WorksPhysics and Astronomy
Steady states in asymmetric exclusion processes with reactions: Fluctuations and non-mean field behaviour
Abstract
We study an totally asymmetric simple exclusion process (TASEP) in an open geometry. We couple TASEP with simple reactions in the form of production (birth) and degradation (death) due to over crowding taking place in the bulk. We investigate the steady states, which are generically space-dependent. Phase coexistences are also possible for some choices of the model parameters. We study the system by mean-field theories (MFT) supplemented by extensive Monte-Carlo simulations (MCS). Intriguingly, unlike oridinary TASEP and other related models, configuration-dependent bulk reactions in a TASEP model can make nearest-neighbour correlations quantitatively important even in an open driven system where mean-field theory would ordinarily be expected to provide a useful description. We attribute this to the correlation effects ignored in MFT. We numerically quantify this correlation effect.
Introduction
Driven diffusive systems [1, 2, 3, 4] exhibit remarkably complex behavior despite being governed by purely local dynamics, even in one-dimensional settings. A prototypical example is the totally asymmetric simple exclusion process (TASEP) [5, 6, 7, 8], which is defined on a one-dimensional lattice with open boundaries. The dynamics are stochastic and obey the exclusion principle, allowing at most one particle to occupy a site at any given time. Particles are injected into the lattice at one boundary with a specified rate, move unidirectionally from site to site with unit hopping rate, and exit through the opposite boundary at another prescribed rate. Depending on the particle injection and extraction rates, the TASEP exhibits three distinct steady-state phases: a low-density (LD) phase, characterized by a bulk density below one-half; a high-density (HD) phase, where the bulk density exceeds one-half; and a maximal-current (MC) phase, in which the bulk density is exactly one-half.
TASEP was originally conceived as a minimal model to describe proteins synthesis in biological cells [9]. Subsequently it was found to display boundary-induced phase transitions [5, 6, 7]. Many extensions of the TASEP have been developed to better model real-world transport processes. These include studies on transport through network-like structures [10, 11, 12] inspired by molecular motor movement in cells [13], microtubule length regulation by kinesin-8 motors [14, 15], and two-lane TASEP systems with particle switching that can produce phenomena such as jams and domain-wall delocalization [16, 17, 18]. Researchers have also investigated the effects of obstacles and bottlenecks - both static and dynamic - which can hinder particle movement, similar to traffic congestion or protein-induced hindrance in cellular transport; see Refs. [19, 20, 21, 22, 23]. Such spatial variations in hopping rates can lead to nonuniform steady-state density profiles [24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 35, 36, 37, 38, 39, 40, 41, 42]. Effects of multiple lanes [43, 44, 45], weak local nonconserving dynamics [46, 47, 48, 49] and finite resources [50, 51, 52, 53, 54, 55, 56, 57, 58, 59] are also known to give rise to novel steady states.
Reaction–diffusion models [60, 61, 62, 63] describe systems with particles that undergo diffusion and local reactions such as creation, annihilation, coagulation, or branching. These models provide a simple framework for studying nonequilibrium phenomena, including phase transitions, pattern formation, and absorbing-state dynamics. Due to their analytical tractability and rich collective behavior, they have been extensively used in statistical physics, population dynamics, chemical kinetics, and biological transport processes.
In this article, we study one-dimensional (1D) local reactions with open boundaries in the presence of a drift or unidirectional motion. To that end, we combine the ideas of an open TASEP together with the simplest one-component reaction model. Our studies can serve as a conceptual model for population dynamics in the presence of in and out migration and chemical reactions in a moving environment. We investigate the steady states of the model. We find that the stationary densities in general have complex dependence on space (i.e., the lattice site). We mainly focus on the special case of equal birth and death rates for reasons of simplicity. For some choices of the model parameters, the model admits phase-coexistence. We use mean-field theory (MFT) to analyse the system, supplemented by extensive Monte-Carlo simulation (MCS) studies. Unexpectedly, the MFT predictions on the stationary densities have a rather poor agreement with the corresponding MCS studies results. We attribute this discrepancy to the correlation effects neglected in MFT. We use our MCS studies to estimate the correlation effects, which are finite in the present model. Quite interestingly, the mean-field phase diagrams with equal birth and death rates for the present model have strong similarities with that in the TASEP with Langmuir kinetics (TASEP-Lk) model with equal attachment-detachment rates, even though the microscopic origins of the two models are very different. Notwithstanding this unexpected similarity, we show that fluctuations have a much stronger impact on the present model than the TASEP-Lk model, making MFT less accurate here. The rest of this article is organised as follows. In Section II, we construct the model that we use. Section III is devoted to studying the steady states. Next, in Section III A, we set the MFT for this model. Then in Section III B, we solve the MFT equations to obtain the stationary densities. In Section III C, we solve the MFT equations for equal birth and death rates. Next, in Section III D, we obtain the phase diagram of the model in the plane of the control parameters. In Section IV, we compare the MFT predictions on the stationary densities with the corresponding MCS results. Then in Section V, we analyse our conjecture that correlation effects are responsible for the disagreement between MFT and MCS results, and prodice numerical support to it. We conclude and summarise in Section VI.
