression of CP .

Calculation of YB : minimal hidden sector
In the minimal implementation, there are three distinct types of cosmological histories 6 determining the hidden gaugino abundance YB in the early Universe.
Conventional freeze-out. If is sufficiently large that the 2 → 2 scattering processes shown in figure 2 are rapid (compared to the Hubble rate) in the early Universe, theB thermalizes with the SM bath and subsequently undergoes conventional freezeout. While in equilibrium, its number density is given by [26] n eq (B ) = where x = mB /T and K 2 is the modified Bessel function of the second kind. The thermally averaged cross sections for the processes in figure 2 are [12] σv In our studies, we calculate the freezeout abundance YB by numerically solving the Boltzmann equation for theB (for details, see [12]) with the above interaction terms.
Thermal abundance from decays. For smaller values of , the 2 → 2 scattering processes above are not sufficiently rapid to thermalize theB . Nevertheless, an equilibrium abundance of theB (eq. (3.7)) can be achieved from decays such asf →B f at T > mB . This scenario is realized for α 1 2 > m 0 /M Pl , and results in a large yield YB ≈ 5 × 10 −3 .

JHEP06(2019)096
Freeze-in. For even smaller values of , YB never reaches thermal abundance, and a smaller abundance accumulates instead via the freeze-in process [31]. Of the feeble processes, the contribution from decays of sfermions while they are in equilibrium,f →B f , generally dominates over contributions from scattering processes in the IR, and contributes [31] YB ≈ 0.03 α 1 2 π 3 M Pl m 0 . (3.10)

Calculation of YB : extended hidden sector
The presence of the vector-like fermion X in the hidden sector could drastically change the picture. In this case, theB interacts with X via the t-channel exchange of the sfermioñ X. This is analogous to the familiar case of the freeze-out of the MSSM bino with leptons via the exchange of a slepton mediator, and has the annihilation cross section [32]: where r = m 2B /m 2 f . This cross section is independent of and can therefore dominate the early Universe dynamics of theB despite the Boltzmann suppressed abundance of theB at T < mB . If this interaction controls the freeze-out of theB , the abundance YB can be estimated, in the limit r = m 2B /m 2 In our studies, we will focus on regions of parameter space where the WIMP-like freezeout XX → Z Z sets the correct relic density of dark matter, andB B → XX determines the freeze-out abundance of theB . This requires not only m Z < m X , but also large enough that the visible and hidden sectors are in equilibrium at some point in the early Universe, and mB > m X so that theB freeze-out can occur at T < mB . Note that requiring the correct dark matter abundance fixes a relation between g D and m X : in the limit m Z /m X 1, we have g D ≈ 0.5 m X /TeV.

Washout and dilution effects
In our framework, the main effects that can wash out the produced baryon asymmetry are the baryon number violating inverse decays or 2 → 2 annihilations involving theB, for which the rates are [12]  where z = mB/mB . Washout effects are important if the above processes have not gone out of equilibrium (i.e., are rapid compared to the Hubble rate) at the time of theB decay. However, even in this case, since the decay of the entireB ensemble is not instantaneous, the reducedB abundance YB e −ΓB t remaining at the time the washout processes go out of equilibrium (Γ ID +Γ S < H) can still be sufficient to produce the desired baryon asymmetry. Another effect that can suppress the baryon asymmetry is entropy dilution from the late decays of theB . TheB energy density redshifts slower than radiation for T < mB . Since theB is long-lived, its energy density can therefore grow to dominate the total energy density of the Universe at late times, when YB mB /T > 1. The eventual decay of theB population injects this entropy into the thermal bath, diluting the abundance of the baryon asymmetry. This dilution factor can be estimated as [33] S after S before ≈ 1.83 g (3.14) This significant entropy injection also raises the temperature of the thermal bath to which affects the calculation of washout effects. We take these dilution and reheating effects into account where relevant.

Results and discussion
In this section, we present the results of our calculation of the baryon asymmetry using the formalism described in the previous sections. For both the minimal and extended hidden sector setups, we perform numerical scans over the relevant parameter space, exploring regions that can produce the observed amount of baryon asymmetry in the Universe, Y BA ≈ 10 −11 .

