Highly symmetric D-brane-anti-D-brane effective actions

The entire S-matrix elements of four, five and six point functions of D-brane-anti D-brane system are explored. To deal with symmetries of string amplitudes as well as their all order α′ corrections we first address a four point function of one closed string Ramond-Ramond (RR) and two real tachyons on the world volume of brane-anti brane system. We then focus on symmetries of string theory as well as universal tachyon expansion to achieve both string and effective field theory of an RR and three tachyons where the complete algebraic analysis for the whole S-matrix \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ < {V}_{C^{-1}}{V}_{T^{-1}}{V}_{T^0}{V}_{T^0}> $$\end{document} was also revealed. Lastly, we employ all the conformal field theory techniques to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ < {V}_{C^{-1}}{V}_{T^{-1}}{V}_{T^0}{V}_{T^0}{V}_{T^0}> $$\end{document}, working out with symmetries of theory and find out the expansion for the amplitude to be able to precisely discover all order singularity structures of D-brane-anti-D-brane effective actions of string theory. Various remarks about the so called generalized Veneziano amplitude and new string couplings are elaborated as well.


JHEP09(2017)025
only are we able to discover new couplings based on direct computations but also one fixes for once all the exact coefficients of string corrections in both IIB, IIA. 1 One even can go further and deal with the thermodynamical facets of brane anti brane system, where it is also recognized that at finite temperature this D-brane anti D-brane is being stabilised and can be entirely associated to black holes for which some implications have been devoted to either AdS/CFT or M-theory approach [49,50]. D-brane-Anti-D-brane system has also been affecting on the stability of KKLT or in Large Volume Scenario and string compactifications [51,52]. Given a tied relation between D-branes and RR fields [53] we just point out on the bound states of the branes [54] as well. Having set the fact that no duality transformation exists for non-BPS branes, one would have to emphasize that the only way of getting exact all order α corrections of effective couplings in string theory is just through CFT and scattering methods. To notify further comments we just demonstrate [44]. Last but not least one might head off reading all standard EFT methods for both Wess-Zumino (WZ) and (non)-BPS DBI effective actions that are verified in detail in [55,56].
The paper is organised so that first we warm up with details on 3 and 4 point functions of D-brane anti-D-brane system, basically we fully address the S-matrix of a closed string Ramond-Ramond (RR) field and 2 tachyons [57] and explain how one does reconstruct all order exact α corrections in this context.
We then move on to observe more hidden symmetries in non-BPS context and as such we employ all the Conformal Field Theory (CFT) methods to a five point function of < V C −1 V T −1 V T 0 V T 0 > in both type IIA and IIB. Given the selection rules [58], EFT and the entire algebraic solutions for integrals on 5-point functions we guess an expansion and test our guess with exact solutions of the integrals related to that amplitude and then produce all the infinite massless/ tachyon poles and come to an agreement with both string and field theory. Eventually we try to address for the first time a six point function of D-brane anti-D-brane system, deriving < V C −1 V T −1 V T 0 V T 0 V T 0 > S-matrix and given all symmetries of amplitude we discover its expansion. We also show that in a particular limit that is being called soft limit 4k 2 .p → 1 the algebraic solutions for all 6-point functions can be found out. It is worth mentioning that within this limit all massless poles of the S-Matrix can be clearly observed and to be regenerated from an EFT argument as well. Lastly, we construct all order higher derivative corrections to four tachyon couplings in the context of brane anti-brane system.