Construction of the model
In this work, we consider a 1D lattice of sites, which are labelled by , ; see Figure 1.
Figure 1. Schematic model diagram. Various microscopic dynamical processes are shown.
A site can have occupation (filled) or (empty). The dynamical update rules of this model consist of the following steps:
(a) Hopping: Particles may enter the TASEP lane at the left end () at rate , stochastically hop from to unidirectionally subject to exclusion at rate unity, and leave the lane at the right end () at rate [41].
(b) Birth: If a site is occupied and either site or is vacant, then a particle is created at site or at rate .
(c) Death: If sites and or and are occupied, then they annihilate each other giving two empty sites at rate . This is death due to over population.
The above dynamical rules (a)-(d) define our model.
Steady states
In this Section, we first construct the MFT for the model defined above, and use that to calculate the stationary densities.
Construction of the MFT
We now construct the MFT equation of motion following the principles of MFT in a pure, open TASEP [64]. MFT entails ignoring spatial correlations. The density at site follows the MFT equation
In (1), obviously the birth and death terms do not conserve numbers. Furthermore, the birth and death terms violate the particle-hole symmetry of an open TASEP [41]. In the thermodynamic limit, we use a quasi-continuous spatial coordinate with . Next, to ensure competition between the hopping dynamics of TASEP and the reaction processes (birth and death), we set
where . This is similar to the rescaling of the attachment-detachment rates in TASEP with Langmuir kinetics (Lk) problem; see Refs. [43, 46, 47, 48, 49]. With this and in the steady state setting , we get
where . (3) is the steady state MFT equation, which can be solved subject to suitable boundary conditions. It allows us to extract an MFT expression for the steady state current given by
where any boundary layers are not considered in the thermodynamic limit.
Stationary densities
We now solve (3) to calculate the stationary densities, subject to the usual boundary conditions and . We note from (3) that , a constant, is a solution to (3), provided
giving
assuming . This solution can be more or less or equal to and clear does not satisfy any of the boundary conditions at . The latter property is reminiscent of the MC phase in an open TASEP, and hence is a generalised MC phase; see also Refs. [46, 47, 48, 49] in the context of the TASEP-Lk model. However, this is not the only solution to . The general solution for may be written as
where is as given by (6). Here, is yet to be determined, -dependent function, giving an -dependent . By substituting (7), we get the equation for :
where . We now solve (8), which can be recast as
Integrating, we obtain
where is a constant of integration. Simplifying,
where is also a constant of integration. Equation (11) gives an implicit solution of . Constant can be evaluated from the boundary conditions on at . There are two possible cases:
(i) Injection limited solution: This is the analogue of the LD phase in a pure, open TASEP [64, 65]. Use the boundary condition . This gives after using (7)
where
(ii) Extraction limited solution: This is the analogue of the HD phase in a pure, open TASEP [64, 65]. Use the boundary condition . Once again, using (7)
where
Therefore, we find that there are three basic stationary solutions for , which are , and , the last one being a constant. For reasons of convenience in analytical manipulations and explicit calculations, we consider the special case with , giving . This allows us to obtain explicit analytical solution for from the MFT. We then compare that with MCS studies results. These are presented below.
In this case, and . From (11), we then obtain
giving
giving an explicit solution of as a function of . Here, is a constant. We then get
in addition to (i.e., ) that is identical to the MC phase solution of an isolated open TASEP. We now use the boundary conditions on to calculate . At , , giving . Thus
This is a growing solution in . This in turn means for this solution in general. If , in the range , reminiscent of the LD phase in a pure open TASEP [64, 65]. Given that is controlled by the “left” boundary condition , we are thus tempted to identify , a generalised LD phase stationary density profile. There are, however, two major differences vis-à-vis the usual LD phase density in an open TASEP, viz. (i) it is space dependent here, whereas in a pure open TASEP, it is constant in the bulk; (ii) as a consequence of the space-dependence of , the LD phase may span only a part of the TASEP lane, depending upon ; see also below.
