Searcharxiv⌕ Search

arXiv subjects

Michael K. Kinyon

Publications and source records attributed to Michael K. Kinyon.

At least 19 recordsLinked to original sources
arXiv2009

Right product quasigroups and loops

Right groups are direct products of right zero semigroups and groups and they play a significant role in the semilattice decomposition theory of semigroups. Right groups can be characterized as associative right quasigroups (magmas in which left translations are bijective). If we do not assume associativity we get right quasigroups which are not necessarily representable as direct products of right zero semigroups and quasigroups. To obtain such a representation, we need stronger assumptions which lead us to the notion of \emph{right product quasigroup}. If the quasigroup component is a (one-sided) loop, then we have a \emph{right product (left, right) loop}. We find a system of identities which axiomatizes right product quasigroups, and use this to find axiom systems for right product (left, right) loops; in fact, we can obtain each of the latter by adjoining just one appropriate axiom to the right product quasigroup axiom system. We derive other properties of right product quasigroups and loops, and conclude by showing that the axioms for right product quasigroups are independent.

math.GR↗
arXiv2008

Admissible orders of Jordan loops

A commutative loop is Jordan if it satisfies the identity $x^2 (y x) = (x^2 y) x$. Using an amalgam construction and its generalizations, we prove that a nonassociative Jordan loop of order $n$ exists if and only if $n\geq 6$ and $n\neq 9$. We also consider whether powers of elements in Jordan loops are well-defined, and we construct an infinite family of finite simple nonassociative Jordan loops.

math.GR↗
arXiv2007

Buchsteiner loops

Buchsteiner loops are those which satisfy the identity $x\backslash (xy \cdot z) = (y \cdot zx)/ x$. We show that a Buchsteiner loop modulo its nucleus is an abelian group of exponent four, and construct an example where the factor achieves this exponent.

math.GR↗
arXiv2007

The Structure of F-Quasigroups

We solve a problem of Belousov which has been open since 1967: to characterize the loop isotopes of F-quasigroups. We show that every F-quasigroup has a Moufang loop isotope which is a central product of its nucleus and Moufang center. We then use the loop to reveal the structure of the associated F-quasigroup.

math.GR↗
arXiv2007

Primary decompositions in varieties of commutative diassociative loops

The decomposition theorem for torsion abelian groups holds analogously for torsion commutative diassociative loops. With this theorem in mind, we investigate commutative diassociative loops satisfying the additional condition (trivially satisfied in the abelian group case) that all $n$th powers are central, for a fixed $n$. For $n=2$, we get precisely commutative $C$ loops. For $n=3$, a prominent variety is that of commutative Moufang loops. Many analogies between commutative C and Moufang loops have been noted in the literature, often obtained by interchanging the role of the primes 2 and 3. We show that the correct encompassing variety for these two classes of loops is the variety of commutative RIF loops. In particular, when $Q$ is a commutative RIF loop: all squares in $Q$ are Moufang elements, all cubes are $C$ elements, Moufang elements of $Q$ form a normal subloop $M_0(Q)$ such that $Q/M_0(Q)$ is a C loop of exponent 2 (a Steiner loop), C elements of $L$ form a normal subloop $C_0(Q)$ such that $Q/C_0(Q)$ is a Moufang loop of exponent 3. Since squares (resp. cubes) are central in commutative C (resp. Moufang) loops, it follows that $Q$ modulo its center is of exponent 6. Returning to the decomposition theorem, we find that every torsion, commutative RIF loop is a direct product of a C 2-loop, a Moufang 3-loop, and an abelian group with each element of order prime to 6. We also discuss Moufang elements, and a class of quasigroups associated with commutative RIF loops.

math.GR↗
arXiv2006

Leibniz algebras, Lie racks, and digroups

The "coquecigrue" problem for Leibniz algebras is that of finding an appropriate generalization of Lie's third theorem, that is, of finding a generalization of the notion of group such that Leibniz algebras are the corresponding tangent algebra structures. The difficulty is determining exactly what properties this generalization should have. Here we show that \emph{Lie racks}, smooth left distributive structures, have Leibniz algebra structures on their tangent spaces at certain distinguished points. One way of producing racks is by conjugation in \emph{digroups}, a generalization of group which is essentially due to Loday. Using semigroup theory, we show that every digroup is a product of a group and a trivial digroup. We partially solve the coquecigrue problem by showing that to each Leibniz algebra that splits over its ideal generated by squares, there exists a special type of Lie digroup with tangent algebra isomorphic to the given Leibniz algebra. The general coquecigrue problem remains open, but Lie racks seem to be a promising direction.

math.RA↗
arXiv2006

When is the commutant of a Bol loop a subloop?

