Spin(7)-manifolds as generalized connected sums and 3d \( \mathcal{N}=1 \) theories

Journal of High Energy Physics, Jun 2018

Abstract M-theory on compact eight-manifolds with Spin(7)-holonomy is a framework for geometric engineering of 3d \( \mathcal{N}=1 \) gauge theories coupled to gravity. We propose a new construction of such Spin(7)-manifolds, based on a generalized connected sum, where the building blocks are a Calabi-Yau four-fold and a G2-holonomy manifold times a circle, respectively, which both asymptote to a Calabi-Yau three-fold times a cylinder. The generalized connected sum construction is first exemplified for Joyce orbifolds, and is then used to construct examples of new compact manifolds with Spin(7)-holonomy. In instances when there is a K3-fibration of the Spin(7)-manifold, we test the spectra using duality to heterotic on a T 3-fibered G2-holonomy manifold, which are shown to be precisely the recently discovered twisted-connected sum constructions.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:


Spin(7)-manifolds as generalized connected sums and 3d \( \mathcal{N}=1 \) theories

HJE Spin(7)-manifolds as generalized connected sums and Andreas P. Braun 0 1 Sakura Schafer-Nameki 0 1 0 Woodstock Road , Oxford, OX2 6GG , U.K 1 Mathematical Institute, University of Oxford M-theory on compact eight-manifolds with Spin(7)-holonomy is a framework for geometric engineering of 3d N = 1 gauge theories coupled to gravity. We propose a new construction of such Spin(7)-manifolds, based on a generalized connected sum, where the building blocks are a Calabi-Yau four-fold and a G2-holonomy manifold times a circle, respectively, which both asymptote to a Calabi-Yau three-fold times a cylinder. The generalized connected sum construction is rst exempli ed for Joyce orbifolds, and is then used to construct examples of new compact manifolds with Spin(7)-holonomy. In instances when there is a K3- bration of the Spin(7)-manifold, we test the spectra using duality to heterotic on a T 3- bered G2-holonomy manifold, which are shown to be precisely the recently discovered twisted-connected sum constructions. Di erential and Algebraic Geometry; String Duality; M-Theory - N = 4 Spin(7)-manifolds as generalized connected sums The construction Calibrating forms Topology of GCS Spin(7)-manifolds GCS Spin(7)-manifolds as quotients of CY4 A simple example 4.4 3d eld theory and sectors of enhanced supersymmetry 5 M-theory on GCS Spin(7)/heterotic on TCS G2 5.3.1 5.3.2 5.3.3 5.4.1 5.4.2 5.5.1 5.5.2 5.5.3 5.5.4 Heterotic string theory on G2-manifolds GCS Spin(7)-manifolds and M-theory/heterotic duality Duality for the building blocks Geometric preparation: dual pairs of G2 and CY3 The acyl G2-manifold and its dual acyl CY3 The acyl CY4 and its dual acyl CY3 5.4 First example of dual pairs: new Spin(7)- and its dual G2-manifold 5.5 Second example of dual pairs The M-theory on GCS Spin(7)-manifold The dual heterotic model on TCS G2-manifold The acyl Calabi-Yau four-fold Z+ The acyl G2-manifold Z M-theory spectrum on the GCS Spin(7)-manifold The dual heterotic model 1 Introduction 2 3 2.1 2.2 3.1 3.2 3.3 3.4 4.1 4.2 4.3 4.5 4.6 5.1 5.2 5.3 Spin(7)-holonomy, M-theory and 3d N = 1 theories Spin(7)-holonomy M-theory on Spin(7)-manifolds A Spin(7) Joyce orbifold as a generalized connected sum Setup and motivation Joyce orbifolds: Spin(7)- and G2-holonomy TCS-decomposition of the Joyce G2-manifold Uplift to GCS-decomposition of the Spin(7)-manifold 6 Discussion and outlook A TCS-construction of G2-manifolds { i { Introduction Geometric engineering of supersymmetric gauge theories is fairly well-understood in theories with at least four real supercharges. Most prominently in recent years, F-theory has established a precise dictionary between geometric data (and uxes) and 6d and 4d theories with N = 1 supersymmetry, and dually, M-theory of course has a well-established dictionary between 3d and 5d theories and Calabi-Yau geometries. For 4d N = 1 the realization in terms of M-theory compati cations on G2-holonomy manifolds is already much less well understood, in particular due to the scarcity of compact geometries of this type. Even less is known about M-theory on Spin(7)-holonomy compacti cations, which yield theory, and we relate these constructions in certain instances to heterotic string theory on G2-holonomy manifolds. A second motivation is to study Spin(7) compacti cation in the context of M/F-duality and uplifting these to supersymmetry breaking vacua in F-theory [1] based on the observations that 3d N = 1 supersymmetry can be related to the circle-reduction of a not necessarily supersymmetric theory in 4d [2, 3]. The construction that we propose, and refer to as generalized connected sum (GCS) construction of Spin(7)-manifolds is motivated by a recent development for compact G2holonomy manifolds in Mathematics, where a large class of compact G2-holonomy manifolds were constructed using a twisted connected sum (TCS) construction [4{6]. This construction relies on the decomposition of the G2-holonomy manifold in terms of two asymptotically cylindrical (acyl) Calabi-Yau three-folds | for a sketch see gure 1. Following these mathematical developments, there has been a resurgence in interest in the string and M-theory compacti cations, which for TCS-manifolds have been studied in [7{14]. For a review of M-theory on G2 and Spin(7)-holonomy manifolds related to earlier results in the '90s and early '00s see [15]. One of the nice features of TCS G2-manifolds is that the connected sum gives rise to a eld theoretic decomposition in terms higher supersymmetric subsectors [11, 13]: the asymptotic regions that are acyl Calabi-Yau three-folds (times a circle), give rise to 4d N = 2 subsectors, whereas the asymptotic neck region where the two building blocks are glued together is by itself K3 times a cylinder times S1, and corresponds to a 4d N = 4 subsector, which is present in the decoupling limit (in nite neck limit). The theory breaks to N = 1, when the asymptotic region is of nite size and the states of the N = 4 vector multiplet become massive, leaving only an N = 1 massless vector. This observation may be used to study non-perturbative corrections, e.g. M2-brane instantons [16], which has been initiated for TCS-manifolds in [7, 8, 14]. For Spin(7)-manifolds the type of constructions have thus far has been rather limited. There are (to our knowledge) two constructions of compact manifolds with Spin(7)holonomy due to Joyce: either in terms of a Joyce orbifold T 8= [17] or a quotient by an anti-holomorphic involution of a Calabi-Yau four-fold (CY4) [18]. There are variations of { 1 { + HKR K3 1 x Z \S-0 S S 3d N = 1 theory, which is crucially related to the singular loci of the Calabi-Yau quotient, and not only the sector that is 3d N = 2. In the present paper we provide a di erent approach to constructing Spin(7)-manifolds, which is closer in spirit to the TCS construction. We will motivate a generalized connected sum (GCS)-construction where the building blocks are an open asymptotically cylindrical (acyl) Calabi-Yau four-fold (CY4) and a open asymptotically cylindrical (acyl) G2holonomy manifold (times S1), which asymptote to a Calabi-Yau three-fold (times a cylinder/line). This is shown in gure 2. Explicitly, we construct the acyl CY4 as CY3 ,! CY4 ! C ; (1.1) and the G2 = (CY3 R)=Z2, where Z2 acts as a re ection on the circle and anti-holomorphic involution on the CY3. In case the anti-holomorphic involution acts freely, these are socalled `barely G2-manifolds' [16], which have holonomy group SU(3) o Z2. Field theoretically we will show that each of the building blocks will give rise to a 3d N = 2 subsector, and the asymptotic region CY3 cylinder results in the limit of an in nite neck region in a 3d N = 4 sector. At nite distance, some of the modes of the vector multiplet of 3d N = 4 become massive and the theory is broken to 3d N = 1. { 2 { HJEP06(218)3 + Z =CY 2 x S1 this paper: the left hand building block is an the acyl Calabi-Yau four-fold building block. We require this to be asymptotically approach a Calabi-Yau three-fold times a cylinder. One way to realize this is in terms of a CY3 bration over an open P1. The right hand building block is a circle times a G2-manifold, which asymptotes to a CY3 I times a circle. The simplest way to realize this is in terms of G2 = (CY3 region is CY3 cylinder. R)=Z2 where Z2 acts as an anti-holomorphic involution. The asymptotic We motivate this construction in section 3 by considering a simple Spin(7) Joyce orbifold, T 8= , which is known to have a related G2 Joyce orbifold, where the precise connection is through M-theory on K3 to Heterotic on T 3 duality applied to this toroidal setting [22]. It is known that such Joyce orbifolds have a TCS-decomposition [12], which we apply to the G2 Joyce orbifold, and subsequently uplift the decomposition to the Spin(7) Joyce orbifold. The TCS-building blocks then map precisely to an open CY4 and G2 respectively. This motivates our general construction of GCS Spin(7)-manifolds, which will S1, be given in section 4. One may ask whether these constructions in fact globally t together to (resolutions of) CY4 quotients, like in Joyce's constructions in [18]. Whenever the acyl G2 manifold is realized as (the resolution of) a quotient of X3 R, this is indeed the case and we can write our GCS Spin(7)-manifolds globally as resolutions of quotients of CY4 by an anti-holomorphic involution. Here, one starts with a Calabi-Yau four-fold bered by a Calabi-Yau three-fold CY3 as in (1.1), and acts on it with an anti-holomorphic involution on the CY3- ber combined with an action on the base P1 given by zi $ zj , i 6= j, where zi are the homogeneous coordinates. The xed locus of this on the base is a circle and the quotient space is half-CY4, which is a CY3-fold bered over a disc, and in the vicinity of the boundary circle we obtain (CY3 R)=Z2 which is a (barely) G2-manifold, times an S1. Pulling the base half sphere apart, gives the decomposition into the GCS building blocks. This is sketched in gure 3. The key di erence to the examples and constructions in [18] is however that the singularities are either absent or occur over a locus of real dimension at least 1. In Joyce's construction via CY4= the singularities are only point-like and get resolved using an ALE-space. It seems likely that the analog to this procedure in our construction is gluing in the open, acyl G2-manifold. { 3 { HJEP06(218)3 bered over a disc, which is locally a CY4. Above the boundary circle (dark green) the quotient acts non-trivially on the CY3 interval, shown in light green, which results in a G2. Pulling the base apart, results in the GCS-decomposition of the resulting Spin(7)-manifold. The paper is structured as follows: we begin with some background material on Spin(7)-holonomy and M-theory compacti cations to 3d N = 1 in section 2. In section 3 we motivate our construction by rst considering a Joyce orbifold example. Section 4 contains our proposal for the general construction of GCS Spin(7)-manifolds. We also comment on how this construction relates to and di ers from other known setups and provide an example of a new Spin(7)-manifold. For some examples of GCS constructions there is a K3bration of the Spin(7)-manifold, and we utilize this to apply M-theory/heterotic duality and construct the dual heterotic on G2 compacti cations in section 5, and match the spectra of the theories. We conclude with a discussion and outlook in section 6. For reference we include a brief summary of the TCS-construction of G2-manifolds in appendix A. 2 Spin(7)-holonomy, M-theory and 3d N = 1 theories As a preparation for this work and to introduce some notation, this section reviews some basic aspects of manifolds with holonomy group Spin(7), see [23] for a detailed discussion. 