Minimal hidden sector
In the minimal scenario, the free parameters are mB , mB, m 0 , , and λ . To understand the various factors at play, we first plot various regions of interest in the mB − plane in figure 3, fixing the remaining parameters as specified in the caption. As discussed earlier, there are several classes of viable cosmological histories that produce the observed baryon asymmetry: theB can thermalize and freeze out via scattering processes (green), reach thermal abundance due to the decay processf →B f (yellow), or build up a small abundance from freeze-in (blue). Other colors indicate problematic regions, where theB decay occurs after big bang nucleosynthesis (BBN), taken to be T = 1 MeV (red), or where the baryon number violating inverse decay and/or annihilation washout processes suppress Y BA below the desired value (black). In the dark green region, washout processes are active at the temperature at which theB decays, but a sufficient amount of baryon asymmetry survives, as discussed earlier. (See discussion after eq. (3.13).) It is straightforward to understand the transitions between these regions as a function of . Extremely small leads to too-long lifetime (red). Increasing not only reduces the JHEP06(2019)096 lifetime (white) but also leads to greater production of theB through the freeze-in process f →B f , eventually enabling sufficient production of the baryon asymmetry (blue). As is further increased, this decay process leads to a thermal abundance of theB from decay processes (yellow). Eventually scattering processes (figure 2) also become rapid, and control the freeze-out of theB (green). For even larger values of , the decay lifetime can become so short that the decay occurs while washout processes are still active, partially erasing the baryon asymmetry (dark green) and eventually making it impossible to realize the desired abundance (black).
The change in the behavior of the black and red regions below mB = 100 GeV is due to our choice of m 0 = max(1 TeV, 10 mB ), in view of LHC bounds on sub-TeV squarks. We also note that, in the majority of the yellow and blue points, theB dominates the energy density of the Universe at the time of its decay, resulting in dilution of the baryon asymmetry (see discussion surrounding eqs. (3.14), (3.15)); the final abundance can nevertheless match the observed value.
Next, we perform an extensive scan over parameter space covering the following ranges: • 10 GeV < mB < 10 8 GeV.
The distribution of points compatible with all constraints (sufficient amount of baryon asymmetry, theB decay before BBN, and washout processes inactive at the time ofB decay) in the mB − plane is shown in figure 4. We see that the desired amount of baryon asymmetry can be produced over a wide range of parameter space, spanning several orders of magnitude in mB and . The three major constraints that bound this region are: the requirement that theB decays before BBN (towards the bottom left of the figure), the requirement of sufficientB production (bottom right), and the absence of strong washout effects (top left). The smallest possible value of in the compatible region of parameter space is ≈ 10 −8 , for mB ≈ 100 GeV. The desired baryon asymmetry can be generated for mB as small as 10 GeV (the observed patterns suggest that viable regions can be found even for GeV scaleB ; however, for such low masses the details of the exact decay channel and phase space suppression might become relevant), extending to extremely heavy masses ( > 10 8 GeV).
The color coding represents various values of m 0 /mB (see caption), and suggests a correlation between large m 0 /mB and large . This correlation is enforced by the need to keep various interaction rates sufficiently rapid to maintain sufficient production of the asymmetry and/or avoid constraints (decay before BBN, avoid washout of asymmetry). Various other parameter combinations are similarly constrained from their effect on CP (see eq. (3.6)). For instance, from our scans we find that λ < 10 −3 cannot produce the desired amount of baryon asymmetry; this can be understood from noting that λ < 10 −3 forces CP ∼ < 10 −9 (see eq. (3.6)), making it impossible to obtain a sufficiently large asymmetry since YB ∼

Extended hidden sector
We now study how the parameter space changes with the extended hidden sector, in particular, whether the baryogensis mechanism discussed above is compatible with the existence of a WIMP dark matter candidate charged under U(1) . For simplicity, we focus on regions of parameter space where the two sectors are in equilibrium (however, as seen in the previous subsection, even when this does not occur, a sufficient amount of baryon asymmetry can still be produced via freeze-in production of theB ), and work with the mass hierarchy mB > m X > m Z . In this framework, there are two potential concerns: • TheB B → XX process, which proceeds with gauge coupling g D strength, can keep theB in equilibrium even at temperatures below mB . This significantly suppresses theB freeze-out abundance YB , which in turn can suppress the baryon asymmetry.
• If theB has a long enough lifetime that it dominates the energy density of the Universe, the subsequent entropy dilution from its decays can suppress the relic density of dark matter, spoiling the attractive success of the "WIMP miracle".
We study these effects with a numerical scan. Figure 5 shows the various regions of parameter space in the mB − plane, with the other parameters fixed as specified in the caption. Here, for concreteness we have assumed mX = m 0 , i.e. the same sfermion masses in both the hidden and visible sectors. This plot shows that both of the above concerns are realized in different parts of parameter space. In the red region, the suppressedB abundance from freeze-out with the dark matter particle makes it impossible to produce JHEP06(2019)096 the desired amount of baryon asymmetry. In the opposite extreme (blue region), thẽ B dominates the energy density of the Universe at the time of its decay, diluting the abundance of the baryon asymmetry as well as of dark matter. We show this region in two shades: in the light blue region, the abundance is diluted only by an O(1) factor, so that the correct dark matter abundance can still be realized via the WIMP miracle; in the dark blue region, abundances are diluted by over an order of magnitude, and the connection with the WIMP miracle is more tenuous. Nevertheless, we see that the baryogenesis mechanism studied in this paper is compatible with hidden sector WIMP dark matter in large regions of parameter space (green region). The shapes of these regions can be understood by noting that, with m X and m 0 /mB fixed, YB is proportional to mB , see eq. (3.12).