Next, at , , giving . Thus
This is also a growing solution in . This in turn means that for this solution in general. In fact, if , i.e., , in the range , which reminds us of the HD phase in pure open TASEP. Given that is controlled by the “right” boundary condition , we are thus tempted to identify , a generalised HD phase stationary density profile. Similar to the LD phase, there are, however, two major differences vis-à-vis the usual HD phase density in an open TASEP, viz. (i) it is space dependent here, whereas in a pure open TASEP, it is constant in the bulk; (ii) as a consequence of the space-dependence of , the HD phase may span only a part of the TASEP lane, depending upon ; see also below. Notice the similarity between the three basic solutions here and the three basic solutions of the TASEP-Lk problem with equal attachment-detachment rate [46, 47]. Similar to the TASEP-Lk problem, the space-dependent nature of and , this model allow for mixed phases, leading to phase coexistence. However, the -dependence of and is nonlinear in in the present model, unlike in the TASEP-Lk model, where such dependence is linear with equal attachment-detachment limit.
Phase diagram
With the three basic solutions , , , we can now construct the MFT phase diagram in the plane and mark the various phases. The phase boundaries are determined by equality of the currents in the different phases. This holds because, although the birth-death processes make the model nonconserved, the probability of occurrence of birth and death at a given site is vanishingly small in the limit of . This ensures that the steady state currents are locally conserved; see also Refs. [47]. As mentioned above, , being a rising function, may exceed at , which should mark the beginning of the MC phase. Equating the currents in the LD phase and MC phase at , we get
such that the system is in the LD phase for and is in the MC phase between , giving the LD-MC phase. Clearly, marks the changeover from the LD to LC-MC phase. Setting in (21) above, we get
as the phase boundary between the LD and LD-MC phase in the plane. On the other hand, sets the condition for the changeover from LD-MC to MC phase. Setting in (21), we get
which is identical to the LD-MC phase boundary in the plane in a pure, open TASEP. Here, it is however the phase boundary between a distinct set of phases, viz., LD, pure phase, and LD-MC, a mixed phase.
Next, we consider the fact that the solution can be less than at some . To proceed quantitatively, let us assume that at , . This gives
which gives
The boundary between the MC and HD-MC phase is determined by , giving
which implies
Furthermore, mixed phase HD-MC to pure phase HD transition is given by the condition , which gives
giving
Although this condition is identical to the one at the HD-MC transition in a pure, open TASEP, here is refers to a different transition, between a pure HD phase and a mixed HD-MC phase.
We have thus found that the model admits several mixed phases, LD-MC and HD-MC. Similar to the TASEP-Lk problem, it is possible to have three-phase coexistence if
or
If the condition (31) is satisfied, between for , in the range and for , giving a three-phase coexistence.
Three phase coexistence naturally requires . For , one instead has a two-phase coexistence - LD-HD phase, which is a localised domain wall (LDW). Since the nonconserving reaction terms are locally vanishingly small in the limit , the TASEP hopping current should be locally conserved. This is similar to Ref. [43, 47]. This gives
where is the position of the LDW. We then find
See Figure 2 for the MFT phase diagrams in the plane.
Figure 2. MFT phase diagrams in the plane in (a)-(e) for respectively.
Notice that with increasing the mixed phases occupy larger areas in the phase diagram. These mixed phases do not exist in the pure TASEP limit with . These features can be explained from the MFT equation (3), which in the limit becomes
The rhs of (34), the “reaction source”, is positive for , negative for and vanishes when . Therefore, at the level of this simple MFT, the reaction processes tend to drive the density towards : In the LD phase, the reactions tend to increase the density towards , whereas in the HD phase, reactions tend to drive down the density towards . This suggests that the uniform LD or HD phase in a pure TASEP can get modified to produce the mixed phases. Since the reaction term in (34) scales with , the extent of the mixed phases increase with . Since the MC phase density is unaffected by the reaction term in (34), the MC phase in the MFT phase diagram should remain unaffected by the reaction. This is consistent with the phase diagrams in Figure 2.
Just as the basic MFT solutions for the stationary solutions in the present problem have a degree of correspondence with those in the TASEP-Lk problem, the phase diagrams in two problems are also similar. Nonetheless, there are significant difference at the level of the microscopic dynamics. In the TASEP-Lk problem, the attachment-detachment processes are purely local, being unaffected by the occupancy of the neighbouring sites. In the present model, reactions processes involve neighbouring sites, which develop correlations, which in turn ultimately make the MFT predictions not agreeing with the MCS results; see below.
Stationary Densities - MFT versus MCS
In this Section, we show the MFT stationary density profiles in the different phases, and compare them with the corresponding MCS results; see the representative plots for the stationary densities in the LD and HD phases in Figure 3. We connect the discrepancy between the MFT and MCS results to the correlation effects.