Lower order D-brane-anti-D-brane effective actions
The so called effective actions of a D pDp -brane of IIA(B) may be found out by inserting tachyonic modes in WZ and DBI effective actions. By taking into account just 2 unstable branes of IIB(A) and projecting them out through (−1) F L operator one derives such an action. In order to simplify the field content, either 2 tachyons and a guage (scalar) fields JHEP09(2017)025 take part in the action whereas RR will contribute among Chern-Simons or WZ action and other fields act on DBI as follows [59] Trace in (2.1) has to be symmetric for F ab , D a T , T matrices. The definitions of the entire matrices are given by a )T . Having made use of the ordinary trace (2.1) gets substituted to A. Sen's action [60] if one makes all kinetic terms symmetrized and evaluates the trace, however, it is shown by CFT and scattering amplitudes in [59,61] that Sen's effective action does not produce consistent result with string amplitudes. Tachyon's potential in the context of type II scattering amplitude is shown by where m 2 T = −1/(2α ) and T p is the tension of a p-brane. The expansion is also made consistent result with potential of V (|T |) = e πα m 2 |T | 2 that came from BSFT as well [62].
We argued in [59] that just above effective action can entirely and consistently generate all infinite singularities as well as contact terms of sting theory. Indeed we also dicussed that the presence of new mixing couplings like F (1) ·F (2) as well as Dφ (1) ·Dφ (2) is necessary to derive the actual and consistent results that are matched string computations with EFT. The Lagrangian of SYM (at quadratic level) in the presence of tachyons must be modified and has to have the following structures/interactions [59]: WZ action for brane anti brane system with C becomes a sum on RR potentials In [66,67] it was justified how to consider the tachyons in the effective actions where another approach would be making contact with superconnection of the non-commutative JHEP09(2017)025 geometry [68][69][70] to be where the curvature and super-connection are defined accordingly as In [59] it is also verified how to derive the curvature as follows a )T ]dx a . Lastly one does extract various couplings of the WZ action (2.5) to be able to lead to diverse WZ couplings that are needed for consistent result of EFT with string side as follows The world volume of a non-BPS brane of both type IIA (IIB) string theory includes a real tachyon and the three point function of an RR and a tachyon < V C −1 V T −1 > based on direct CFT methods was given in [71] to be Using momentum conservation one reveals that p a p a = k 2 = − 1 4 , the string amplitude (2.7) can be reconstructed out in an EFT by (2πα β µ p C p ∧ DT ) coupling where there is no singularity structure for this amplitude at all.
On the other hand, the world volume of a D-brane-anti D-brane system involves two real tachyons. Using direct CFT methods [72] the four point function of an RR and two JHEP09(2017)025 To get familiar with notations we just warm up with this calculation where the vertices with their CP factors for brane-anti brane are world-sheet is taken as disk so that all open strings and RR are located on the boundary and the middle of disk accordingly. On-shell conditions are p 2 = 0, k 2 1 = k 2 2 = 1/4, notations are also followed by where for type IIA (type IIB) n = 2, 4,a n = i (n = 1, 3, 5,a n = 1) stands. To deal with standard holomorphic functions and based on various change of variables, the doubling trick has been set as follows with further ingredients as Making use of the above doubling trick, we can now head off and start working with the following two point functions for all the fields of X µ , ψ µ , φ as follows Using gauge fixing as ( and the final result can be read [57]

JHEP09(2017)025
where µ p is RR charge and the trace for γ 11 kept fixed for the following as well p > 3, H n = * H 10−n , n ≥ 5.
We actually discussed all the proper expansion of S-matrices within detail in [73], so that the entire S-matrix elements are recovered in an EFT by sending either k i .k j → 0 or (k i + k j ) 2 → 0 which means that one indeed is able to regenerate massless /tachyonic singularities of different configurations. We knew that a non-zero coupling C p−1 ∧ F exists, and also there is non-vanishing two tachyons and a gauge field coupling, thus using momentum conservation along the world volume of brane the correct expansion for D-brane-anti D-brane was explored to be as also clarified in [61]. Note that as we have seen in an RR and a tachyon amplitude this constraint for non-BPS branes gets enhanced to The only u-channel massless gauge field pole of this particular CT T S-matrix can be regenerated by following sub-amplitude in an EFT where the presence of Chern-Simons coupling on the brane anti-brane system is needed as follows 14) The off-shell propagator must be A (1) and A (2) to be able to have consistent result with both string and an EFT. The propagator and vertices are accordingly defined by

