Bi-atomic classes of positive semirings (Q2037076)

A ``factorizable monoid'' (F-monoid) \(M=\langle M,\ast ,e\rangle \) is a commutative and cancellative semigroup with identity where \(\ast \) is the monoid operation and \(e\) is the identity of \(M\) under \( \ast \) and has the natural partial order induced by divisibility: \(x|_{M}y\) \( \Leftrightarrow \exists z\in M(y=x\ast z)\). That is, our monoid acts exactly like the multiplicative monoid of nonzero elements of an integral domain. One advantage in making this blanket assumption is that whatever went on in integral domains by way of factorizations, say in the AAZ paper [J. Pure Appl. Algebra 69, 1--19 (1990)] can be mimicked for F-monoids. The usual operations on F-monoids are multiplication (\(\cdot \)), with \(e=1,\) and addition (\(+\)), with \(e=0.\) The study of Puiseux monoids by \textit{S. T. Chapman} et al. [Am. Math. Mon. 128, No. 4, 302--321 (2021; Zbl 1471.13049)] shows how this works and how ``atomic domains'' ACCP domains and BF domains and FF domains can be cloned into ``atomic monoids'' ``ACCPMs etc. Of course monoids show up in other algebraic systems such as rings and semirings \([\langle S\backslash \{0\},\cdot ,1\rangle \), \(\langle S,+,0\rangle ]\) where both \(\langle S\backslash \{0\},\cdot ,1\rangle \) and \( \langle S,+,0\rangle \) are monoids, \(1\neq 0,\) and \(\cdot \) distributes over \(+\) and of course we can insist on focusing only on F-monoids. Since multiplication by negative numbers has the potential of messing up the divisibility induced partial order in \(\langle S,+,0\rangle \), it is best to deal with semirings whose elements can be expressed in terms of nonnegative numbers. Constructions such as \(R^{+}[X]\) (resp., \(N_{0}[X]\)) of polynomials with nonnegative coefficients, where \(R^{+}\) denotes the nonnegative real numbers and \(N_{0}\) the nonnegative integers, are examples of (positive) semirings. Other notable constructions, called valuations, come from, \( N_{0}[X]\) by replacing \(X\) by \(\tau \), a suitable quadratic integer, as indicated by [\textit{P. Cesarz} et al., in: Commutative algebra and its applications. Proceedings of the fifth international Fez conference on commutative algebra and applications, Fez, Morocco, June 23--28, 2009. Berlin: Walter de Gruyter. 89--101 (2009; Zbl 1182.16038)] and from Puiseux monoids generated by geometric sequences, as studied by \textit{S. T. Chapman} et al. [Commun. Algebra 48, No. 1, 380--396 (2020; Zbl 1442.20036)]. Now what do we call a semiring consisting of F-monoids that are both atomic (resp., ACCM, BFM, FFM or HFM etc.)? The authors of the paper under review call them bi-atomic semirings, bi-BFSs etc.. The paper under review is a sort of survey of the study of positive semirings. A number of examples are given in the paper to fill the reader in on the state of the art. Of course the bi-paradise does have some snakes. For instance as indicated by \textit{F. Campanini} and \textit{A. Facchini} [Semigroup Forum 99, No. 2, 317--332 (2019; Zbl 1454.20108)], \(N_{0}[X]\) is bi-atomic but not bi-HFS. All in all, the paper is an interesting read and a good source of references.