Figure 3. Stationary densities in the LD and HD phases obtained from MFT (blue line) and MCS (red points). MFT predictions for both the LD (top) and HD (bottom) phase stationary densities are consistently above the MCS results. See text.
We find that there is a significant disagreement between the MFT predictions on the stationary densities and the corresponding MCS results; see Figure 3. This is rather unexpected, because in an open TASEP-like model there is no global number conservation. As a result, the density fluctuations are short-ranged, which in turn should make MFT reasonably accurate. This expectation usually holds good in different open TASEP or its variants, including TASEP-Lk [46]. Significant deviations are usually found only in some models with a strict number conservation, e.g., [48, 50, 51]. The unexpected disagreements between MFT and MCS results in the present model require careful considerations, which we discuss below.
We now argue that the correlation effects, neglected in the derivation of the MFT Eq. (1), are responsible for this disagreement between the MFT and MCS results. To investigate the correlation effects, we first study the system in a limit in which the dynamics is overwhelmingly dominated by the reaction part. This is achieved by removing the -scaling that was introduced to the birth and death rates to ensure competition between the reaction process and the TASEP hopping dynamics. If our argument explaining the lack of accuracy of MFT is indeed correct, then that should affect even in this extreme case of the reaction-dominated dynamics. In this limit, neglecting the TASEP hopping dynamics, which is now irrelevant in the bulk, the MFT predicts that the bulk density should be
To verify the result (35), we study the model with MCS, after removing the -scaling, as discussed above. The MCS results are shown in Figure 4. We made the following observations from Figure 4. First of all, the bulk
Figure 4. Stationary densities in the limit in which the reactions dominate over the TASEP hopping for large systems for various in the MCS studies. We find that . See text.
density remains unchanged if both and are enhanced or decreased by the same factor. For instance in Figure 4, the bulk density is same for , and , implying that if change keeping the ratio unchanged. However, the MCS density in the bulk , violating the MFT prediction. Thus the MFT predictions appear to overestimate the stationary densities, as we also find in Figure 3.
Why is MFT inaccurate here?
We now discuss why MFT is quantitatively inaccurate, in particular why it overestimates the stationary densities. The derivation of the MFT Eq. (1) involves replacing the death term by , where . This factorisation becomes questionable in 1D for the following reason. The birth process tends to create clusters of occupied sites along the TASEP lane. Once a cluster is formed, the birth process can take place only along the edges of the cluster. In contrast, the death process can take place inside the bulk of the cluster. This suppresses the birth events, effectively suppressing the birth rate, relative to the death rate. As a result, the MCS gives a lower steady state density in comparison with the MFT prediction. Equivalently, it is expected that , giving the true loss term , its mean field value. This is another way of seeing that the “true” steady state density should be lower than that predicted by MFT. We now make quantitative estimates of relative to from our MCS results in the stationary states in pure TASEP and in the present model. If MFT works well, we expect , else .
The actual reaction term on the rhs of (34) is
whereas the MFT reaction term is
Clearly, if , the actual death rate is larger (smaller) than one appears in the MFT equation (34). If it is larger (smaller), then the actual stationary density should be smaller (larger) than that predicted by MFT. We check this numerically. We calculate in the different phases and also in pure TASEP (without any reaction). If the factorisation works, then in our MCS studies, else .
Across all the figures Figure 5-Figure 10, our MCS studies reveal that in general is space-dependent and larger than 1 in all the phases admitted by our model. This is in contrast to the pure TASEP limit, for which always. This shows that the effective decay rate is space-dependent and is bigger than its mean-field value. This in turn explains why MFT over-estimates the stationary densities and is always larger than the corresponding MCS result. Indeed, in the regions where is bigger, the disagreement between MFT and MCS results are also bigger. This establishes that neglecting the correlation effects lies at the heart of the quantitative inaccuracy of MFT.
Figure 5. (top) Stationary density for , , from MFT (blue line) and MCS (red points). MFT density is higher than MCS density. (bottom) Plot of versus for pure TASEP (red points) and TASEP with reaction (blue points). For the latter, and decreases with . This suggests an effective death rate larger than that decreases with , which is consistent with the top figure.
Figure 10. (top) Stationary density for , , from MFT (blue line) and MCS (red points). MFT density is higher than MCS density. (bottom) Plot of versus for pure TASEP (red points) and TASEP with reaction (blue points). For the latter, and grows with . This suggests an effective death rate larger than that grows with , which is consistent with the top figure.
Figure 6. (top) Stationary density for , , from MFT (blue line) and MCS (red points). MFT density is higher than MCS density. (bottom) Plot of versus for pure TASEP (red points) and TASEP with reaction (blue points). For the latter, and has a peak towards the exit end (). This suggests an effective death rate larger than that has a peak towards the exit end (), which is consistent with the top figure.
