# FC-Pure Injectivity and FC-Pure Flatness and Its Relationship to Two Famous Problems of Kothe and Kaplansky

Let S be a class of left R-modules. An R-module M is said to be ** S-pure injective** (resp.

**) if M has the injective (resp. projective) property respect to each S-pure exact sequence. Also, an R-module M is said to**

*S-pure projective***if M has the flat property respect to each S-pure exact sequence. As Puninski (1994), let us fix a few of the most common purities S of R-modules.**

*be S-pure flat*-
. If S = { the class of all finitely presented left modules over R}, then we have the left Cohn’s purity. In this case, S-pure exact, S-pure projective, S-pure injective and S-pure flat are commonly called*Purity*and*pure exact, pure projective, pure injective*, respectively*pure flat* -
. If S = {R/rR | r \in R} (R/rR is called cyclically presented), then we have the RD-purity. In this case, S-pure exact, S-pure projective, S-pure injective and S-pure flat are commonly called*RD-purity*(or*RD-exact*)*relative divisible exact*and*, RD-projective, RD-injective*respectively.*RD-flat,* -
(or*I-purity*) . If S = {R^n/K | K is a cyclic submodule of the R-module R^n}, then we have I-purity or*cyclically purity*. In this case, S-pure exact, S-pure projective, S-pure injective and S-pure flat are commonly called*cyclically purity*and*I-pure exact, I-pure projective, I-pure injective*, respectively.*I-pure flat* -
If S = {R/I | I is a left ideal of R}, then we have the cyclically purity (or c-purity). In this case, S-pure exact, S-pure projective, S-pure injective and S-pure flat are commonly called C-pure exact, C-pure projective, C-pure pure injective and C-pure flat, respectively.*C-purity.* -
Let S = {R/I | I is a finitely generated left ideal of R}. In this case, since each element R/I of S is finitely presented cyclic R-module, we have the FC-purity (or finitely presented cyclically purity). Note that Puninski, use the phrase Warfield purity to mean FC-purity. In this case, S-pure exact, S-pure projective, S-pure injective and S-pure flat are commonly called*FC-purity.*,*FC-pure exact**FC-pure projective**,*e and*FC-pure injectiv*, respectively.*FC-pure flat*

The question of which commutative rings have the property that every finitely generated module is a direct sum of cyclic modules has been around for many years. We will call these rings FGC rings. The problem originated in I. Kaplanskys papers. For several years, this is one of the major open problems in ring theory. A deep and difficult study was made by Brandal, Shores-R. Wiegand, S. Wiegand, Wiegand and V´amos , leading to a complete solution of the problem in the commutative case**.** Also, it was shown by Kothe and Cohen-Kaplansky: “a commutative ring R has the property that every modules is a direct sum of cyclic modules if and only if R is an Artinian principal ideal ring” (recently, a generalization of the Kothe-Cohen-Kaplansky theorem have been given by Behboodi et al. [Proc. Amer. Math. Soc.(2011)(to appear)]. for the non-commutative setting, when all idempotents of R are central). In general, the corresponding problems in the non-commutative case are still open; see [Non-associative algebra and its applications, Lecture Notes in Pure and Applied Mathematics 246 (2006), Appendix B: Unsolved Problems in the Theory of Rings and Modules. Pages 461-516] in which the following problems are considered.

**G. Kothe’s Problem**: Describe the rings over which each left (or each left, and each right) module can be decomposed as a direct sum of cyclic modules

**I. Kaplansky’s Problem **(Reported by A. A. Tuganbaev): Describe the rings over which every finitely generated module can be decomposed as a direct sum of cyclic modules.

The goal of this proposal is to study the structures and basic properties of FC-pure injectivity, FC-pure injective envelope, I-pure injectivity, FC-pure flatness and I-pure flatness of modules. In particular, we will study the relationship between these notions and the above two problems of Kothe and Kaplansky. For instance, we prove the following relationship between FC-pure injectivity and the Kothe and Kaplansky problems:

**Theorem A: ** Every left R-module is FC-pure injective (projective) if and only if R is a left Kothe ring (i.e., every left R-module is a direct sum of cyclic modules).

**Theorem B: ** Every finitely generated left R-module is FC-pure injective (projective) if and only if R is a left FGC- ring (i.e., every finitely generated left R-module is a direct sum of cyclic modules).

Related Papers:

1- **M. Behboodi (Joint with A. Ghorbani, A. Moradzadeh-Dehkordi, and S.H. Shojaee)**** **

*On FC-purity and I-purity of modules and Kothe rings*

Comm. Algebra **42** (2014), 2061-2081

2- **M. Behboodi (Joint with A. Ghorbani, A. Moradzadeh-Dehkordi, and S. H. Shojaee)**** **

*C-pure projective modules*

Comm. Algebra **41** (2013), 4559-4575