JHEP09(2017)025
Taking into account k a = (k 1 + k 2 ) a = −p a as well as replacing them in EFT amplitude, we gain the field theory amplitude as follows which is exactly as derived in string theory amplitude (2.11). In [74] we also constructed all order α corrections of one RR and two tachyons of D-brane anti D-brane system as All the correlators of this S-matrix can be investigated. We define One finds out the amplitude after the gauge fixing (x 1 , x 2 , x 3 , z,z) = (0, 1, ∞, z,z) to be Notice that all integrations on upper half plane are carried out on RR position as follows and the final result on those moduli spaces has been verified in detail for d = 0, 1 in [75] and for d = 2 in [73]. Using some algebraic analysis the ultimate form of S-matrix is obtained to be where the functions N 1 , N 2 are given as One normalizes the amplitude by 3π −1/2 β µ p /8 to be able to get consistent result with EFT. As can be explicitly seen, the amplitude is entirely symmetric with respect to interchanging JHEP09(2017)025 s, t, u and for p − 1 = n case and it has an infinite massless gauge field poles in all s, t, u− channels and also for p + 1 = n case it does have an infinite tachyon poles in (s + t + u + 1)−channel poles( s = s + 1 2 , t = t + 1 2 , u = u + 1 2 ). Given these arguments, the fact that the on-shell condition is and p a p a → 1/4 for non-BPS D-branes, one obtains uniquely the expansion for CT T T as 3) The expansion of N 1 around (3.2) is where some coefficients are The string amplitude for this case is given by These infinite u-channel gauge field poles can be produced in an EFT by the following sub-amplitude where the vertices are k is off-shell's gauge field momentum. Finally all contact interactions of this part of Smatrix or all order higher derivative corrections of C p−2 ∧ DT ∧ DT ∧ DT can be derived by the following coupling 8β (πα 2 )µ p ∞ p,n,m=0 All infinite tachyon poles are produced in an EFT by the following couplings and vertices The propagator and vertex V (C p , T ) are Following Lagrangian is also needed −T p Tr (πα )m 2 T 2 + (πα )D a T D a T − (πα ) 2 F ab F ba + T 4 (3.10) Using (3.10) and the fact that off-shell tachyon in propagator is abelian, the vertex V β (T, T, T, T ) is constructed to be V β (T, T, T, T ) = −12iT p Tr (λ 1 λ 2 λ 3 Λ β ). Considering it in (3.9), one reveals the tachyon pole of string amplitude in an EFT to be 24iβ µ p 1 (p + 1)!(s + t + u + 1) a 0 ···ap H a 0 ···ap In the next section we would like to perform an RR and four tachyon couplings on Dbrane-anti-D-brane's world volume which is the generalization of Veneziano amplitude of four open string tachyons [76]. We observe that all infinite tachyon poles are reconstructed by (2πα ) 2 µ p β 2 C p−1 ∧ DT ∧ DT and all order extensions of four tachyon couplings.
Hence CT T T T S-matrix makes sense for p − 2 = n, p = n cases accordingly. If we use the particular limit 4k 2 .p → 1 that we called soft limit 3 then one reveals that the algebraic solutions for all the integrals on upper half plane can be entirely achieved, as explained in [73]. In this particular limit, the ultimate form of the amplitude for p − 2 = n case will be given by The second part of the amplitude makes sense for p = n case and after simplifications we obtain it as follows Applying momentum conservation along the brane we derive Given (4.3), the EFT methods that imposed to have an infinite massless either u or tchannel gauge field poles, also highly symmetries [77] of this amplitude ( it should be