Our representative MCS studies on the stationary densities reveal LD, HD and LD-HD two phase coexistences, i.e., domain walls. Our limited studies have not detected the MC phase or mixed phase densities involving the MC phase. Detection of these phases require additional extensive parameter scans, which are beyond the scope of the present study.
Figure 7. (top) Stationary density for , , from MFT (blue line) and MCS (red points). MFT density is higher than MCS density. (bottom) Plot of versus for pure TASEP (red points) and TASEP with reaction (blue points). For the latter, and has a peak towards the exit end (). This suggests an effective death rate larger than that has a peak towards the exit end (), which is consistent with the top figure.
Figure 8. (top) Stationary density for , , from MFT (blue line) and MCS (red points). MFT density is higher than MCS density. (bottom) Plot of versus for pure TASEP (red points) and TASEP with reaction (blue points). For the latter, and grows with . This suggests an effective death rate larger than that grows with , which is consistent with the top figure.
Summary and outlook
In this work, we have studied the steady states of an open totally asymmetric simple exclusion process (TASEP) coupled to bulk birth and death processes. The reactions are configuration-dependent: particle production requires an occupied site adjacent to a vacancy, whereas particle annihilation occurs between neighbouring occupied sites. This coupling of directed transport, exclusion and local reactions provides a minimal setting in which to investigate the interplay between nonequilibrium transport and density-dependent particle production and annihilation.
We have constructed a mean-field theory (MFT) for this model and analysed its stationary solutions in the hydrodynamic limit. For equal birth and death rates, the theory admits spatially varying density profiles in addition to a uniform solution with density . The resulting mean-field phase diagram contains low-density and high-density regimes, a maximal-current regime, and mixed phases in which distinct stationary density profiles occupy different portions of the system. The phase structure resembles that of TASEP with Langmuir kinetics, despite the different microscopic origins of the particle-nonconserving processes. In the present model, the spatial variation of the stationary density is nonlinear in the continuum coordinate and is determined by the coupled transport-reaction balance.
Our Monte Carlo simulations reveal a substantial quantitative discrepancy between the mean-field stationary density profiles and those obtained from the stochastic dynamics. In the parameter regimes examined, MFT generally predicts higher densities than the simulations.
Figure 9. (top) Stationary density for , , from MFT (blue line) and MCS (red points). MFT density is higher than MCS density. (bottom) Plot of versus for pure TASEP (red points) and TASEP with reaction (blue points). For the latter, and has a large peak for near 1. This suggests an effective death rate larger than that has a large value for near 1, which is consistent with the top figure.
This discrepancy is noteworthy because mean-field descriptions often provide useful quantitative predictions for open driven systems with short-ranged density correlations. The present results indicate that the microscopic structure of the reaction processes can be important even when the corresponding mean-field phase structure appears conventional.
We have investigated the origin of this discrepancy by examining nearest-neighbour occupation correlations. The exact evolution equation for the mean density depends on the probability of finding two neighbouring occupied sites, whereas MFT replaces this pair probability by the product of the corresponding one-site densities. This factorisation neglects correlations generated by the reaction dynamics. In particular, the birth process requires occupied-vacant pairs, while the death process acts on occupied-occupied pairs.
Consequently, correlations between neighbouring sites modify the relative frequencies of production and annihilation events and can shift the stationary density away from the mean-field prediction.
Our numerical measurements of the nearest-neighbour pair probability provide evidence that this correlation effect is important in the present model. The results are consistent with the physical picture that the formation of occupied clusters suppresses particle production relative to the factorised estimate and enhances the contribution of pair annihilation. These effects can qualitatively account for the observed tendency of MFT to overestimate the stationary density. A complete quantitative description, however, requires the correlation corrections to be incorporated into the transport-reaction balance, including their spatial dependence.
The comparison with TASEP with Langmuir kinetics highlights an important distinction between microscopic reaction mechanisms and their mean-field representations. Models with similar mean-field equations can exhibit different quantitative behaviour when their reactions depend differently on local particle configurations. In the present model, the correlations generated by the reactions can substantially affect the predictions of the mean-field theory. This observation motivates a broader investigation of the conditions under which mean-field descriptions remain reliable in driven systems with particle-nonconserving dynamics.