2.1 Spin(7)-holonomy A compact orientable 8-dimensional manifold Z which has a Ricci- at metric g with holonomy group contained in Spin(7) supports a closed, self-dual four-form = g , which can be expressed in local coordinates (in which g is the euclidean metric) as = dx1234 + dx1256 + dx1278 + dx1357 dx1368 dx1458 dx1467 dx2358 dx2367 dx2457 + dx2468 + dx3456 + dx3478 + dx5678 : (2.1) Here, we use dxijkl as a shorthand for dxi ^ dxj ^ dxk ^ dxl. This four-form is in the stabilizer of the action of the holonomy group Spin(7). Such a form de nes a Spin(7) structure, which is called torsion free if is closed and self-dual. { 4 { Having this structure in place does not necessarily mean that the holonomy group is exactly Spin(7) and, as long as Z is simply connected, we may discriminate between di erent cases by computing the A^ genus, which tells us the number of covariantly constant spinors: ^ A 1 2 3 hol(g) Spin(7) SU(4) Sp(2) dimensions, we will only be interested in the case A^ = 1. terested in compacti cations of M-theory which preserve N = 1 supersymmetry in three For eight-manifolds Z with holonomy contained in Spin(7) and A^(Z) = 1, a necessary and su cient condition for the holonomy group to be all of Spin(7) is that Z is simply connected, 1(Z) = 0 [17, 23]. This still allows for cases with non-trivial subgroups of Spin(7), and we will see examples of such spaces later on. We will refer to manifolds Z with A^(Z) = 1 and a metric g with hol(g) Spin(7) as barely Spin(7)-manifolds. The dimension of the moduli space of Ricci- at metrics on a Spin(7)-manifold is given by b4 + 1. Together with A^ = 1, this number is already determined by the Euler characteristic and the two Betti numbers b2 and b3 by using (2.3) b 4 = 3 9 b2 + b3 : Calibrated submanifolds of Spin(7)-manifolds must be of real dimension four and are called Cayley submanifolds. The dimension of the moduli space of such a Cayley submanAll of these manifolds have b1(Z) = 0 and the remaining independent Betti numbers are related to the A^ genus by HJEP06(218)3 1 2 1 2 { 5 { ifold N is [23, 24] mN = (N ) (N ) N N ; 1 2 1 2 so that Spin(7)-manifolds cannot possibly be bered by calibrated four-tori. where (N ) is the Hirzebruch signature and (N ) is the Euler characteristic. Note that this expression evaluations to mN = 4 N N for a Cayley submanifold which is a K3 surface, where (N ) = 16 and (N ) = 24. It hence seems sensible to assume there exist Spin(7)-manifolds bered by K3 surfaces over a four-dimensional base. Using duality to heterotic strings, it is precisely the existence of such brations which we will conjecture and exploit. In contrast, note that for a Cayley submanifold with the topology of a four-dimensional torus T 4, the same computation gives mN = N N M-theory on a Spin(7)-manifold gives rise to 3d N = 1 supersymmetric theory and the spectrum is encoded in the topological data of the Spin(7)-manifold as follows: # b3(Z) b2(Z) b4 (Z) 1 3d N = 1 Multiplet Scalar Vector Scalar Scalar g C3 C3 11d Origin Pib=31 i(3)'i Pib=21 !(2)v PbI4=1( I(4)) Volume modulus where !(2), (3) are a basis of harmonic two and three-forms, and I(4) are a basis of harmonic anti-self-dual four-forms. For 3d N = 1 the scalar multiplet has only a real scalar as its bosonic component and the vector multiplet is just a 3d vector. As we can dualize the 3d vectors to real scalars, compacti cations of M-theory on Spin(7)-manifolds hence give rise to ns = b4 + 1 + b2 + b3 = 8 + 2b3 (Z) 3 NM2 = (Z) 1 24 space-time lling M2-branes [25, 26]. As such M2-branes can freely move on Z, they each contribute a further 8 real degrees of freedom. In the absence of G4- ux, the e ective action for a smooth Spin(7)-manifold is a 3d N = 1 eld theory with b2 abelian vectors, and scalars with the following kinetic terms GIJ Z Z I ^ J(4) ; (4) M Z Z !(2) ^ ?!(2) : The theory for a smooth Spin(7)-manifold is an abelian gauge theory. With singularities this can enhance to non-abelian gauge symmetries. With G4- ux, additional Chern-Simons terms and scalar potential are generated | for an in depth discussion of the e ective theory see [20, 27{33]. In this paper we will not consider uxes but focus on the geometric constructions. Of course it is interesting to consider this in the future and study the e ects of these on supersymmetry breaking and potential obstructions to dualities, both M/F-duality [ 34 ], where in particular 4d Poincare invariance could be broken, as well as obstructions to M-theory/weakly-coupled heterotic duality [35, 36]. 3 3.1 A Spin(7) Joyce orbifold as a generalized connected sum Setup and motivation Our goal is to construct new classes of Spin(7)-manifolds | which we have motivated from various points of view in the introduction. The construction which we will end up with is inspired by combining two ideas: { 6 { ' I 1. The recent construction of G2-holonomy manifolds as twisted connected sums (TCS), with each building block an asymptotically cylindrical Calabi-Yau three-fold. We review this construction in appendix A. 2. M-theory on K3/heterotic on T 3 duality. Combining these ideas will lead us to consider a generalized connected sum (GCS) construction, where | as we will show | the building blocks are acyl CY four-folds and G2-manifolds, respectively. To motive this we start with a well-known construction by Joyce of both G2 and Spin(7)-manifolds and a well-known duality, between M-theory and heterotic strings, which will be discussed in more detail in section 5.2, between M-theory on K3 ! Heterotic on T 3 (3.1) This duality in 7d is based on the agreement of the moduli spaces of the two compacti cations Mhet=T 3 = MM=K3 = [SO(3; Z) SO(10; Z)] nSO(3; 19)= [SO(3) SO(10)] R+; (3.2) which is both the Narain moduli space of heterotic string theory on T 3 and the moduli space of Einstein metrics on K3 [37] (for a recent exposition in the context of ber-wise application see [12]). The string coupling on the heterotic side is matched with the volume modulus of the K3. We can ber this over a four-manifold M4 in such a way that there is a duality of 3d N = 1 theories M-theory on Spin(7)-manifold Z8 Heterotic on G2-manifold J7 (3.3) where Z8 is K3- bered over M4 and J7 has T 3- bers. This duality has been tested in the case when both manifolds are Joyce orbifolds in [22]. More generally, berwise application of this duality will lead us to a correspondence between two realizations of a 3d N = 1 theory, in terms of a Spin(7)-compacti cation of M-theory, and a T 3- bered G2-holonomy manifold.1 It is this setup which will motivate our construction of GCS Spin(7)-manifolds: it is the analog of the TCS-decomposition for G2-manifolds on the heterotic side, mapped to M-theory using the duality. In this way we obtain a dual pair of connected sums: M-theory on GCS Spin(7)-manifold Heterotic on TCS G2-manifold (3.4) As a warmup we now show how this works for a simple G2 Joyce orbifold, which has a TCS-decomposition and determine what this decomposition corresponds to on the Mtheory side. This will give a rst hint as to what the general connected sum construction 1The existence of T 3- brations of G2 manifolds has also been conjectured in the context of mirror will be for Spin(7)-manifolds. symmetry for G2 manifolds in [13]. ! ! { 7 { We will start the construction of a Spin(7)-manifold Z8 by considering a Joyce orbifold T 8= , where = Z24, where each generator of the order two subgroups acts as follows [23]: The entries 12 denote x ! x + 12 . The singular sets are locally given by x1 + 1 2 x2 + 1 2 + S : 4 S : 4 S : 2 S : 2 C2=Z2 C2=Z2 C2=Z2 , the singularities of which are modelled on C4=Z2. It was shown by Joyce that the resolution of the orbifold in this way yields a Spin(7)-holonomy manifold. From this, the Betti numbers are computed as follows. First note that the only even classes in H (T 8) under the Z24 (besides H0(T 8) and H8(T 8)) are given by 14 classes in H4(T 8). Resolving the 8 singularities of the form C2=Z2 T 4= f 1g gives whereas the 4 singularities of the form C2=Z2 T 4 yield b b b 2 3 4 b b b 2 3 4 6 0 1 6 4 1 C2=Z2 T 4= f 1 g C^2=Z2 T 4= Hence each of these 12 singularities contributes b2 = 1 and b4 = 6 and only the second 4 contribute a non-zero b3 = 4. Finally, the resolution of the 64 singularities of the form C4=Z22 contribute b4 = 1 and nothing else. Altogether we can compute is the stabilizer of x6 and so The notation T(32;6;8) indicates the three-torus along the coordinates x2; x6; x8. We now need to identify the action of and on this space. Again, there will be two components, which we denote by X0 and X1=4, located at x6 = 0; 1=4, which are given by x6 : ( ; ; ) = ; 1 2 ; + x6 x6 + 12 , and thus x6 2 I = [0; 1=4]. The generator X0 : X1=4 : >>:z31=4 = x6 + ix2 Each of these halves is an open, K3- bered CY three-fold. To see this introduce the coordinates G2-manifold as follows: consider the T 7 given by x1; x2; x3; x5; x6; x7; x8, and act with Z2. This is a Joyce construction of G2-manifolds by orbifolds as already observed TCS-decomposition of the Joyce G2-manifold Our goal is to rst identify the TCS description of this Joyce G2-orbifold and to then lift this to a connected sum description, GCS, for the Spin(7)-manifold. First note that T 7=h ; ; i is bered over an interval HJEP06(218)3 (3.10) (3.11) (3.12) (3.13) (3.14) The K3s are along z1; z2, where acts non-trivially. The remaining coordinates combine with the interval coordinate x6. At x6 = 0(1=4) the x8(x2) circle pinches. A sketch is shown in gure 4. The holomorphic three-form is invariant under these actions, making these open CY3. 3.4 Uplift to GCS-decomposition of the Spin(7)-manifold Adding back the circle along the x4 coordinate, as well as the additional orbifold generator z , 0 1 ! acts trivially on x6, so that the bration over the interval remains intact, but : z10, so that the (3;0) form is not invariant any longer. Let us now again de ne two building blocks along the x6 interval, close to either boundary. The interesting observation is that the two halves behave quite di erently: rst consider x6 = 0. De ne another complex coordinate with the action of z 0 4 x2 + ix4 ! 0 z4 : { 9 { Z1=4, which are an open Calabi-Yau four-fold and an open G2-manifold times S(18). These are glued together along the x6 coordinate. Both asymptote to Xs times a cylinder, where Xs is the Schoen Calabi-Yau three-fold. In this particular example each building block contains a K3, which is also present in the neck region of the Schoen Calabi-Yau. Together with the action of on zi0, we see that (4;0) = dz10 ^ dz20 ^ dz30 ^ dz40 is invariant, 0 and we obtain a building block which is an open Calabi-Yau four-fold At the other end of the x6 interval we have where M7 will be shown to be a G2 holonomy manifold. The action of is From this we can de ne the G2-form 3 = dx4 ^ !(1;1) + Re (3;0) ; 3 = 2 1 !(1;1) ^ !(1;1) dx4 ^ Im (3;0) ; (3.15) (3.16) (3.17) (3.18) which is invariant, as follows from the action of and the coordinates z1=4. In summary we obtained a GCS-construction of a Spin(7) Joyce orbifold Z, which has two building blocks, an open CY4 and an open G2, W times a circle, respectively. The geometry in the middle of the x6 internal is Q = Xs where the Xs is a CY3 along the directions 1; 2; 3; 4; 5; 7. Not too surprisingly, this is in fact the Schoen Calabi-Yau three-fold. The action along these coordinates, written in terms of the complex coordinates of Z0 is (see [12] for a detailed discussion of that). Inside Xs, there is a K3 surface along z10; z20. On the other hand the CY4 Z0 is bered by Xs over x6; x8. Note that we may also think about this Spin(7) orbifold as a quotient of a CalabiYau four-fold X4 by an anti-holomorphic involution. The action of ; ; respects the holomorphic coordinates zi0 on T 8 and leaves 0 (4;0) invariant, so that it produces a CalabiYau orbifold. The full Spin(7) orbifold is then formed by acting with , which acts as an anti-holomorphic involution. We expect this structure to persist after resolution. The Joyce orbifold (3.5) has several generalizations, which can be parametrized in terms of the following data: (3.19) (3.20) (3.21) (3.22) (3.23) where the shift vectors can take the following values Case I is the orbifold studied above. Instead of the bration over the interval (3.10), in this general setting, we consider pulling these orbifolds apart along x3, with the interval x3 2 [0; 1=4]. There is an asymptotic middle-region, where the space is Xs = T(61;2;5;6;7;8)=h ; i, which in fact is again the Schoen Calabi-Yau three-fold for all of these Joyce orbifolds. The asymptotic manifolds at the and G2-manifold M7, respectively, given by xed points x3 = 0; 14 are again open a Calabi-Yau four-fold x3 + + d3 x4 + + + x5 + c5 d5 (1; 1; 1; 1) ; (1; 0; 1; 0) ; (1; 1; 1; 0) ; (1; 0; 1; 0) ; x6 + c6 + 1 2 1 2 1 2 1 2 d = d = d = d = x7 + + d7 x8 + + + (0; 1; 1; 1) (0; 1; 1; 1) (0; 1; 1; 0) (0; 1; 1; 0) : I : II : III : IV : x1 + c1 d1 x2 c2 + + 1 2 1 2 1 2 1 2 c = c = c = c = The analysis will follow very much along the lines of the rst example we studied and we will instead now move to generalize this construction beyond Joyce orbifolds. We have seen that the Joyce orbifold Z8 (3.5) has a natural decomposition in terms of a GCS-construction, where one building block is an open Calabi-Yau four-fold, and the other is a G2-manifold times a circle, where both geometries asymptote to a Calabi-Yau three-fold Xs times a cylinder. With this example in mind, we now turn to providing a generalization of this construction in the next section. 