JHEP09(2017)025
symmetric under exchanging Mandelstam variables u, t and other variables), k i .k j → 0 and the fact that for tachyonic strings p a p a → 1 4 we obtain the following expansion for this particular soft limit of the scattering amplitude as follows t → 0, r → − 1 4 , (u, s, → − 1 2 ) Given the so called selection rules for non-BPS couplings of string theory [58] and also the fact that all kinetic terms have already been fixed in DBI action, one comes to know that for p − 2 = n case we have just an infinite single massless gauge field poles in both u and t-channels. The expansion of M 1 around the first above expansion is As it is clear from (4.2) and M 1 for p − 2 = n case the amplitude has an infinite u,t-channel massless gauge field poles and due to symmetries we just produce u-channel poles as follows.
where k b = (k 2 +k 3 ) b is the momentum of off-shell gauge field. Note that V a (C p−3 , A, T 1 , T 4 ) was derived from coupling If we normalize the amplitude by , then one observes that the first u-channel gauge field pole is reconstructed in an EFT by (4.5) to be The vertex of V b (A (1) , T 2 , T 3 ) and single pole are fixed and have no corrections so to be able to generate all other massless u-channel poles one needs to apply higher derivative corrections to above WZ coupling (4.6), however, given the entire explanations of the previous sections we are no longer interested in doing so. We would rather carry out the rest of the analysis of the amplitude which has something to the extensions of four tachyon couplings.
On the other hand, having set the EFT arguments, one would become aware of the fact that for p = n case, the amplitude has an infinite (u + w + r + 1) channel (and due to symmetries (t + s + u + 1), (s + v + w + 1), (t + v + r + 1) channel as well) tachyon singularity structures and a double pole that can be shown later on.
As it is clear from A 2 , it has an infinite (u + w + r + 1) channel tachyon poles that can be explored in an EFT within the following sub-amplitude as follows To produce an infinite tachyon poles, one needs to employ the higher derivative corrections to four tachyon couplings as follows with following definition for D nm and D nm derivative operators as D nm (EF GH) ≡ D b 1 · · · D bm D a 1 · · · D an EF D a 1 · · · D an GD b 1 · · · D bm H D nm (EF GH) ≡ D b 1 · · · D bm D a 1 · · · D an ED a 1 · · · D an F GD b 1 · · · D bm H JHEP09(2017)025 All order four tachyon extensions of V β (T, T 2 , T 3 , T 4 ) might be explored from the higher derivative couplings (4.8) as below 2iT p (α ) n+m (a n,m − b n,m )Tr (λ 2 λ 3 λ 4 Λ β ) u r (k 1 ·k) m (k 2 ·k 4 ) n + (k 1 ·k) n (k 2 ·k 4 ) m +(k 2 ·k) m (k 2 ·k 3 ) n + (k 2 ·k) n (k 2 ·k 3 ) m + (k 1 ·k) m (k 2 ·k) n + (k 1 ·k) n (k 2 ·k) m +(k 2 ·k 3 ) m (k 2 ·k 4 ) n + (k 2 ·k 3 ) n (k 2 ·k 4 ) m + (u, r → u, w) + (u, r → r, w) k a is off-shell tachyon's momentum. Taking into account this vertex in (4.7), one gains all tachyon poles in an EFT as follows (a n,m − b n,m ) × u r (u m r n + u n r m ) + u w (u m w n + u n w m ) + r w (r m w n + r n w m ) One may re-write down the above amplitude as follows ∞ n,m=0 e n,m (u + r ) n (u r ) m+1 + (u + w ) n (u w ) m+1 + (r + w ) n (r w ) m+1 These tachyon poles are exactly the same singularities that appeared in second part of string amplitude. EFT imposes to us that the amplitude does have just a double pole for C p−1 case as and the off-shell gauge field A needs to be both A (1) and A (2) with the following vertices derived G ab (A) = iδ ab (2πα ) 2 T p u (4.10) with k is off-shell gauge field's momentum. Inserting above vertices in (4.9) we are then able to reconstruct the double pole in an EFT as follows β 2 (2πα )µ p −1 u(u + r + w + 1) 1 2p! a 0 ···a p−1 c H a 0 ···a p−1 (p + k 1 ) c (r − w) (4.12)

JHEP09(2017)025
It is worth to pointing out that the appearance of F (1) · F (2) and also Dφ i(1) · Dφ i(2) couplings for D-brane-Anti D-brane DBI action have been confirmed by direct consistent string amplitudes of [59] and [74] accordingly. More crucially, we got to know that C p−1 ∧F does not receive higher derivative correction of WZ couplings. The other important result that we found is that as the kinetic terms donot get any corrections hence all non-leading tachyonic singularities provide some information not only about the structure of higher derivative corrections to some couplings such as C p−1 ∧ DT ∧ DT but also could fix their coefficients on the world volume of brane-anti brane system as well. We could further move on and talk about all the contact terms of the string amplitude, for instance a new coupling such as C p−1 ∧F T T * can also be confirmed. We hope to further analyze the higher derivative corrections to those new couplings in near future and to be able to eventually fix all the higher derivative corrections to brane anti-brane DBI as well as WZ effective actions.
Remarks are in order. As we discussed these non-BPS couplings are worked out in the presence of the constraint p a p a → 1/4, hence these couplings cannot be compared with BSFT couplings, although the tachyon potential is the same as the one that appeared in BSFT (V (T ) = e πα m 2 T 2 [62]) i.e., where m 2 T = −1/(2α ). However, note that there is a symmetrized trace on the σ factors in DBI action, carrying out the symmetrized trace, one finds The tachyon does condensate at T → ∞ and hence, the tachyon potential goes to zero at that point as well.