Complementary phenomenology
In this section, we briefly discuss some other phenomenological signatures that might be possible within the baryogenesis framework discussed in this paper. As we saw in the previous section, the desired amount of baryon asymmetry can be produced over a large range of mass scales. While experimental evidence will prove difficult for heavy scales, additional phenomenology is possible on several fronts if the superpartners are light, as we now describe.
Collider and direct signatures. There are stringent constraints on squarks from the LHC even with RPV [34][35][36][37][38], where resonant production provides additional search modes [39]. Under various assumptions, such constraints currently lie at the O(100) GeV-1 TeV scale. In our studies in section 4, we restricted m 0 to lie above 1 TeV in light of such constraints, but baryogenesis can be realized with lower masses. In our framework, TeV scale squarks (assuming a large enough Higgs boson mass can be generated) can therefore be accessible to collider searches, e.g. forq → qχ 0 → q(qqq). This is in contrast with the MSSM implementation of baryogenesis in [11,12], where the sfermions are required to be several orders of magnitude heavier in order to satisfy the out of equilibrium condition and are therefore beyond the reach of colliders. Likewise, displaced RPV decays of light binos produced at the LHC could also be observable [40] at detectors such as CODEX-b [41], FASER [42], and MATHUSLA [43].
Likewise, direct searches also constrain the existence of a kinetically mixed light Z . Recall that the gauge boson kinetic mixing angle η is in principle different from the gaugino mixing angle , but one generally expects η ∼ . Current limits (see figure 6 in [44]) are at the level of η ∼ 10 −3 for GeV scale Z (deteriorating to η ∼ 10 −2 above 100 GeV), but much stronger below the GeV scale due to beam dump and SN1987A limits. In our framework, such probes are therefore relevant only if the Z is extremely light (below GeV scale), or if ∼ 10 −3 .
Low energy signatures. The presence of light superpartners with RPV couplings can give rise to several low energy signatures. These depend on the flavor structure of the RPV coupling λ ijk , i.e., which quarks are involved in the RPV interactions. We briefly discuss the most interesting signatures relevant to our framework here, and refer the interested JHEP06(2019)096 reader to [8] for a more comprehensive discussion of various low energy constraints on RPV supersymmetry.
If the coupling involves first generation quarks, a particularly strong test of baryon number violation comes from searches for neutron-antineutron oscillations n −n. 7 In our scenario, since the coupling of theB to SM is suppressed by , the n −n oscillation is primarily mediated by the MSSM binoB, via effectiveBudd four-fermion couplings obtained by integrating out the squark. Note that the RPV coupling λ 111 = 0, hence generating this effective vertex requires a flavor off-diagonal mixing term in the squark sector, leading to a suppression of the signal [8]. Current n −n constraints [48,49] can be translated to the following bound on this off-diagonal mixing angle: Upcoming experiments [50,51] will improve on this sensitivity by more than an order of magnitude and could probe our framework in the presence of such flavor off-diagonal mixing.
There are other processes that do not suffer from this unknown mixing suppression, such as double nucleon decays (such as p p → K + K + or n n → K 0 K 0 ). Such processes most strongly constrain λ 112 , λ 113 ( ∼ < 10 −4 , 10 −1 respectively for TeV scale superpartners), but involve large uncertainties from hadronic and nuclear matrix elements, see [8,52] for details. EDM measurements are a powerful probe of CP violation and are known to constrain even PeV scale sfermions [14,15]. In our framework, the EDM contribution arises through a gaugino-sfermion loop diagram. However, since the CP phase relevant to baryogenesis resides in the combination of the two gaugino masses, Arg(mB m * B ), both gaugino mass insertions are required on the gaugino propagator in the loop to generate the EDMs. This suppresses the amplitude by 2 , greatly suppressing the EDMs, which can be estimated (for down-type fermions) as: (5.2) Future experimental limits are expected to reach d e,p,n ∼ 10 −29 e cm [53], with electron, proton, and neutron EDMs expected to provide comparable reach if all sfermions are at the same scale. We see that the 2 contribution can suppress the EDMs below observable limits even for TeV scale sfermions and gauginos, 8 making them likely unobservable even with improved future measurements unless ∼ 10 −3 .
Dark matter signatures. If the hidden sector also contains a WIMP dark matter particle, as in the extended hidden sector scenario we considered, additional signatures are possible at both direct and indirect detection experiments.