4 Spin(7)-manifolds as generalized connected sums In the discussion of the last section, we were led to think about Spin(7) Joyce orbifolds as being glued from two eight-manifolds: a non-compact Calabi-Yau four-fold and the product of a non-compact G2-manifold with a circle. These two halves are glued along their overlap, which is the product of a Calabi-Yau three-fold with an open cylinder I S 1 | see gure 2. We are now going to propose a generalization of this procedure in which both the two building blocks Z and the resulting Spin(7)-manifold Z are no longer (resolutions of) orbifolds. To start this discussion, let us rst propose the following de nitions: De nition. An asymptotically cylindrical (acyl) Calabi-Yau four-fold Z+ is a non-compact algebraic four-fold, which admits a Ricci- at metric, is simply connected, and is di eomorphic (as a real manifold) to the product of a cylinder I S1 and a compact Calabi-Yau three-fold X3 outside of a compact submanifold . The Ricci- at metric of Z+ exponentially asymptotes to the Ricci at metric on the product I S 1 X3 = X4 n . Asymptotically Calabi-Yau manifolds were discussed in [38]. Similar to the explicit construction of acyl three-folds in [5, 6], we expect to able to construct Z+ by excising a ber X3 from a compact four-fold Z~+ with c1(Z~+) = [X3], which is bered by Calabi-Yau three-folds. This implies that we can think of producing Z+ by appropriately cutting a compact, CY3- bered Calabi-Yau four-fold in half, or via toric methods as in [9]. Likewise De nition. An asymptotically Calabi-Yau (acyl) G2-manifold Z is a non-compact manifold of G2 holonomy, which is di eomorphic to the product of an interval I with a compact Calabi-Yau three-fold X3 outside of a compact submanifold of Z , and the Ricci- at metric on Z exponentially asymptotes to the Ricci- at metric on the product manifold I X3. Such manifolds were discussed in [39]. Note that such G2-manifolds are easily constructed as (resolutions of) orbifolds (X3 Rt)=Z2 with the Z2 acting as t ! t and as an anti-holomorphic involution on the Calabi-Yau three-fold X3. The quotient by such involutions can be thought of as a Calabi-Yau three-fold X3 bered over an half-open interval which undergoes a degeneration at one end. In particular, it is not surprising if the Ricci- at metric on such a resolution of (X3 R)=Z2 asymptotes to the Ricci- at metric on X3 R far away from the origin of R. A particularly simple case is given by freely acting anti-holomorphic involutions. In this case no resolution is required, and the Ricci at metric on (X3 R)=Z2 is simply the quotient of the Ricci- at metric on X3 R. The simplicity of this examples comes at a price, however, as the holonomy group of such acyl G2-manifolds is not the full G2 [39, 40], but only SU(3) o Z2, so that we can call them acyl barely G2-manifolds. Such acyl barely G2-manifolds in particular have a nontrivial fundamental group Z2. We are now ready to propose our construction of what we will call generalized conZ Z nected sum (GCS) Spin(7)-manifolds. Take an acyl Calabi-Yau four-fold Z+ with asymptotic neck region isomorphic to It+ 1 S + X3+ for t+ = 0 with asymptotic neck region isomorphic to It X3 for t l and a acyl G2-manifold = 0 l. Then Z+ and 1 can be glued as topological manifolds to a manifold Z by identifying the neck regions It S 1 X3 such that t+ = l + = t : X3+ = X3 (4.1) for a biholomorphic map . If the Ricci- at metrics on Z asymptote to the Ricci- at metrics on X3 , this takes us close to a Ricci- at metric on Z. We conjecture that for long enough neck regions (l large enough) there exists a Ricci- at metric g associated with a torsion free Spin(7) structure on Z, which is found by a small perturbation of the Ricci- at metrics on Z+ and Z S1. If Z are both simply connected, it follows from the Seifert-van Kampen theorem that Z is a simply connected eight-manifold. This means that the holonomy group of Z must be equal to Spin(7) (and not a subgroup) if Z has a torsion-free Spin(7) structure. Before exploring the consequences of our proposed construction further, let us brie y discuss the mathematical work needed to put our proposal on rm ground. The work [38{40] on acyl Calabi-Yau four-folds and acyl G2-manifolds and their deformation theory, together with clear criteria when the asymptotic Calabi-Yau three-folds X3 allow a biholomorphic map , should clarify under which circumstances a gluing can be found for a given pair of such manifolds. In our examples, we can easily nd such di eomorphisms by realizing X3 as hypersurfaces in toric varieties, so that a di eomorphic pair can be simply constructed by writing down identical algebraic equations. The crucial task is then to show the existence of a Ricci- at metric with holonomy Spin(7) on the resulting topological manifold Z. By making the neck regions very long, we expect that the torsion introduced when gluing Z+ and Z S1 can be made su ciently small for a torsion-free Spin(7) structure to exist nearby. As we shall discuss in more detail in section 4.5, there are instances of GCS Spin(7)manifolds which can globally be realized as (resolutions) of quotients of Calabi-Yau fourfolds. Quotients by free actions, where no resolution is needed, or situations in which resolutions with Ricci- at metrics can be constructed provide non-trivial examples of our construction in which their existence is proven by other means. 4.2 Both the acyl Calabi-Yau four-fold Z+ and the product of the acyl G2-manifold Z with a circle have a torsion free Spin(7) structure. In this section we show how the identication (4.1) between the asymptotic neck regions produces a globally de ned Spin(7) structure. Let us denote the Kahler form and holomorphic four-form of Z+ by !+ and +. The Cayley four-form de ning a torsion-free Spin(7) structure on Z+ is then given by In the asymptotic neck region, we can introduce coordinates t+ and + and decompose the SU(4) structure as + = Re + + !+ ^ !+ : (4.2) 3;+ are the Kahler form and holomorphic three-form on X3+. This means that the Spin(7) structure in the neck region is 1 2 1 2 which is noting but the map (4.1) proposed earlier. t+ = + = 3;+ = !3;+ = !3; t 3; (4.3) (4.4) (4.5) (4.6) (4.8) + = d + ^ Re 3;+ dt+ ^ Im 3;+ + !3;+ ^ !3;+ + d + ^ dt+ ^ !3;+ : Similarly, the Spin(7) structure on Z S 1 is given by in terms of the G2 structure ' on Z . In the neck region, this is further decomposed as + = d ^ ' + ' ' ' = = dt ^ !3; + Re 3; 1 2 !3; ^ !3; + dt ^ Im 3; Spin(7) structure on the neck region of Z S1 can be written as in terms of the Kahler form and holomorphic three-form on X3 . We hence nd that the = d ^ dt ^ !3; + d ^ Re 3; + 1 2 !3; ^ !3; + dt ^ Im 3; : (4.7) This means that the two Spin(7) structure are consistently glued together under a di eomorphism which identi es the neck regions as The easiest topological number to determine for the GCS Spin(7)-manifold Z is given by the Euler characteristic (Z). As it is additive, we have (Z) = (Z+) + (Z S1) Z+ \ (Z S1) = (Z+) ; (4.9) where we have used the fact that the Euler characteristic vanishes for any manifold with an S1 factor. Two copies Z+, Z+0 of the acyl Calabi-Yau four-fold Z+ can be glued to a compact Calabi-Yau four-fold X4 = Z+ [ Z+0 such that Z+ \ Z+0 = I three-fold X3. It follows that (X4) = 2 (Z+) = 2 (Z). As the Euler characteristic of any Calabi-Yau four-fold is divisible by six [41], this implies that (Z) is divisible by 3, so S 1 X3 for a Calabi-Yau that b4 in (2.4) is always an integer. As we have de ned an acyl Calabi-Yau four-fold to be simply connected, but left room for the possibility of an acyl G2-manifold to have a non-trivial fundamental group, the Seifert-van Kampen theorem tells us that Z is simply connected if an only if Z is simply connected. If Z is not simply connected, Z can have a non-trivial fundamental group, which signals that the holonomy group of Z is smaller than Spin(7) [17, 23]. The easiest way to construct acyl G2-manifolds is by a free quotient X3 R by Z2, in which case the fundamental group of Z is not trivial but equal to Z2. This works as follows. Consider two points identi ed by the anti-holomorphic involution on X3 over the origin of R. Any path connecting two such points becomes a closed loop in the quotient and, as the involution acts freely, cannot be homotopic to a point. If X3 is simply connected, all such loops are homotopic, so that 1((X3 R)=Z2) = Z2, which implies that Z only has holonomy group SU(3) o Z2, i.e. is a barely G2 manifold. On Z, such loops give rise to a non-trivial fundamental group, which is Z2 as well, so that Z does not have the full holonomy group Spin(7). From the point of view of physics, compacti cation on such `barely' Spin(7)-manifolds does not give rise to extended supersymmetry as there is still only one covariantly constant spinor. In fact, it is not hard to see that Z has holonomy group SU(4) o Z2 in this case: the holonomy group SU(4) of Z+ and the holonomy group SU(3) o Z2 of Z share a common SU(3), the holonomy group of X3. We can compute the cohomology groups of Z in terms of the cohomology groups of Z and the pull-backs to X3 by using the Mayer-Vietoris exact sequence, which implies that where can be expressed in terms of the restriction maps Hi(Z; Z) = ker i coker i 1 ; i : Hi(Z+; Z) Hi(Z S1; Z) ! Hi(X3 S1; Z) +i : Hi(Z+; Z) ! Hi(X3 i : Hi(Z ; Z) ! Hi(X3; Z) : S1; Z) Note that both i and i 1 feature in i due to the product S1. (4.10) (4.11) (4.12) Under a few assumptions, which are met in the examples to be discussed later, we can now explicitly work out the various contributions to H (Z). First of all, let us assume that the acyl Calabi-Yau four-fold Z+ can be constructed from an algebraic four-fold Z~ which is bered by Calabi-Yau three-folds by excising a ber X3. Furthermore, let us assume that the images of +2 and +4 are surjective and that b3(Z+) = b5(Z+) = 0. As noted above, we assume that we can construct Z as (a resolution) of the quotient (X3 Rt)=Z2 by an anti-holomorphic involution. The resolution must be such that it preserves a metric of holonomy G2 on Z and the property of Z being asymptotically cylindrical. We will denote the numbers of even/odd classes of X3 under this Z2 by bie and bio, respectively. Finally, the kernels of the restriction maps i and their ranks are abbreviated as N i ker i n i jN i j : With this notation and under the assumptions we have made, (4.10) implies that b1(Z) = 0 b2(Z) = n2+ + n2 + be2 b3(Z) = n 2 + n3 b5(Z) = n 4 + n5 b6(Z) = n6+ + n5 + bo4 b7(Z) = 0 : b4(Z) = n 3 + n4 + n4+ + bo2 + bo3 + be3 + be4 (4.13) (4.14) This can be seen as follows. For H1(Z), the unique class in H1(X3 S1) is in the image of 1 and solely originates from H1(Z coker( 0) = 0 implies b1(Z) = 0. S1). Hence j ker( 1)j = 0, which together with Let us now consider b2(Z). As the unique class in H1(Z S1) is in the image of 1 we nd that coker( 1) = 0 so that H2(Z) = ker( 2) follows. In turn, ker( 2) has three di erent contributions: ker( 2) = ker( +2) ker( 2 ) im( +2) \ im( 2 ) : (4.15) As +2 is surjective by assumption, the last term is simply given be2 and we recover the expression in (4.14). The computation for b3(Z) is made particularly simple by our assumptions. As contribution to b3(Z) is from ker( 2 ) ker( 3 ) = N 2 N 3 . j coker( 2 )j = 0, it only receives a contribution from ker( 3). As b3(Z+) = 0, the only For b4(Z), all potential contributions are non-trivial. First of all, there is a contribution n4+ from ker +4, as well as n3 + n4 from ker 3 + ker 4 . Furthermore, there are bo2 classes in coker( 