Several questions remain open, which can form potential future research directions. First, it would be useful to explore the finite-size scaling of the discrepancy between MFT and Monte Carlo results and to determine whether it persists in the thermodynamic limit. Second, a systematic study of the connected nearest-neighbour correlation function across the different stationary regimes could clarify the spatial structure of the corrections to correlations and their relation to the density-profile errors. Third, extending the mean-field theory to include pair correlations, or employing a suitable cluster approximation, may improve the quantitative predictions. Such approaches would also help to determine whether the observed discrepancy can be understood predominantly through nearest-neighbour correlations or whether considering higher-order correlations are necessary. For reasons of simplicity, our present study is limited to equal birth and death rates. How unequal birth and death rates may modify our results would be an important question to answer in the future. It would also be interesting to investigate whether the competition between directed transport and configuration-dependent reactions produces additional regimes when the birth and death rates are unequal. Lastly, developing a systematic theory of the correlation effects, and establishing their consequences for the stationary phases, remain important directions for future work.
Acknowledgement:- The authors thank Prof. Abhik Basu for his valuable discussions. SM thanks ANRF (India) for partial financial support through the ARG (MATRICS) programme (file no.: ANRF/ARGM/2025/000748/TS).
References
- [1]S. Katz, J. L. Lebowitz, and H. Spohn, Phase transitions in stationary nonequilibrium states of model lattice systems, Phys. Rev. B 28, 1655 (1983).DOI
- [2]S. Katz, J. L. Lebowitz, and H. Spohn, Nonequilibrium steady states of stochastic lattice gas models of fast ionic conductors, J. Stat. Phys. 34 497 (1984).
- [3]B. Schmittmann and R. K.-P. Zia, in Phase Transitions and Critical Phenomena, edited by J. L. Lebowitz and C. Domb, Vol. 17 (Academic Press, London, 1995).
- [4]Nonequilibrium Statistical Mechanics in One Dimension, edited by V. Privman (Cambridge University Press, Cambridge, 1997).DOI
- [5]J. Krug, Phys. Rev. Lett. 67, 1882 (1991).
- [6]B. Derrida, S. A. Janowsky, J. L. Lebowitz, and E. R. Speer, Exact solution of the totally asymmetric simple exclusion process: Shock profiles, J. Stat. Phys. 73, 813 (1993).DOI
- [7]B. Derrida and M. R Evans, Exact correlation functions in an asymmetric exclusion model with open boundaries, J. Phys. I 3, 311 (1993).DOI
- [8]B. Derrida, M. R. Evans, V. Hakim, and V. Pasquier, Exact solution of a 1D asymmetric exclusion model using a matrix formulation, J. Phys. A: Math. Gen. 26, 1493 (1993).DOI
- [9]C. T. MacDonald, J. H. Gibbs, and A. C. Pipkin, “Kinetics of biopolymerization on nucleic acid templates”, Biopolymers 6, 1 (1968).pubmed.ncbi.nlm.nih.gov/5641411
- [10]B. Embley, A. Parmeggiani, and N. Kern, “Understanding totally asymmetric simple exclusion process transport on networks: Generic analysis via effective rates and explicit vertices,” Phys. Rev. E 80, 041128 (2009).pubmed.ncbi.nlm.nih.gov/19905294
- [11]R. Chatterjee, A. K. Chandra, and A. Basu, Phase transition and phase coexistence in coupled rings with driven exclusion processes, Phys. Rev. E 87, 032157 (2013).DOI
- [12]R. Chatterjee, A. K. Chandra, and A. Basu, Asymmetric exclusion processes on a closed network with bottlenecks, J. Stat. Mech.: Theory Exp. 2015, P01012.
- [13]J. Howard and R. L. Clark, “Mechanics of motor proteins and the cytoskeleton,” Appl. Mech. Rev. 55, B39–B49 (2002).DOI
- [14]A. Melbinger, L. Reese, and E. Frey, Microtubule length regulation by molecular motors, Phys. Rev. Lett. 108, 258104 (2012).DOI
- [15]L. Reese, A. Melbinger, and E. Frey, Crowding of molecular motors determines microtubule depolymerization, Biophys. J. 101, 2190–2200 (2011).
- [16]T. Reichenbach, T. Franosch, and E. Frey, Exclusion processes with internal states, Phys. Rev. Lett. 97, 050603 (2006).DOI
- [17]T. Reichenbach, E. Frey, and T. Franosch, Traffic jams induced by rare switching events in two-lane transport, New J. Phys. 9, 159 (2007).DOI
- [18]T. Reichenbach, T. Franosch, and E. Frey, Domain wall delocalization, dynamics and fluctuations in an exclusion process with two internal states, Eur. Phys. J. E 27, 47 (2008).DOI
- [19]D. Helbing, Traffic and related self-driven many-particle systems, Rev. Mod. Phys. 73, 1067–1141 (2001).DOI
- [20]B. S. Kerner and H. Rehborn, Experimental features and characteristics of traffic jams, Phys. Rev. E 53, R4275–R4278 (1998).
- [21]M. Treiber and A. Kesting, Traffic Flow Dynamics: Data, Models and Simulation, Springer, Berlin (2013).