JHEP06(2019)096
The Dirac nature of the dark matter particle X gives rise to a spin-independent direct detection cross section with nuclei mediated by Z, Z gauge bosons; this cross section scales approximately as (see [44,54] for details) This could provide an observable signal if ∼ < 10 −3 and m Z ∼ m Z . (See e.g. the projected sensitivity in figure 6 of [44].) Indirect detection signals, on the other hand, hold more promise as they are not suppressed by the small parameter. The dark matter annihilation process XX → Z Z proceeds with weak scale cross sections, and the Z subsequently decay into SM fermions, yielding observable signals at indirect detection instruments. Such cascade decays of hidden sector dark matter along with prospects/constraints from various indirect detection searches, such as with CMB measurements, from dwarf galaxies, at AMS-02, or at Cherenkov telescopes, have been studied extensively in [55] (see also [44,[56][57][58]). These studies suggest that such measurements can probe TeV scale hidden sector WIMP dark matter.

Summary
In this paper, we studied baryogenesis via a gaugino portal, with out-of-equilibrium decays of a hidden sector gauginoB via its portal mixing with the MSSM bino producing the baryon asymmetry of the Universe. We summarize our main findings below: • This mechanism can produce the desired baryon asymmetry in a minimal hidden sector framework, where all relevant phenomenology is controlled by the portal coupling of theB with the visible sector, as well as within an extended hidden sector that contains additional particles interacting with theB . In particular, the baryogenesis mechanism can coexist with hidden sector WIMP dark matter, which is an attractive possibility given the lack of a dark matter candidate in the visible sector due to R-parity violation.
• The desired baryon asymmetry can be produced across wide ranges of values of mB and , spanning several orders of magnitude. In particular, the mechanism is consistent with mB as low as 10 GeV. Such a low scale baryogenesis mechanism is an attractive prospect due to the possibility of additional phenomenological signatures.
• The small portal coupling is crucial in two respects. It makes theB sufficiently longlived to evade washout of the baryon asymmetry from inverse decay and annihilation effects without suppressing CP , the fraction ofB decays that produce a baryon asymmetry. Furthermore, while CP violating low scale physics generally gives rise to too-large EDMs, in this framework the EDMs are also 2 suppressed, making a low scale baryogenesis implementation consistent with stringent EDM constraints.
It is attractive that baryogenesis from long-lived particle decays, previously studied in other contexts, may be naturally implemented in a framework involving supersymmetry and a hidden sector, both well-motivated for other reasons.

JHEP06(2019)096
Our study contains several aspects worth pursuing in greater detail: for instance, it will be interesting to explore other hidden sector dark matter candidates that are compatible with the baryogenesis scenario discussed here.
Our work also contains several possible experimental signatures that might motivate additional study, ranging from colliders, direct searches, low energy probes, to indirect dark matter searches. The detailed nature of these signals, complementary to the parameter space for baryogenesis, depend on several additional underlying details of the model, as discussed in section 5, but can broadly be separated into two categories. There are those whose observability requires ∼ > 10 −3 (close to the upper limit we have studied); these includes signals such as direct Z production, EDMs, and direct detection of dark matter. However, there are other experimental signatures that are independent of the portal coupling -squarks at the LHC, neutron antineutron oscillation, and indirect detection of dark matter -which can therefore be observable even with much smaller values of . In contrast to most baryogenesis scenarios, which operate at high energies and do not carry testable low energy signatures, our baryogenesis mechanism therefore contains rich phenomenology that is testable on several experimental fronts.