3), which correspond to three-forms with one leg on the product S1 of the cylinder region, and bo3 classes in coker( 3) which correspond to three-forms purely on X3. Finally, there is a contribution (im +4) \ (im 4 ) (im 3 ) ; (4.16) which simply has dimension be4 + be3 as +4 is surjective. Together, all of these contributions give the expression for b4(Z) in (4.14). The computation of b5(Z) again only receives a contribution from ker 5 as coker 4 is empty by assumption. Furthermore, 5 is only non-trivial on Z S1, so that j ker 5j = n The computation of b6(Z) is similar to b2(Z), it receives a term bo4 from coker( 5), as well as terms n5 and n6+ from ker( 6). Note that N 6 = 0 as Z is a G2 manifold. Finally, coker( 6) = 0 and ker( 7) = 0 gives b7(Z) = 0. The various contributions n i and hie; hio are not all independent and their relations make Poincare duality for (4.14) manifest. For anti-holomorphic involutions of a CalabiYau three-fold X3, the Lefschetz xed-point theorem combined with Poincare duality implies that bo2 = be4 and be2 = bo4. Furthermore, Poincare duality on the compact G2 manifold (X3 S1)=Z2 implies that n2 = n 5 and n3 = n4 . Together with n2+ = n6+, these relations imply that bi(Z) = b8 i(Z) in (4.14). The number of deformations of the Ricci- at metric is given by b4 + 1 for manifolds of Spin(7)-holonomy. Using (2.4), this can be written as (4.17) (4.18) (4.19) which becomes 1 3 1 3 b4 + 1 = 8 + 2 b2 + b3 + b4 ; b4 + 1 = 8 + n3 + 2 + n4+ n2+ + 2bo2 be2 + be3 + bo3 in our GCS-construction. The Euler characteristic of Z is given by (Z) = 2 + 2b2(X3) + b3(X3) + n4+ + 2n2+ ; and only depends on the data of Z+, as expected. Even though the topology of the resolutions of Joyce's Spin(7)-orbifolds discussed in section 3 is most conveniently computed otherwise, one can also use their decomposition as GCS Spin(7)-manifolds together with the relations of this sections to nd their topology from resolutions of two building blocks Z . 4.4 3d eld theory and sectors of enhanced supersymmetry In compacti cations of M-Theory on a smooth Spin(7)-manifold, the low energy e ective theory at the classical level gives a 3d N = 1 supergravity theory with nv = b2(Z) massless U(1) vector multiplets and nr = b3(Z)+b4 (Z)+1 massless uncharged real scalar multiplets. As the Spin(7)-manifolds considered here are glued from pieces with holonomy SU(4) and G2, respectively, we expect to nd subsectors of enhanced supersymmetry in our spectrum, similar to the observations made in [11] regarding TCS G2 manifolds, to arise from localized forms in the building blocks. Consider rst the multiplets arising from localized forms on Z S1. By (4.14), each two-form in N 2 gives rise to both a two-form and a three-form on Z, which combine to form the bosonic eld content of a 3d N = 2 vector multiplet. Furthermore, each three-form in N 3 gives rise to a real scalar due to its appearance in the formula for b3(Z) in (4.14). As for compacti cations of M-Theory on a G2-manifold times a circle, the moduli associated to these three-forms pair up with deformations of the metric to form 3d N = 2 chiral multiplets: this can be seen from (4.18), which shows that each three-form in N 3 corresponds to an anti-self-dual four-form, i.e. a deformation of the metric. Let us now turn to multiplets which arise from localized forms on Z+. Each two-form of Z+ in N+2 gives rise to a 3d N = 1 vector multiplet due to its contribution to b2(Z). As for compacti cations of M-Theory on Calabi-Yau four-folds, we expect this degree of freedom to pair up with a real scalar to the bosonic degrees of freedom of a 3d N = 2 vector multiplet. This degree of freedom must originate from an anti-self-dual four-form in N+4 . Furthermore, we expect the remaining anti-self-dual four-forms in N+4 to appear pairwise, so as to combine into 3d N = 2 chiral multiplets. To see how this comes about, we need to exploit the fact that Z+ and its compactication Z~+ are Kahler manifolds and carry a Lefschetz SU(2) action. The upshot is that the anti-self-dual four-forms on Z~+ are given by the four-forms in H3;1(Z~+) H1;3(Z~+), together with four-forms of the type !k ^ J , where J is the Kahler form on Z~+ and the !k are (1; 1) forms on Z~+ such that !k ^ J 3 = 0. As every two-form in N+2 gives a four-form in N+4 upon wedging with J , we can hence associate an anti-self-dual four-form in N+4 on Z+ to every element of N+2 (modulo the (1; 1) forms in the image of +2). The remaining anti-self-dual four-forms in N+4 then have to appear pairwise, as they must correspond to H3;1(Z~+) H1;3(Z~+). As we can think about the forms in N+4 as being localized on the acyl Calabi-Yau four-fold Z+ far away from the gluing region, (anti) self-duality on Z+ will imply (anti) self-duality of the corresponding forms on Z. Note that this argument precisely re ects how these degrees of freedom originate in Physics. For two-forms in N+2 and self-dual four-forms in N+4 , there are deformations of the Ricci- at metric of Z+ (Kahler and complex structure deformations), which do not alter the cylindrical region in which Z+ is glued to Z S1. As we expect the Ricci- at metric of Z to be well approximated by the Ricci- at metrics of Z+ far away from the gluing region, these become deformations of the Spin(7)-manifold Z if we stretch the neck region su ciently long. The metric deformations associated with N+2 and N+4 , together with the 3d vectors originating from the C- eld on N+2 , only see the geometry of a Calabi-Yau fourfold and hence give rise to a subsector with 3d N = 2 supersymmetry. As the property of forms being (anti) self-dual is a topological constraint, the counting and identi cations we have performed will persists throughout the moduli space of Z, so that the subsectors with enhanced N = 2 supersymmetry in 3d will persist. We have summarized the result of this discussion in table 1. The remaining moduli of M-Theory on the GCS Spin(7)-manifold Z do not appear in multiplets with enhanced supersymmetry, as they are associated with the gluing along X3 S1. These moduli are, however, in one-to-one correspondence with the geometric deformations of the Calabi-Yau three-fold X3 in the neck region, i.e. Kahler and complex structure deformations of X3. While compacti cation of M-Theory on X3 S 1 S1 results in a theory with N = 4 supersymmetry in 3d, these multiplets are truncated to N = 1 for MTheory compacti ed on Z. Similar to the case of TCS G2 manifolds [11], the eigenvalues unHJEP06(218)3 N = 2 vector N = 2 chiral N = 4 vector N = 2 vector N = 2 chiral Origin N+2 h3;1(Z~+) b2(X3), b3(X3) N 2 N 3 multiple from the neck region is only massless in the in nite neck limit and away from this becomes der the Laplace operator of the appropriate non-harmonic forms needed to lift these N = 1 multiplets to N = 4 multiplets go to zero in the limit in which the neck region is stretched. GCS Spin(7)-manifolds as quotients of CY4 Our construction has a natural connection to Spin(7)-manifolds constructed as quotients of Calabi-Yau four-folds by anti-holomorphic involutions, however we will see that there are key di erence to the previous constructions in [18, 20]. Consider a compact Calabi-Yau four-fold X4, which is bered by Calabi-Yau three-folds over a base P coordinates [z1 : z2] and let z = z1=z2. Let us denote the ber over a point p of the P1 base by X3(p). For appropriate brations, we may then tune the complex structure (de ning 1 with homogeneous equation) of X4 such that For appropriate anti-holomorphic involutions construct an anti-holomorphic involution acting on X3(p) for all p, we may then X3(z) = X3(1=z) : = (z1 ! z2) z2 ! z1 1 Sf (X3 R)= : acting in X4. This will produce a singular Spin(7)-manifold Zs = X4= , and we will assume that it can be resolved into a smooth Spin(7)-manifold Z. In the P1 base the xed locus of this involution will be given by the circle Sf1 = fjzj2 = 1g and the quotient e ectively truncates the base P1 to a disc with boundary. We may make this disc very large and furthermore con ne all of the singular bers of the bration of X3(p) near the origin. If the bration of X3(p) over the xed Sf1 is trivial, the bers over Sf1 all become identical and we may simply denote them by X3. Cutting along a circle of xed radius now produces one half near the boundary of the disc which may be described as In the limit in which the disc is very large, a resolution of Zs to Z is equivalent to a resolution of (X3 R)= to a smooth acyl G2 manifold Z . The other half of Z near the origin of the disc does not require resolution and becomes an acyl Calabi-Yau four-fold Z+ which (4.20) (4.21) (4.22) asymptotes to X3. In the middle region, the two spaces Z+ and Z the product of an interval and X3 given above. An illustration is given in gure 3. Sf1 . This is precisely the gluing construction we have Sf1 are glued along Let us highlight two aspects of the topology of Z which are immediately recovered from this point of view. The resolution of Zs to Z only introduces new cycles sitting over a circle, so that the topological Euler characteristic of Z is the same as that of Zs. But the topological Euler characteristic of Zs is simply given by half of the Euler characteristic of X4. In fact, we may think of X4 as being glued from two copies of the acyl CalabiYau four-fold Z+, which implies that the Euler characteristic of Z is equal to that of Z+, which is the same result found from our gluing construction earlier. Furthermore, we have observed that we produce a barely Spin(7)-manifold with fundamental group Z2, whenever Z is constructed from a free quotient of X3 R by . This means that we may write Z as the free quotient X4= , so that we immediately recover 1(Z) = Z2. As we have explained, a GCS Spin(7)-manifold can be described as a resolution of a quotient of a Calabi-Yau four-fold by an anti-holomorphic involution whenever the acyl G2 manifold Z can be described as a resolution of (X3 R)= for an anti-holomorphic involution . In the absence of a di erent construction for acyl G2 manifolds, our GCS construction is hence equivalent to forming quotients of a speci c class of Calabi-Yau fourfolds. In this context, it implies that a resolution of singularities of the quotient by is already captured by resolving quotients (X3 R)= . Furthermore, it gives a completely new point of view on such geometries; it shows how to distinguish subsectors of enhanced supersymmetry and, as we shall see, allows to construct dual heterotic backgrounds on TCS G2 manifolds. Finally, the realization that many GCS Spin(7)-manifolds are simply (resolutions of) quotients X4= in fact proves that our construction produces eight-dimensional manifolds with Ricci- at metrics with holonomy group Spin(7) (or SU(4) o Z2). In [18] Joyce proposed another construction of Spin(7)-manifolds base on non-free quotients of Calabi-Yau four-folds with orbifold singularities was given. Crucially, the resulting singular Spin(7) orbifolds are taken to only have isolated singularities here, which can be resolved by gluing in appropriate ALE spaces. Even though similar in spirit, the antiholomorphic involution appearing here are either free or have xed loci of real dimension at least one, so these constructions are distinct. It seems likely, however, that there are examples for which the resolutions can be described by gluing in a acyl G2 manifold times a circle, so that there will be many Spin(7)-manifolds which can be found using both 4.6 A simple example We now construct a new Spin(7)-holonomy manifold using the GCS construction. There will be further examples in the next section, which are geared toward the application of M-theory/Heterotic duality. Consider a smooth anti-canonical hypersurface in weighted projective space P11114. The resulting space is a Calabi-Yau three-fold X1;149 with Hodge numbers h1;1 = 1 and h2;1 = 149. We now show how to construct an acyl G2-manifold Z which asymptotes to X1;149 R and an acyl Calabi-Yau four-fold Z+ which asymptotes to X1;149 S 1 R, and then glue Z+ and Z S1 to a Spin(7)-manifold Z as described above. Let us rst describe the acyl G2-manifold Z . The Calabi-Yau three-fold X1;149 admits a freely acting anti-holomorphic involution, which is obvious by choosing a Fermat type hypersurface where [x1 : x2 : x3 : x4 : x5] are the homogeneous coordinates of P11114. Letting