- [22]R. Dixit, J. L. Ross, Y. E. Goldman, and E. L. F. Holzbaur, Differential regulation of dynein and kinesin motor proteins by tau, Science 319, 1086–1089 (2008), doi:10.1126/science.1152993.DOI
- [23]I. A. Telley, P. Bieling, and T. Surrey, Obstacles on the microtubule reduce the processivity of kinesin-1 in a minimal in vitro system and in cell extract, Biophys. J. 96, 3341–3353 (2009), doi:10.1016/j.bpj.2009.01.015.DOI
- [24]S. A. Janowsky and J. L. Lebowitz, Finite-size effects and shock fluctuations in the asymmetric simple-exclusion process, Phys. Rev. A 45, 618 (1992).pubmed.ncbi.nlm.nih.gov/9907026
- [25]N. Sarkar and A. A. Basu, Nonequilibrium steady states in asymmetric exclusion processes on a ring with bottlenecks, Phys. Rev. E 90, 022109 (2014).DOI
- [26]A. B. Kolomeisky, Asymmetric simple exclusion model with local inhomogeneity, J. Phys. A: Math. Gen. 31, 1153 (1998).DOI
- [27]K. Qiu, X. Yang, W. Zhang, D. Sun, and Y. Zhao, Density profiles in the totally asymmetric exclusion processes with both local inhomogeneity and Langmuir kinetics, Physica A 373, 1 (2007).DOI
- [28]A. Goswami, R. Chatterjee, S. Mukherjee, Defect versus defect: stationary states of single file marching in periodic landscapes with road blocks, Phys. Rev. E 110, 064149 (2024).DOI
- [29]S. Pal and A. Basu, Universality of shocks in conserved driven single-file motions with bottlenecks, Phys. Rev. E 113, L042102 (2026).
- [30]L. B. Shaw, A. B. Kolomeisky, and K. H. Lee, Local inhomogeneity in asymmetric simple exclusion processes with extended objects, J. Phys. A: Math. Gen. 37, 2105 (2004).DOI
- [31]G. Tripathy and M. Barma, Driven lattice gases with quenched disorder: Exact results and different macroscopic regimes, Phys. Rev. E 58, 1911 (1998).DOI
- [32]H. Hinsch and E. Frey, Bulk-driven nonequilibrium phase transitions in a mesoscopic ring, Phys. Rev. Lett. 97, 095701 (2006).pubmed.ncbi.nlm.nih.gov/17026375
- [33]T. Banerjee, N. Sarkar, and A. Basu, Generic nonequilibrium steady states in an exclusion process on an inhomogeneous ring, J. Stat. Mech. 2015, P01024.DOI
- [34]P. Roy, A. K. Chandra, and A. Basu, Pinned or moving: States of a single shock in a ring, Phys. Rev. E 102, 012105 (2020).DOI
- [35]T. Banerjee and A. Basu, Smooth or shock: Universality in closed inhomogeneous driven single-file motions, Phys. Rev. Research 2, 013025 (2020).DOI
- [36]P. Pierobon, M. Mobilia, R. Kouyos, and E. Frey, Bottleneck-induced transitions in a minimal model for intracellular transport, Phys. Rev. E 74, 031906 (2006).pubmed.ncbi.nlm.nih.gov/17025666
- [37]S. Klumpp and R. Lipowsky, Active diffusion of motor particles, Phys. Rev. Lett. 95, 268102 (2005).
- [38]B. Waclaw, J. Cholewa-Waclaw, and P. Greulich, Totally asymmetric exclusion process with site-wise dynamic disorder, J. Phys. A: Math. Theor. 52, 065002 (2019).DOI
- [39]A. Jindal, A. B. Kolomeisky, and A. K. Gupta, The role of dynamic defects in transport of interacting molecular motors, J. Stat. Mech. 2020, 043206.DOI
- [40]M. Sahoo, J. D. Dong, and S. Klumpp, Dynamic blockage in an exclusion process, J. Phys. A: Math. Theor. 48, 015007 (2014).
- [41]A. Goswami, M. Chatterjee, and S. Mukherjee, Steady states and phase transitions in heterogeneous asymmetric exclusion processes, J. Stat. Mech. (2022) 123209.DOI
- [42]S. Mukherjee and A. Basu, Nonuniform asymmetric exclusion process: Stationary densities and domain walls, arXiv: 2410.22516, accepted for publication in Phys. Rev. E.arxiv.org/abs/2410.22516
- [43]A. Goswami, U. Dey, S. Mukherjee, Nonequilibrium steady states in coupled asymmetric and symmetric exclusion processes, Phys. Rev. E 108, 054122 (2023).DOI
- [44]I. R. Graf and E. Frey, Generic Transport Mechanisms for Molecular Traffic in Cellular Protrusions, Phys. Rev. Lett. 118, 128101 (2017).DOI
- [45]M. Bojler, I. R. Graf, and E. Frey, Self-organized system-size oscillation of a stochastic lattice-gas model, Phys. Rev. E 98, 012410 (2018).DOI
- [46]A. Parmeggiani, T. Franosch, and E. Frey, Phase coexistence in driven one-dimensional transport, Phys. Rev. Lett. 90, 086601 (2003).