xi ! xi then acts as a xed-point free anti-holomorphic involution, as the above equation has no solutions purely over the reals. The odd/even Hodge numbers are he2 = 0 ; ho2 = 1 ; he3 = ho3 = 149 ; and (X1;149 R)=Z2 is an acyl G2-manifold. Note that ni = 0 for all i as no resolution of singularities is required on Z . To nd an acyl Calabi-Yau four-fold, we consider an algebraic four-fold Z~+ which is a bration of X1;149 over P1 and has c1(Z~+) = [X1;149]. Such a Z~+ can easily be constructed by as the zero locus of a generic section of O(8; 1) over P11114 h1;1(Z~+) = 3, which means that n2+ = 1. Furthermore, the resulting acyl Calabi-Yau P1. Such a four-fold has four-fold Z+ = Z~+ n X1;149 has (Z+) = (Z~+) (X1;149) = 1680 = (Z). Note that we may choose an element of the algebraic family Z~+ such that the ber over some chosen point in the base P1 is described by (4.23), so that Z+ manifestly asymptotes to a Calabi-Yau three-fold which is isomorphic to the Calabi-Yau three-fold in the neck region of Z . For the resulting GCS Spin(7)-manifold constructed from this choice of Z+ and Z , it now follows that (4.23) (4.24) HJEP06(218)3 b2(Z) = 1 b3(Z) = 0 b4 (Z) = 551 b4+(Z) = 1125 : (4.25) Using (2.4), this is most e ciently calculated by using (Z) = 1680. One can also work out that n4+ = 1374, which reproduces b4(Z) = 1676 using (4.14). Due to the free action of the anti-holomorphic involution on X1;149, this example produces a barely Spin(7)manifold. By considering more general anti-holomorphic involutions and resolving the resulting singularities, similar examples with full holonomy Spin(7) can be constructed from this procedure. Hodge numbers of X4 are In agreement with our general discussion in section 4.5, Z may also be found as the free quotient of a Calabi-Yau four-fold X4 realized as a hypersurface in P11114 P1. The h1;1(X4) = 4 h2;1(X4) = 0 h3;1(X4) = 548 h2;2(X4) = 2252 (4.26) and (X4) = 3360. The free involution acts as xi ! xi z1 ! z2 z2 ! z1 ; (4.27) where [z1 : z2] are homogeneous coordinates on the base P1 and xi are the homogeneous coordinates on P11114. We may choose a smooth invariant hypersurface on which acts without xed points as 4 i (z12 + a5z1z2 + z22)x52 + X(z12 + aiz1z2 + z22)xi8 = 0 ; (4.28) for a set of real parameters ai. A little thought reveals that a one-dimensional subspace of the forms in H1;1(X4) is even under , which together with (X4) = 3360 and b3(X4) = 0 implies (4.25). HJEP06(218)3 5 M-theory on GCS Spin(7)/heterotic on TCS G2 In this section we construct a new class of GCS Spin(7) manifolds Z following our general discussion in section 4, where the building blocks are an open acyl Calabi-Yau four-fold Z+ and an open acyl G2-manifold Z (times a circle), which are both asymptote to a cylinder times the same Calabi-Yau three-fold X3. The additional input in this section is that we will consider these in the context of the duality to Heterotic strings theory on G2-manifolds, which we show to be TCS-manifolds, with building blocks X . Our strategy in nding dual compacti cations will be similar to the one we used in [12], i.e. using berwise duality between M-theory and heterotic string theory, we will identify dual pairs of building blocks, which can then be glued to nd the dual Spin(7) and G2manifolds. As this requires to carefully dissect geometries on both sides, the following discussion will unfortunately be rather technical. After motivating our construction for a dual pair of geometries, we will summarize our results in section 5.4 before checking the spectra. In our rst example, the E8 E8 gauge symmetry is broken by a (quotient of a) Wilson line, and we can easily match the light degrees of freedom of M-theory to those of the dual heterotic string theory. In general, such a match is much more di cult, as it requires to determine the number of moduli associated with a vector bundle on a G2-manifold. We present a second closely related example in which we use the proposed duality to instead predict the number of bundle moduli on the G2 manifold. We begin with a brief summary of heterotic G2-systems in section 5.1 and a discussion of the duality in section 5.2, before then constructing examples of dual pairs. 5.1 Heterotic string theory on G2-manifolds Before we begin with a detailed discussion of dual pairs, we should explain how to construct 3d N = 1 theories from heterotic string theory on a G2-holonomy manifold together with a vector bundle. Here we will discuss brie y heterotic compacti cations on G2-holonomy manifolds. A compact orientable 7-dimensional manifold J which has a Ricci- at metric g with holonomy group contained in G2 supports a closed three-form , the Hodge- -dual of which is closed as well, which can be expressed in local coordinates (in which g is the euclidean = dx123 + dx145 + dx167 + dx246 dx257 dx347 dx356 : (5.1) This three-form and its Hodge- -dual are preserved by the action of G2. Besides having a Ricci- at metric, G2-manifolds support a single covariantly constant spinor. In contrast to Spin(7)-manifolds, a necessary and su cient condition for the holonomy group of a manifold J with a Ricci- at metric g and hol(g) G2,to be equal to G2 is that the fundamental group of J is nite rather than trivial. We will refer to manifolds which support a single covariantly constant spinor, but have holonomy group SU(3) o Z2 as barely G2-manifolds. The only non-trivial Betti numbers of a G2-manifold are b2 and b3 as b1 = 0, and the moduli space of Ricci- at metrics for a G2-manifold has real dimension b3. For G2-manifolds, calibrated submanifolds can have dimension three (associative cycles), in which case they are calibrated by , oder dimension four (co-associative cycles), in which case they are calibrated by . Let us now discuss compacti cation of heterotic E8 E8 strings on G2-manifolds. Earlier studies of this in the context of resolutions of Joyce orbifolds T 7= have appeared in [22, 42, 43]. The most in depth analysis of the conditions that such a compacti cation needs to satisfy were derived in [44] and are given in terms of a heterotic G2-system. This is comprised of a G2-holonomy manifold and its tangent bundle (J; T J ) as well as an E8 E8vector bundle with connection (V; A). The curvatures R and F of T J and V satisfy Furthermore there can be NS5-branes wrapped on associative three-cycles, subject to the anomaly condition R ^ = 0 ; F ^ = 0 : 0 4 (Tr F ^ F Tr R ^ R) = dH ; (5.2) (5.3) where dH =[NS5] the Poincare dual of the homology class corresponding to associatives wrapped by NS5-branes. 5.2 GCS Spin(7)-manifolds and M-theory/heterotic duality Up to this point, our construction of GCS Spin(7)-manifold was largely motivated for constructions of new geometries. An alternative motivation is to realize that these give rise to M-theory duals of heterotic strings on TCS G2-manifolds. A review of the TCS construction is given in appendix A. Let us hence consider heterotic strings compacti ed on a manifold of G2 holonomy J , together with an appropriately chosen vector bundle, giving an e ective 3d theory with N = 1 supersymmetry. Fiberwise duality to M-theory is possible if J admits a calibrated bration by three-tori T 3, which are replaced by K3 surfaces in the dual compacti cation of M-theory. As 3d compacti cations of M-theory with N = 1 supersymmetry are found upon compacti cation on manifolds of Spin(7) holonomy, we expect the resulting geometry to be in this class. We already discussed the 7d duality between M/K3 and Het/T 3 in section 3.1. We now apply this berwise to construct dual pairs of TCS and GCS constructions, in cases when there is a K3- bration on the GCS Spin(7)-manifold and dually on the heterotic side there is a T 3- bered TCS G2-manifold. In this section we make some initial observations about the general features of these compacti cations. If J is a TCS G2-manifold, mirror symmetry for Type II strings implies the existence of a calibrated T 3- bration, which realizes a mirror map in the spirit of Strominger-YauZaslow (SYZ) [45] via three T-dualities [13]. Such a bration exists if one of the two building blocks, say X+, used in the TCS-construction of J is bered by K3 surfaces S+, which in turn admit an bration by an elliptic curve E E ,! S+ ! P1 : (5.4) E 1 Sb; becomes the SYZ ber of X . This implies that X+ admits an elliptic bration with ber E as well, and the T 3- bers of J restricted to E 1 Se;+ on X+ Se1;+. The Donaldson matching (A.1) then maps E to a sLag ber of the K3-surface S , and Se;+ to Sb1; . Going away from the neck region of X , 1 Next let us consider how this maps under Heterotic/M-theory duality: HJEP06(218)3 X+: we replace the T 3- bration given by E 1 Se;+ on X+ Se1;+, with a K3 surface S+. Then only two out of the three forms of the hyper-Kahler structure of S+ have a non-trivial bration over the base. In particular, we may choose to described the resulting K3 bration algebraically and end up replacing the product of the acyl 1 Calabi-Yau three-fold X+ with Se;+ by an acyl Calabi-Yau four-fold Z+. X : here, the SYZ- ber of X is bered such that application of Heterotic/M-theory duality leads to a K3- bration in which all three (1,1)-forms of the hyper-Kahler structure have a non-trivial variation over the base. As in the 4d N = 1 duality between heterotic strings and M-theory, we replace a Calabi-Yau three-fold by a G2manifold. In the present setup, this means replacing an acyl Calabi-Yau three-fold X with an acyl G2-manifold Z . In the cylinder region of J , where J is di eomorphic to a product S simply nd X3 1 S S 1 S 1 I for a K3-surface S with a calibrated T 2- bration,2 we I by application of Heterotic/M-theory duality. We hence nd the statement that M-theory duals of heterotic strings on TCS G2-manifolds are compacti ed on Spin(7)-manifolds, which allows a construction as proposed in our GCSconstruction: an acyl Calabi-Yau four-fold Z+ and an acyl G2-manifold Z are glued along a cylinder times a Calabi-Yau three-fold. times an S1 In our discussion, we have so far ignored that the heterotic compacti cation comes equipped with a vector bundle on J , and we need to identify what this is mapped to on the dual M-theory geometry. For X+, where only E varies non-trivially, replacing X+ 1 Se;+ by Z+ is simply a non-compact version of the 3d duality between heterotic strings and M-theory on Calabi-Yau four-folds. We can hence apply the usual logic of how bundles constructed from spectral covers are translated in F-theory [46]. Restricting J to X Se1; , holomorphic vector bundles on a (non-compact) Calabi-Yau three-fold get translated to the geometry of a (non-compact) G2-manifold, similar to discussed in [12]. Gluing Z+ and Z S1 to a compact Spin(7)-manifold Z then implies that the geometry of Z determines a heterotic G2-system on the G2-manifold J . It is hence tempting to construct such vector bundles on twisted connected sum G2-manifolds by appropriately gluing 2Depending on a choice of complex structure, this is an elliptic or sLag bration. holomorphic vector bundles on the building blocks X , see [47] for a concrete realization of this idea. In section 5.5 we will use this logic to count the number of bundle moduli on a TCS G2-manifold and con rm a matching with the degrees of freedom of the dual M-theory compacti cation on a GCS Spin(7)-manifold. Duality for the building blocks Before heading to the full construction of M-theory/Spin(7) and Heterotic/G2 we rst explore the dual pairs that are relevant for each building block. Recall: GCS Spin(7)-manifold Z is built out of an acyl CY4 building block Z and an acyl G2-building block Z+. TCS G2-manifold J is build out of two acyl CY3, X . We will now construct dual pairs for CY4 and G2 and correspondingly on the heterotic side CY3 in sections 5.3.2 and 5.3.3. These will then be used as building blocks to construct the GCS/TCS dual pairs. As a preparation we rst need to construct pairs of G2 and CY3 and duals for M/het duality. Geometric preparation: dual pairs of G2 and CY3 In this section we discuss a well-suited example for building the G2 building block of the GCS-manifold, by consider M-theory on G2-manifold and the heterotic on CY3 compacti cation. Similar to the strategy used in [16], we can nd such a pair by forming an appropriate quotient of a dual pair of 4d N = 2 compacti cations, which