- [47]A. Parmeggiani, T. Franosch, and E. Frey, Totally asymmetric simple exclusion process with Langmuir kinetics, Phys. Rev. E 70, 046101 (2004).pubmed.ncbi.nlm.nih.gov/15600454
- [48]T. Banerjee, A. K. Chandra, and A. Basu, Phase coexistence and particle nonconservation in a closed asymmetric exclusion process with inhomogeneities, Phys. Rev. E 92, 022121 (2015).pubmed.ncbi.nlm.nih.gov/26382358
- [49]B. Daga, S. Mondal, A. K. Chandra, T. Banerjee, and A. Basu, Nonequilibrium steady states in a closed inhomogeneous asymmetric exclusion process with generic particle conservation, Phys. Rev. E 95, 012113 (2017).DOI
- [50]S. Pal, P. Roy and A. Basu, Availability, storage capacity, and diffusion: Stationary states of an asymmetric exclusion process connected to two reservoirs, Phys. Rev. E 110, 054104 (2024).arxiv.org/abs/2308.08384
- [51]S. Pal, P. Roy and A. Basu, Distributed fixed resources exchanging particles: Phases of an asymmetric exclusion process connected to two reservoirs, Phys. Rev. E 111, 034109 (2025).arxiv.org/abs/2403.05945
- [52]S. Pal, P. Roy and A. Basu, Stationary densities and delocalized domain walls in asymmetric exclusion processes competing for finite pools of resources, Phys. Rev. E 113, 034125 (2026).DOI
- [53]D. A. Adams, B. Schmittmann, and R. K. P. Zia, Far-from-equilibrium transport with constrained resources, J. Stat. Mech.: Theory Exp. 2008, P06009.DOI
- [54]L. Jonathan Cook, R. K. P. Zia, and B. Schmittmann, Competition between multiple totally asymmetric simple exclusion processes for a finite pool of resources, Phys. Rev. E 80, 031142 (2009).DOI
- [55]L. Jonathan Cook and R. K. P. Zia, Feedback and fluctuations in a totally asymmetric simple exclusion process with finite resources, J. Stat. Mech.: Theory Exp. 2009, P02012.DOI
- [56]A. Haldar, P. Roy, and A. Basu, Asymmetric exclusion processes with fixed resources: Reservoir crowding and steady states, Phys. Rev. E 104, 034106 (2021).DOI
- [57]A. Haldar, P. Roy, E. Frey and A. Basu, “Availability versus carrying capacity: Phases of asymmetric exclusion processes competing for finite pools of resources”, Phys. Rev. E 111, 014154 (2025).DOI
- [58]B. Pal and A. K. Gupta, Reservoir crowding in a resource-constrained exclusion process with a dynamic defect, Phys. Rev. E 106, 044130 (2022).DOI
- [59]S. Pal and A. Basu, Stationary densities in a weakly nonconserving asymmetric exclusion process with finite resources, Phys. Rev. E 114, 034121 (2026).DOI
- [60]G. M. Schütz, Exactly Solvable Models for Many-Body Systems Far from Equilibrium, in Phase Transitions and Critical Phenomena, Vol. 19 (Academic Press, 2001).
- [61]J. Marro and R. Dickman, Nonequilibrium Phase Transitions in Lattice Models (Cambridge University Press, 1999).
- [62]H. Hinrichsen, “Nonequilibrium Critical Phenomena and Phase Transitions into Absorbing States,” Advances in Physics 49, 815 (2000).DOI
- [63]U. C. Täuber, Critical Dynamics: A Field Theory Approach to Equilibrium and Non-Equilibrium Scaling Behavior (Cambridge University Press, 2014).
- [64]R. A. Blythe and M. R. Evans, Nonequilibrium steady states of matrix-product form: A solver’s guide, J. Phys. A: Math. Theor. 40, R333 (2007).DOI
- [65]A. B. Kolomeisky, G. M. Schütz, E. B. Kolomeisky and J. P. Straley, Phase diagram of one-dimensional driven lattice gases with open boundaries, J. Phys. A: Math. Gen. 31, 6911 (1998).DOI