are M/X43;43 S1 ! Het/K3 Th2 : (5.5) (5.6) On the heterotic side we choose a trivial vector bundle on the K3 surface and break E8 E8 to U(1)16 by introducing Wilson lines on an the Th2 factor. Tadpole cancellation then necessitates the introduction of 24 NS5-branes wrapped on T 2. Such a compacti cation of the heterotic string theory on K3 Th2 is dual to M-theory compacti ed on X43;43 S1, where X43;43 is a CY3-fold, chosen to be a bration of K3-surfaces with polarizing lattice N = U ( E8) ( E8) over P1. The resulting CY3 is constructed as a hypersurface in a toric variety A associated to the pair of re exive polytopes 0 = BBB 3 3 1 1 12 0 12 CA = BBB 1 1 1 0 6 6 and has h1;1(X43;43) = h2;1(X43;43) = 43. Besides the 4d N = 2 supergravity multiplet (which also contains a graviphoton), both of these compacti cations have 43 N = 2 vector multiplets, 43 hypermultiplets and one universal hypermultiplet. We can now quotient both theories in (5.5) by an involution to nd dual compacti cations of M-theory on a G2-manifold and heterotic strings on a Calabi-Yau three-fold with a vector bundle. This truncates the light excitations to be described by a theory with 20 N = 1 U(1) vectors and 67 N = 1 chiral multiplets. On the type IIA side we act with a free involution Zo2 to produce a G2-manifold3 M as M = (X43;43 S1u)=Zo2 ; (5.7) where Zo2 sends the coordinate u ! u and acts on X43;43 as a antiholomorphic involution. This anti-holomorphic involution is chosen such that it sends all of the homogeneous coordinates of the ambient toric space A for X43;43 to their complex conjugate and furthermore swaps any homogeneous coordinate of A associated with the lattice point (a; b; c; d) on with the homogeneous coordinate associated with the lattice point (a; b; c; d) on . Some thought reveals that this is a symmetry of A and that one can choose the hypersurface equation of X43;43 such that Zo2 becomes a symmetry of X43;43 (without enforcing singularities) and acts as Zo2 : ! , as appropriate for an antiholomorphic involution. There are 47 divisors on X43;43, which descend from A by intersecting X43;43 with the vanishing loci of the homogeneous coordinates of A, out of which 43 are linearly independent. There are 21 pairs of homogeneous coordinates fzi ; i = 1 21g on which Zo2 acts as HJEP06(218)3 Zo2 : Zo2 : z i ! z i wj ! wj : and the remaining 5 homogeneous coordinates fwj ; j = 0 5g are acted on as Such an involution acts freely, because a xed point would require that zi to the C action on the homogeneous coordinates of A. For our choice of zi actions imply that xed points require zi+ = zi = 0. For our choice, such intersection vanish for any pair zi , so that there are no xed points. Out of the four linear relations among the divisors of A, only one involves the 42 divisors associated with fzi ; i = 1 21g and the remaining three involve the fwj ; j = 0 combinations of such divisors and we nd that4 g 5 . Hence there are precisely 20 even As Zo2 is an antiholomorphic involution, we also have that b3 (X43;43) = 44, from which h1+;1(X43;43) = 20 h1;1(X43;43) = 23 : b2(M ) = h1+;1(X3) b3(M ) = h1;1(X3) + b3+(X3) = 67 ; the P 1 base of the K3 why there are 8 + 12 = 20 even classes in b2(X3). which gives the number of N = 1 U(1) vector and chiral multiplets for compacti cation of M-Theory on M . 3As the involution we consider acts freely, this produces only a `barely' G2-manifold, the holonomy group of which is only the subgroup SU(3) o Z2 of G2. 4Another way to see this is as follows: the three-fold X43;43 is bered by elliptic K3 surfaces with two II over P1 and this involution acts as the antipodal map of the base of this elliptic bration (another way to see why it acts freely) in particular identifying the two II bers. It furthermore acts (non-freely) on bration and identi es the 24 reducible K3 bers on X3 pairwise. This reproduces (5.8) (5.9) zi , up the C (5.10) (5.11) with Hodge numbers h1;1(X11;11) = h2;1(X11;11) = 11. Furthermore, it acts on the bundle data by twisting the two Wilson lines such as to break E8 E8 to U(1)8. We can think about the surviving bundle data as an extension of a line bundle associated with a Wilson E8 to U(1)8, and there are eight complex degrees of freedom specifying The orbifold furthermore identi es the 24 NS5-branes on Th2 pairwise,6 so that the number of 4d N = 1 U(1) vector and chiral multiplets on the heterotic side is given by z ! On the dual heterotic side, the corresponding involution Z2h acts as a combination of z (on the Th2 factor) together with the Enriques involution5 on the K3 surface to produce a Calabi-Yau three-fold X11;11 = (K3 Th2)=Z2h (5.12) (5.13) (5.14) (5.15) (5.16) nv = 8 + 12 nc = 1 + 2 11 + 3 12 + 8 in perfect agreement with the dual M-theory on the G2-manifold M . With this preparation in hand we can now study the duality for the building blocks. The acyl G2-manifold and its dual acyl CY3 We can turn X11;11 and M in the last subsection into a dual pair of an acyl Calabi-Yau three-fold and an acyl G2-manifold by cutting each of them in half. Let us start with M in (5.7). By cutting along the middle of the interval remaining of the circle S1u, M is turned into an acyl (barely) G2-manifold Z which we may represent by Z = (X43;43 Ru) =Zo2 ; where Zo2 acts as before on X43;43 and as u ! fold of Z is X43;43 and there are b2+(X43;43) = 20 classes in the image of the restriction map from H2(Z ; Q) to H2(X43;43; Q), as well as b3+(X43;43) = 44 classes in the image of the restriction map from H3(Z ; Q) to H3(X43;43; Q). There are no elements in the kernels u on Ru. The asymptotic Calabi-Yau threeof these maps. trivial) K3 bration On the heterotic side, Calabi-Yau three-fold X11;11 (5.12) carries a natural (almost K3 ,! X11;11 ! P1h ; where the base P1h is found as the quotient of T 2. Over four points on this P1h, the K3 h ber is truncated to an Enriques surface, but is constant otherwise. Note that this way of thinking reproduces the fact that the Euler characteristic of X11;11 vanishes from (X11;11) = 24 (2 4) + 4 12 = 0 ; 5The Enriques involution is the unique xed-point free non-symplectic involution on a K3 surface and is identi ed by the invariants (r; a; ) = (10; 10; 0) in the classi cation of [48]. 6Equivalently, one can compute that ch2(X11;11) = 12[Th2]. where L(2) denotes a lattice L with its inner form rescaled by a factor of two. The expression for N (X ) is nothing but the even sublattice of H2(K3; Z) under the Enriques involution. Furthermore, X may be compacti ed to a compact building block X~ in the sense of [5] by gluing in a single K3 ber. It now follows from a computation as above that (X~ ) = 24, which together with b2(X~ ) = 11 (and b1(X~ ) = 0 implies that b3(X~ ) = 0. This is all of the data we will need in using X as a building block for a G2-manifold. Note that we have indeed decomposed M and X11;11 by cutting along the same S1. This becomes particularly clear, by noting that these dual compacti cations descend from a compacti cation to ve dimension, in which M-theory is put on X43;43 and heterotic string on K3 S1. The dual models we have constructed are then found by compactifying both side on a further S1u and quotienting by Z2. It is the quotient of S1u, which becomes one of the two circles on Th2, which is cut in half to produce Z and X , respectively. The acyl CY4 and its dual acyl CY3 We now construct a dual pair of an acyl Calabi-Yau four-fold Z+ (as a building block for the Spin(7) GCS) and the heterotic dual acyl Calabi-Yau three-fold X+ (as a building block for the G2 TCS). As we want Z+ to asymptote to X43;43 so that we can glue Z+ with Z S1, we construct (5.17) (5.18) (5.19) (5.20) so that the algebraic three-fold Z~+ satis es Z+ = Z~+ n X403;43 ; X43;43 ,! Z+ ! P1 ; ~ c1(Z~+) = [X403;43] : Note that we may also construct Z+ by cutting a Calabi-Yau four-fold X4, which is a bration of X43;43 over P1 in two halves. This four-fold is a toric hypersurface de ned by the pair where we have excised the four bers which are Enriques surfaces from the base P1h in the rst term and added them back in for the second. If we now cut P1h into two halves, each of which is C and contains two of the Enriques bers over it, we produce an acyl Calabi-Yau as the TCS G2 building block K(X ) = ; ( E8(2)) ; 0 B = BBB B 0 = BBB 1 1 0 0 6 6 6 6 1 C C h1;1(X4) = 68 ; h3;1(X4) = 292 ; h2;2(X4) = 1484 ; (5.21) so that (X4) = 2208. As a generalization of [9] to four-folds, Z~+ can also be constructed from one of the two isomorphic tops in corresponding to the bration by X43;43. Either way, it follows that the Euler characteristic of Z+ is (Z+) = 1104, half of the Euler characteristic of X4. The restriction maps i : Hi(Z+; Q) ! Hi(X403;43 S1; Q) are fairly easy because all of the divisors and none of the three-cycle on X43;43 descend from Z+. Furthermore, there are 12 divisor classes on Z+ associated with reducible bers of the bration by X43;43. Hence HJEP06(218)3 we conclude j im( 2)j = 43 ; j im( 3)j = 0 ; j ker( 2)j = 12 ; j ker( 3)j = 0 ; j coker( 2)j = 0 ; j coker( 3)j = 131 : (5.22) This will be all the data from Z+ we need in the following. Let us now nd the dual geometry for the heterotic string. As Z+ and X4 are bered by elliptic K3 surfaces, this is fairly straightforward, as we simply need to apply the known rules of heterotic-F-theory duality. In particular, Z+ and X4 are bered by elliptic K3 surfaces with Picard lattice U ( E8)2, so that the dual heterotic theory has no E8 E8 vector bundles turned on. Furthermore, Z+ and X4 are elliptic brations over (blow-ups of) C P 1 P1. These blowups capture the presence of the NS5-branes needed on the heterotic side to satisfy the Bianchi identity. Hence the dual heterotic geometry dual to is X4 given by an elliptically bered Calabi-Yau X3;243 with base P The three-fold X3;243 is constructed from a pair of re exive polytopes with vertices P1 times a circle S1. 0 = BBB 3 3 3 1 1 0 0 0 1 2 1 C ; 1 0 = BBB 1 1 0 6 0 0 1 1 1 1 0 6 0 6 6 C : (5.23) and it follows that X3;243 has Hodge numbers (h1;1; h2;1) = (3; 243). The dual acyl CY building block X+ of the acyl Calabi-Yau four-fold Z+ is constructed using by cutting X3;243 along an S1 in the base of its K3 bration. Equivalently, we may glue in a K3 ber to compactify X+ to a building block X~+ which can be constructed using the methods of [9] from a pair of projecting tops 0 = BBB 3 3 3 C 1 1 0 CA 0 0 1 C ; 0 = BBB 1 1 0 6 0 0 1 1 6 6 1 1 6 6 1 1 6 0 1 1 6 0 1 1 1 6 6 1 1 6 6 1 1 1 C 0 C : (5.24) (5.25) 0 1 0 0 1 0 0 0 It follows that N (X+) = U K(X+) = ; h2;1(X~+) = 112 : Z = Z+ [ (Z S1) ; where Z+ \ (Z S1) = X3 S1 = and X3 = X43;43. It immediately follows that (Z+) = 1104. The remaining Betti numbers are found from the Mayer-Vietoris b2(Z) = j ker 2j + jcoker 1j = (20 + 12) + 0 = 32 ; where we have used that 20 classes in h1;1(X43;43) are in the common image of the restriction maps 2 : H2(Z+) H2(Z H2(X43;43) is 12-dimensional. Furthermore, S1) ! H2(X43;43 S1) and that the kernel of +2 : H2(Z+) ! b3(Z) = j ker 3j + jcoker 2j = 0 ; as all of the classes in h1;1(X43;43) are in the image of +2, no class in b3(X43;43) is in the image of +3 and the restriction maps 3 have no kernels. We can now use (2.9) to nd b4 (Z) = 327 ; Note that there are Wilson lines breaking E8 E8 to U(1)16 on the product S1 on the heterotic side, which are dual to the volumes of the divisors associated with the ( E8) 2 in the Picard lattice of the K3 bers of X4. First example of dual pairs: new Spin(7)- and its dual G2-manifold Above we have constructed a pair of an acyl Calabi-Yau four-fold Z+ and an acyl G2manifold Z which both asymptote to (a cylinder/an interval times) the same Calabi-Yau three-fold X43;43, together with the two dual acyl Calabi-Yau three-folds X+ and X together with the corresponding bundle data on the heterotic side. We will now glue Z+ with Z S1 to a GCS Spin(7)-manifold Z and X S1 to a TCS G2-manifold J and verify that the light elds in the e ective eld theories indeed agree. The M-theory on GCS Spin(7)-manifold As detailed in section 4, the Spin(7)-manifold Z is constructed as and it follows that there are nc = 360 real degrees of freedom originating from the metric and three-form for compacti cations of M-theory on Z. Out of these, 32 sit in 3d N = 1 vector multiplets. Furthermore, a consistent compacti cation requires the introduction of 1104=24 = 46 M2 branes. Each of these contributes a further 8 real moduli from moving on Z. Alternatively, this Spin(7)-manifold can also be obtained as a free quotient of the Calabi-Yau four-fold X4 speci ed as a toric hypersurface by (5.20). The dual heterotic model on TCS G2-manifold The dual heterotic side has various components: geometry, vector bundle and NS5-branes. We will determine all the dual data, and compare with the spectrum obtained in the M-theory compacti cation | and nd agreement. (5.26) (5.27) (5.28) (5.29) Geometry. M-theory on Z is dual to heterotic on a G2-manifold J which can be obtained by ber-wise duality as J = (X+ Se1;+) [ (X Se1; ) ; where X are the acyl Calabi-Yau three-folds introduce in sections 5.3.3 and 5.3.2 above. Recall that the lattices N (X ) are E8(2) ; h2;1(X ) we hence immediately nd [6] and there is a matching such that N (X+) \ N (X ) = ; . Together with K(X ) and (5.30) (5.31) (5.32) b2(J ) = K(X+) + K(X ) = 0 b2(J ) + b3(J ) = 23 + 2 (K(X+) + K(X )) + 2 h2;1(X+) + h2;1(X ) = 247 ; giving b3(J ) = 247. We hence nd 247 real degrees of freedom from the metric of a heterotic compacti cation on J and no moduli associated with the B- eld. Bundle data. Recall from section 5.3.1 that the heterotic compacti cation on X11;11 = X dual to M-theory on M = Z has eight complex bundle degrees of freedom, associated with a bundle which is an extension of appropriate Wilson lines on the double cover of X11;11. The origin of the corresponding degrees of freedom in the dual M-theory suggests to associate 4 complex bundle degrees of freedom with each of the S1s in the K3 Th2 double cover of X11;11. For X , one of these two S1s has disappeared, so that we are left with 8 real bundle degrees of freedom there. In particular, these give the non-trivial holonomies in E8 1 E8 along Sb; in the base C of the K3 G2-manifold, these get mapped to Wilson lines on Se1;. Note that this is precisely what is expected for the duality between M-theory on the Calabi-Yau four-fold X4 and heterotic strings on X3;243 Se1: the non-zero volumes of the divisors on Z+ associated with the ( E8) 2 in the Picard lattice of the K3 bers of Z+ are mapped to Wilson lines on Se1+ on the heterotic side. We hence conclude that J carries a vector bundle V inherited from a bration on X . In the TCS bundle is captured by the volumes of the curves in ( E8 2) of the K3 bers of Z . bundle on X which breaks E8 E8 ! U(1)8 and has eight real moduli. The data of this NS5-branes. Furthermore, our compacti cations of the heterotic string require the addition of NS5-branes for consistency as the gauge bundle V does not have a second Chern character. We will infer the NS5-branes which we need to include by exploiting piecewise duality. In particular, we can think about a representative of the rst Pontrjagin class of J as being glued from representatives of c2(X~ ). The second Chern class of X~+ is c2(X~+) = 46HH^+ + 46H+ + + 23H^+ + + 11 +2 = 46H+H^+ + 24H+ + + 12H^+ + ; (5.33) where H^+ is the divisor class corresponding to xing a point on the base of the K3 bration on X~+ and H+ is the class corresponding to xing a point on the other P1 in the P1 P^1 base of the elliptic bration on X~+. The divisor class + is represented by the section of this elliptic bration and these classes satisfy the relation that c1(X~+) = [H^+]. The 46 NS5-branes on the curve HH^ are wrapped on the elliptic ber of X+ times 1 the auxiliary S+;e multiplying X+ in the construction of J . As such, they are wrapped precisely on the conjectured ber of the T 3 bration of J used for the duality and we can associated them with the 46 M2 branes present on the M-theory side. Let us con rm this by counting the number of moduli. Each of these NS5-branes has 4 real degrees of freedom by displacing the elliptic ber on X+. Furthermore, there is one real degree of freedom from the scalar in the tensor multiplet (the position along the 11th direction in heterotic M-theory) and three real degrees of freedom from the self-dual two-form B on the b1(T 3) = b2(T 3) = 3 cycles of the T 3 they are wrapped on. This makes 8 degrees of +( + + 2H+ + H^+) = 0. Note freedom matching the counting for the M2 branes. The 12 NS5-branes on H^+ + are points on the P1 base of the K3 bration of X~+ so that they are wrapped on a three-manifold with the topology S2 S1 in J . Within X+, each such NS5-brane has 2 real moduli associated with displacement,7 one modulus associated with the worldvolume scalar , and one modulus from the self-dual two-form B. At least in the Kovalev limit, we are hence led to associate 4 real moduli with each such NS5-brane. Finally, there are the 24 NS5-branes in the class H+ + of X~+ Se1;+. On X+ = X~+ n S+0, they are wrapped on the whole of the open base C 1 Se;+ of the K3 bration on X+ Se1;+. Similarly, there are 12 NS5-branes wrapped on a double cover the open base C K3 bration on X 1 Se; as we have seen in section 5.3.1. Note that this sector of NS51 Se; of the branes becomes 24 copies of the T 2 factor times the interval in (X+ Se1;+) \ (X Se1; ) = I T 2 K3. On the G2-manifold, these NS5-branes can hence be joined to form 12 irreducible NS5-branes. This has several e ects. First of all, the relative positions of the 24 branes on X+ S1+;e, which are points on the K3 ber S+ of X+ , are pairwise xed to be symmetric under the Enriques involution acting on S . Second, each such pair of NS5-branes only has two real moduli of deformation. Because they are wrapped on the C-base of the K3 bration on X+ and the elliptic bration of X+ is non-trivial there, they cannot be displaced in the direction of the elliptic curve of X+. Third, the two branches of each pair of NS5-branes are swapped when encircling two special points in the base of the K3 bration on X , so that we should think of each such pair of NS5-branes as a single NS5-brane wrapped on a three-manifold L which is a double cover of S3 branched along two unlinked S1s. We conclude that there are 12 NS5-branes wrapped on three-manifolds L inside J with two real deformations each. It is not hard to see that b1(L) = b2(L) = 1, so that each of these NS5-branes is associated with 4 real degrees of freedom: 2 real moduli of displacement together with 2 real moduli from and B adding up to 4 real moduli each. Counting of degrees of freedom. We are now ready to count the number of degrees of freedom on the heterotic side. To do so, we can neglect the 46 NS5-branes wrapped on the T 3 ber of J , as these are mapped to M2 branes in M-theory and we have already matched 7The associated holomorphic curves which lift to associatives are xed to lie on the section of the elliptic bration of X+. their degrees of freedom. Summarizing the di erent contributions discussed above, the remaining light elds for heterotic strings on J are hence the 247 moduli of the Ricci- at metric on J , the 8 real degrees of freedom in the bundle V , 8 U(1) vectors which remain massless in E8 E8, 24 NS5-branes with 4 real moduli each, and the dilaton, giving a total of 360 real degrees of freedom. Out of these, there are 32 3d N = 1 vector multiplets, originating from the 8 surviving U(1)s together with the 24 NS5-branes wrapped on threecycles with b1 = 1. We have hence veri ed that the number of light elds between heterotic string theory on J and M-theory on Z precisely agrees. Second example of dual pairs Let us now make consider a variation of the previous example and work out the topology of the Spin(7) associated with putting an E8 E8 vector bundle on J . To describe such a situation, we intend to replace Z and Z+ by an acyl G2-manifolds and acyl CY4 originating from a K3 bration with Picard lattice U instead of U ( E8) 2. The acyl Calabi-Yau four-fold Z+ Following the usual rules of heterotic-M-theory (F-theory) duality Z+ is now found as one half of a compact Calabi-Yau four-fold X4 described by a generic Weierstrass elliptic bration over P1 P 1 P1 and Z+ \ (Z S1) = X3;243 S 1 I. Here, X3;243 has already appeared in section 5.3.3 and X4 is found from the pair of re exive polytope: 0 B = BBB B 0 B = BBB B 1 0 0 0 0 2 1 0 0 0 0 1 0 0 0 1 1 0 0 0 3 3 3 3 C 0 1 0 0 CC 0 0 0 1 CA 1 0 1 0 C 1 1 1 1 C C 6 6 CC ; (5.34) S 1 I and the (5.36) which implies and (Z+) = 8784. j im +2j = 3 ; j im +3j = 0 ; j ker +2j = 0 j ker +3j = 0 h2;1(X4) = 0 ; h2;2(X4) = 11724 ; (5.35) and (X4) = 17568. We can write Z+ [ Z+ = X4 with Z+ \ Z+ = X3;243 data of Z+ relevant for its use in the construction of Z are To construct Z , we rst consider M = Z = X3;243 S1u=Z2. The three-fold X3;243 was constructed as a toric hypersurface associated with the pair of re exive polytopes shown in (5.23). Its homogeneous coordinates have the toric weights Similar to our procedure in section 5.3.1 we can introduce an antiholomorphic involution on X3;243 by letting8 y x w z1 z2 z^1 z^2 3 2 1 6 4 0 6 4 0 0 1 0 0 1 0 0 0 1 0 0 1 : (y; x; w; z^1; z^2) ! (y; x; w; z^1; z^2) (z1; z2) ! ( z2; z1) : b2(Z) = b3(Z) = 0 : b4 (Z) = 2919 This involution acts as the antipodal map on the P1 with homogeneous coordinates [z1; z2] The resulting barely G2-manifold M hence has Betti numbers h1+;1(X3;243) = 0 h2+;1(X3;243) = 244 h1;1(X3;243) = 3 h2;1(X3;243) = 244 b2(M ) = 0 ; b3(M ) = 247 : We can now form an acyl G2-manifold Z and Z = S1 X3;243. The restriction maps are by cutting S1u in the middle, i.e. M = Z j im 2 j = 0 ; j im 3 j = 244 ; j ker 2 j = 0 j ker 3 j = 0 : 5.5.3 M-theory spectrum on the GCS Spin(7)-manifold From the acyl Calabi-Yau four-fold Z+ and the acyl G2-manifold Z S we can now form a Spin(7)-manifold Z = Z+ [ Z S. It follows from the restriction maps that Furthermore (Z) = 8784. Hence and compacti cations of M-theory on Z have 2920 real degrees of freedom from the metric which all sit in real multiplets, as well 8784=24 = 366 space-time lling M2-branes. Again, this Spin(7)-manifold can also be obtained as a free quotient of the Calabi-Yau four-fold X4 speci ed as a toric hypersurface by (5.34). 8In fact, X43;43 is connected to X3;243 by several singular transitions and the actions (5.8) and (5.9) imply the one given here. (5.37) (5.38) (5.39) (5.40) (5.41) (5.42) (5.43) The dual heterotic compacti cation lives on the same G2-manifold J as discussed in section 5.4.2. Here, the metric contributes 247 real degrees of freedom. Now, however, we choose to switch on a vector bundle V completely breaking E8 E8, such that V is at on the T 3 bers of J for the duality to hold berwise. It now follows from the proposed duality that such a bundle must have mV = 2920 1 = 2672 (5.44) real moduli. Furthermore, the heterotic Bianchi identity in this context must force to include 366 NS5-branes wrapped on the T 3 ber of J . We can reproduce the number of bundle moduli by studying the moduli of holomorphic vector bundles on X together with a gluing condition. For X , such bundles are inherited from E8 E8 vector bundles on K3 symmetric under the Enriques involution. From the point of view of the K3 bration on X , the bundle data are constant over the base C of X . A holomorphic vector bundle on K3 has 4 112 real moduli (they all sit in hypermultiplets for heterotic string theory on K3) and the restriction to be invariant imposes a restriction xing half, i.e. 224 real moduli of such a bundle. For X+, the number of bundle moduli can be inferred as follows. Consider heterotic strings on X3;243 = X+ [ X+. By a straightforward application of duality to F-theory (which is compacti ed on the manifold X4 discussed in section 5.5.1) immediately gives that a generic vector E8 E8 model W , which is at on the elliptic ber of X3;243, has mW = 5344 real moduli [46, 49]. An E8 E8 vector bundle on X+ \ X+ = I S 1 K3 has 4 112 real moduli, which gives the number of conditions when matching two generic E8 E8 bundles W on X . We hence expect mW+ + mW 448 = mV : (5.45) As furthermore mW+ = mW , we nd that mW+ = 2896. On the G2-manifold J , we can hence construct a suitable vector bundle V by letting V jX+ = W+. Such a bundle restricts to a vector bundle on S+ which then has to be appropriately restricted to match V jX Se1; . As we have seen VS is not a generic vector bundle on a K3 surface, but there are 224 conditions arising at it must be symmetric under the Enriques involution. Once this condition is met, VS uniquely de nes V jX Se1; . As we have seen, the Enriques involution gives 224 real restrictions, so that we conclude that mV = 2896 224 = 2672, which matches the expectation from the counting of degrees of freedom on the Spin(7)-manifold Z. Ignoring the 366 NS5-branes wrapped on the T 3 ber of J , all of the moduli of this model sit in real multiplets, matching the result that b2(Z) = 0 on the M-Theory side. 6 Discussion and outlook We proposed a new construction of eight-manifolds with Spin(7)-holonomy, based on a generalized connected sum (GCS), where two building blocks | a Calabi-Yau four-fold and a G2-holonomy manifold times S1 | are glued together along an asymptotic region that is a Calabi-Yau three-fold times a cylinder. This construction is in part inspired by the recent twisted connected sum (TCS) realization of G2-holonomy manifolds, which has resulted in a multitude of new examples of compact G2-manifolds, and thereby a resurgence of interest in the string/M-theory context. Likewise the GCS-construction that we propose, provides an avenue to construct large classes of compact Spin(7)-manifolds systematically. In particular the construction of acyl Calabi-Yau three-folds using semi-Fano three-folds [5] or tops [9] which is useful for TCS-constructions, has an obvious generalization to acyl Calabi-Yau four-folds, which is useful for expanding the set of examples of GCS-constructions. We gave an alternative description of the GCS Spin(7)-manifolds in terms of a quotient by an anti-holomorphic involution of a Calabi-Yau three-fold bered Calabi-Yau four-fold in section 4.5, which is similar in spirit to the constructions of Joyce, however the key di erence is that instead of gluing in ALE-spaces at point-like orbifold singularities, we glue in G2-manifolds with suitable asymptotics. There is a multitude of future directions to consider: 1. M/F-duality for Spin(7): M-theory on Spin(7) results 3d N = 1 theories, and we have seen that there are subsectors of the e ective eld theory, which in the limit of in nite asymptotic region enjoy enhanced supersymmetry, as discussed in section 4.4. It would clearly be very interesting to apply M/F-duality to the GCS Spin(7)-manifolds and determine how the supersymmetry breaking in the 4d F-theory vacuum is realized [1]. Needless to say it is not di cult to construct GCS-examples that have an elliptic bration and we will return to this shortly elsewhere. Again key for this will be also to understand the uxes in the GCS Spin(7)-manifolds. 2. M/Heterotic-duality and heterotic G2-systems: as was exempli ed in section 5, some GCS-constructions have a K3- bration, so that M-theory compacti cation on these has a dual description in terms of Heterotic on G2-manifolds with a T 3- bration with vector bundle. We have seen that the GCS-decomposition of the Spin(7)-manifold maps in the dual G2-compacti cation to a TCS-decomposition. Heterotic on G2manifolds has been studied only very sparsely, and this approach may very well provide further insight into the construction and moduli of vector bundles for socalled heterotic G2-systems. Furthermore, this duality also gives evidence for the existence of associative T 3- brations on G2-manifolds as conjecture from G2 mirror symmetry in [13], and these may indeed be less rare as was proposed in [23] despite the obstructions of associatives. 3. Non-abelian gauge groups: furthermore we expect, again through the duality to heterotic, that the GCS-Spin(7) manifolds can give rise to non-abelian gauge symmetries. Locally, these will have a description in the form of an ADE singularities over Cayley four-cycle (for a discussion of such non-compact examples see [29]). It would indeed be interesting to develop this further and understand e.g. a Higgs bundle description of the e ective theories and its relation to the compact Spin(7) manifolds | much like what has been done in recent years for M/F-theory on CY4. 4. Mirror symmetry: mirror symmetry for TCS-G2 manifolds as studied in [10, 13] by applying the mirror map to the Calabi-Yau three-fold building blocks. It would be interesting to see whether there is a similar way to study mirror symmetry for Spin(7) manifolds (as proposed in [50]) by applying the mirror map to the building blocks. Acknowledgments We thank for Sebastjan Cizel, Magdalena Larfors, Xenia de la Ossa, James Sparks and Yuuji Tanaka for discussions related to the present work. APB and SSN are supported by the ERC Consolidator Grant 682608 \Higgs bundles: Supersymmetric Gauge Theories and Geometry (HIGGSBNDL)". A TCS-construction of G2-manifolds The TCS G2-manifolds [4{6] have the topology of a K3- bration over an S3 and are obtained by gluing two building blocks Z that are algebraic three-folds, which are K3- bered over S2. We will denote the generic K3- bers of each building block by S . Excising a ber, the two building blocks asymptote to K3 S1 S1 I, where I is an interval. The building blocks are not Calabi-Yau but have c1(Z ) = [ S ], which means that removing a ber gives rise to acyl Calabi-Yau three-folds X . These are shown in gure 1. Here we summarize the salient points and introduce some notation that will be used in the main text. We refer the reader for a more detailed exposition to [4{6, 12]. The TCS-construction glues these two building blocks times an S1 together, with a hyper-Kahler rotation (HKR), or Donaldson matching, acting on the Kahler forms ! and holomorphic (2,0)-forms as N T = im( = N ? H2(S ; Z) K(Z ) = ker( )=[S ] : 8 Re (2;0) Im (2;0) induces an isometry on the cohomologies of the bers: H2(S+; Z) = H2(S ; Z). and an exchange of the circles in the base Se1; 1 = Sb; as shown in gure 1. In particular What will be most important for this paper is how to determine the cohomologies of the G2-manifold in terms of data of the TCS gluing. For each building block we de ne the restriction map on the second cohomology as H2(Z ; Z) H2(S ; Z) = 3;19 = U 3 ( E8)2 ; where we used the cohomology of the K3-surfaces, where U is the hyperbolic lattice of rank 2. The cohomology of the G2-manifold, can then be written in terms of the following lattices (A.1) (A.2) (A.3) Applying the Mayer-Vietoris sequence to this problem gives the cohomology of the G2 H1(J; Z) = 0 H2(J; Z) = (N+ \ N ) H3(J; Z) = Z[S] K(Z+) K(Z ) H3(Z+) 3;19=(N+ + N ) H3(Z ) \ T+) (N+ \ T ) K(Z+) K(Z ) H4(J; Z) = H4(S) (T+ \ T ) 3;19=(N + T+) 3;19=(N+ + T ) H3(Z+) H3(Z ) K(Z+) K(Z ) H5(J; Z) = 3;19=(T+ + T ) K(Z+) K(Z ) : We will determine a similar relation for the Spin(7) GCS-construction. Open Access. Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited. [hep-th/9409111] [INSPIRE]. 2153 [hep-th/9506101] [INSPIRE]. [1] C. Vafa, Evidence for F-theory, Nucl. Phys. B 469 (1996) 403 [hep-th/9602022] [INSPIRE]. [2] E. Witten, Is supersymmetry really broken?, Int. J. Mod. Phys. A 10 (1995) 1247 [3] E. Witten, Strong coupling and the cosmological constant, Mod. Phys. Lett. A 10 (1995) [4] A. Kovalev, Twisted connected sums and special Riemannian holonomy, J. Reine Angew. Math. 565 (2003) 125 [math/0012189] [INSPIRE]. [5] A. Corti, M. Haskins, J. Nordstrom and T. Pacini, Asymptotically cylindrical Calabi-Yau 3-folds from weak Fano 3-folds, Geom. Topol. 17 (2013) 1955 [arXiv:1206.2277]. [6] A. Corti, M. Haskins, J. Nordstrom and T. Pacini, G2-manifolds and associative submanifolds via semi-Fano 3-folds, Duke Math. J. 164 (2015) 1971 [arXiv:1207.4470] [7] J. Halverson and D.R. Morrison, The landscape of M-theory compacti cations on seven-manifolds with G2 holonomy, JHEP 04 (2015) 047 [arXiv:1412.4123] [INSPIRE]. [8] J. Halverson and D.R. Morrison, On gauge enhancement and singular limits in G2 compacti cations of M-theory, JHEP 04 (2016) 100 [arXiv:1507.05965] [INSPIRE]. [9] A.P. Braun, Tops as building blocks for G2 manifolds, JHEP 10 (2017) 083 [arXiv:1602.03521] [INSPIRE]. [10] A.P. Braun and M. Del Zotto, Mirror Symmetry for G2-Manifolds: Twisted Connected Sums and Dual Tops, JHEP 05 (2017) 080 [arXiv:1701.05202] [INSPIRE]. [11] T.C. da C. Guio, H. Jockers, A. Klemm and H.-Y. Yeh, E ective Action from M-theory on Twisted Connected Sum G2-Manifolds, Commun. Math. Phys. 359 (2018) 535 [arXiv:1702.05435] [INSPIRE]. [12] A.P. Braun and S. Schafer-Nameki, Compact, Singular G2-Holonomy Manifolds and M/Heterotic/F-Theory Duality, JHEP 04 (2018) 126 [arXiv:1708.07215] [INSPIRE]. [13] A.P. Braun and M. Del Zotto, Towards Generalized Mirror Symmetry for Twisted Connected Sum G2 Manifolds, JHEP 03 (2018) 082 [arXiv:1712.06571] [INSPIRE]. [14] A.P. Braun, M. Del Zotto, J. Halverson, M. Larfors, D.R. Morrison and S. Schafer-Nameki, In nitely Many M2-instanton Corrections to M-theory on G2-manifolds, arXiv:1803.02343 [15] B.S. Acharya and S. Gukov, M theory and singularities of exceptional holonomy manifolds, Phys. Rept. 392 (2004) 121 [hep-th/0409191] [INSPIRE]. [16] J.A. Harvey and G.W. Moore, Superpotentials and membrane instantons, hep-th/9907026 [17] D.D. Joyce, Compact 8-manifolds with holonomy Spin(7), Invent. Math. 123 (1996) 507. [18] D. Joyce, A new construction of compact 8-manifolds with holonomy Spin(7), J. Di . Geom. 53 (1999) 89 [math/9910002] [INSPIRE]. arXiv:1309.5027. [19] A. Kovalev, Asymptotically cylindrical manifolds with holonomy Spin(7). I, [20] F. Bonetti, T.W. Grimm and T.G. Pugh, Non-Supersymmetric F-theory Compacti cations on Spin(7) Manifolds, JHEP 01 (2014) 112 [arXiv:1307.5858] [INSPIRE]. [21] F. Bonetti, T.W. Grimm, E. Palti and T.G. Pugh, F-Theory on Spin(7) Manifolds: Weak-Coupling Limit, JHEP 02 (2014) 076 [arXiv:1309.2287] [INSPIRE]. [22] B.S. Acharya, N = 1 M-theory heterotic duality in three-dimensions and Joyce manifolds, [23] D. Joyce, Compact Manifolds with Special Holonomy, Oxford Mathematical Monographs, hep-th/9604133 [INSPIRE]. Oxford University Press (2000). [hep-th/9505053] [INSPIRE]. [24] R.C. Mclean, Deformations of calibrated submanifolds, Commun. Analy. Geom 6 (1996) 705. [25] C. Vafa and E. Witten, A one loop test of string duality, Nucl. Phys. B 447 (1995) 261 [26] S. Sethi, C. Vafa and E. Witten, Constraints on low dimensional string compacti cations, Nucl. Phys. B 480 (1996) 213 [hep-th/9606122] [INSPIRE]. [27] G. Papadopoulos and P.K. Townsend, Compacti cation of D = 11 supergravity on spaces of exceptional holonomy, Phys. Lett. B 357 (1995) 300 [hep-th/9506150] [INSPIRE]. [28] K. Becker, A note on compacti cations on Spin(7)-holonomy manifolds, JHEP 05 (2001) 003 [hep-th/0011114] [INSPIRE]. [hep-th/0109025] [INSPIRE]. [29] S. Gukov and J. Sparks, M theory on Spin(7) manifolds. 1, Nucl. Phys. B 625 (2002) 3 [30] B.S. Acharya, X. de la Ossa and S. Gukov, G ux, supersymmetry and Spin(7) manifolds, JHEP 09 (2002) 047 [hep-th/0201227] [INSPIRE]. [31] M. Becker, D. Constantin, S.J. Gates Jr., W.D. Linch III, W. Merrell and J. Phillips, M theory on Spin(7) manifolds, uxes and 3-D, N = 1 supergravity, Nucl. Phys. B 683 (2004) 67 [hep-th/0312040] [INSPIRE]. spin(7) structures, JHEP 04 (2006) 027 [hep-th/0511047] [INSPIRE]. fourfolds, JHEP 09 (2015) 107 [arXiv:1506.07356] [INSPIRE]. 053 [arXiv:1702.08537] [INSPIRE]. HJEP06(218)3 perturbative heterotic compacti cations, arXiv:1707.04613 [INSPIRE]. [hep-th/9503124] [INSPIRE]. J. Di . Geom. 101 (2015) 213 [arXiv:1212.6929]. applications to compact irreducible g2-manifolds, Annals Global Anal. Geom. 38 (2010) 221 (1996) 243 [hep-th/9606040] [INSPIRE]. 187 (1997) 679 [hep-th/9701162] [INSPIRE]. [32] D. Tsimpis , M-theory on eight-manifolds revisited: N = 1 supersymmetry and generalized [33] D. Prins and D. Tsimpis , Type IIA supergravity and M -theory on manifolds with SU(4) structure , Phys. Rev. D 89 ( 2014 ) 064030 [arXiv: 1312 .1692] [INSPIRE]. [34] D. Prins and D. Tsimpis , 3d N = 1 e ective supergravity and F-theory from M-theory on [35] I.V. Melnikov , R. Minasian and S. Sethi , Non-duality in three dimensions , JHEP 10 ( 2017 ) [36] I.V. Melnikov , R. Minasian and S. Sethi , Spacetime supersymmetry in low-dimensional [37] E. Witten , String theory dynamics in various dimensions , Nucl. Phys. B 443 ( 1995 ) 85 Cambridge Phil. Soc. 145 ( 2008 ) 311 .

This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2FJHEP06%282018%29103.pdf

Andreas P. Braun, Sakura Schäfer-Nameki. Spin(7)-manifolds as generalized connected sums and 3d \( \mathcal{N}=1 \) theories, Journal of High Energy Physics, 2018, 103, DOI: 10.1007/JHEP06(2018)103