A left Bol loop is a loop satisfying $x(y(xz)) = (x(yx))z$. The commutant of a loop is the set of elements which commute with all elements of the loop. In a finite Bol loop of odd order or of order $2k$, $k$ odd, the commutant is a subloop. We investigate conditions under which the commutant of a Bol loop is not a subloop. In a finite Bol loop of order relatively prime to 3, the commutant generates an abelian group of order dividing the order of the loop. This generalizes a well-known result for Moufang loops. After describing all extensions of a loop $K$ such that $K$ is in the left and middle nuclei of the resulting loop, we show how to construct classes of Bol loops with non-subloop commutant. In particular, we obtain all Bol loops of order 16 with non-subloop commutant.

math.GR↗
arXiv2006

Power-associative, conjugacy closed loops

We study conjugacy closed loops (CC-loops) and power-associative CC-loops (PACC-loops). If $Q$ is a PACC-loop with nucleus $N$, then $Q/N$ is an abelian group of exponent 12; if in addition $Q$ is finite, then $|Q|$ is divisible by 16 or by 27. There are eight nonassociative PACC-loops of order 16, three of which are not extra loops. There are eight nonassociative PACC-loops of order 27, four of which have the automorphic inverse property. We also study some special elements in loops, such as Moufang elements, weak inverse property (WIP) elements, and extra elements. In a CC-loop, the set of WIP and the set of extra elements are normal subloops. For each $c$ in a PACC-loop, $c^3$ is WIP, $c^6$ is extra, and $c^{12} \in N$.

math.GR↗
arXiv2006

F-quasigroups isotopic to groups

In math.GR/0510298, we showed that every loop isotopic to an F-quasigroup is a Moufang loop. Here we characterize, via two simple identities, the class of F-quasigroups which are isotopic to groups. We call these quasigroups FG-quasigroups. We show that FG-quasigroups are linear over groups. We then use this fact to describe their structure. This gives us, for instance, a complete description of the simple FG-quasigroups. Finally, we show an equivalence of equational classes between pointed FG-quasigroups and central generalized modules over a particular ring.

math.GR↗
arXiv2004

C-loops: extensions and constructions

C-loops are loops satisfying the identity $x(y\cdot yz) = (xy\cdot y)z$. We develop the theory of extensions of C-loops, and characterize all nuclear extensions provided the nucleus is an abelian group. C-loops with central squares have very transparent extensions; they can be built from small blocks arising from the underlying Steiner triple system. Using these extensions, we decide for which abelian groups $K$ and Steiner loops $Q$ there is a nonflexible C-loop $C$ with center $K$ such that $C/K$ is isomorphic to $Q$. We discuss possible orders of associators in C-loops. Finally, we show that the loops of signed basis elements in the standard real Cayley-Dickson algebras are C-loops.

math.GR↗
arXiv2004

Rectangular loops and rectangular quasigroups

We solve two problems posed by Krapež by finding a basis of seven independent axioms for the variety of rectangular loops. Six of these axioms form a basis for the variety of rectangular quasigroups. The proofs of the lemmas showing that the six axioms are sufficient are based on proofs generated by the automated reasoning program OTTER, while most of the models verifying the independence of the axioms were generated by the finite model builder Mace4.

math.GR↗
arXiv2004

The structure of extra loops

The Sylow theorems hold for finite extra loops, as does P. Hall's theorem for finite solvable extra loops. Every finite nonassociative extra loop $Q$ has a nontrivial center, $Z(Q)$. Furthermore, $Q/Z(Q)$ is a group whenever $|Q| < 512$. Loop extensions are used to construct an infinite nonassociative extra loop with a trivial center and a nonassociative extra loop $Q$ of order 512 such that $Q/Z(Q)$ is nonassociative. There are exactly 16 nonassociative extra loops of order $16p$ for each odd prime $p$.

math.GR↗
arXiv2004

On Twisted Subgroups and Bol Loops of Odd Order

In the spirit of Glauberman's fundamental work in B-loops and Moufang loops, we prove Cauchy and strong Lagrange theorems for Bol loops of odd order. We also establish necessary conditions for the existence of a simple Bol loop of odd order, conditions which should be useful in the development of a Feit-Thompson theorem for Bol loops. Bol loops are closely related to Aschbacher's twisted subgroups, and we survey the latter in some detail, especially with regard to the so-called Aschbacher radical.

math.GR↗
arXiv2004

Infinite Simple Bol Loops

If the left multiplication group of a loop is simple, then the loop is simple. We use this observation to give examples of infinite simple Bol loops.

math.GR↗
arXiv2004

Axioms for trimedial quasigroups

We give new equations that axiomatize the variety of trimedial quasigroups. We also improve a standard characterization by showing that right semimedial, left F-quasigroups are trimedial.

math.GR↗
arXiv2004

Finite Bruck Loops

A loop $(X,\circ)$ is said to be a Bruck loop if it satisfies the (right) Bol identity $((z\circ x)\circ y)\circ x = z\circ ((x\circ y)\circ x)$ and the automorphic inverse property $(x\circ y)^{-1}=x^{-1}\circ y^{-1}$. If $X$ is a finite Bruck loop and $G$ is the group generated by all right translations $R(x): y\mapsto y\circ x$, then we show that $X$ and $G$ are central products $X = O^{2'}(X) * O(X)$ and $G = O^{2'}(G) * O(G)$, where $O^{2'}(X)$ ($O^{2'}(G)$) is the subloop (subgroup) generated by all 2-elements, and $O(X)$ ($O(G)$) is the largest normal subloop (subgroup) of odd order. In particular, if $X$ is solvable, then these central products are direct products. We also give a set of necessary conditions that must hold for a finite Bruck loop $X$ to be nonsolvable but have each proper section solvable; in particular, $X$ must be simple and consist of 2-elements, while the quotient of $G$ by its largest normal 2-subgroup must be isomorphic to $PGL_2(q)$, with $q=2^n+1\geq 5$.

math.GR↗
arXiv2002

Diassociativity in Conjugacy Closed Loops

Let $Q$ be a conjugacy closed loop, and $N(Q)$ its nucleus. Then $Z(N(Q))$ contains all associators of elements of $Q$. If in addition $Q$ is diassociative (i.e., an extra loop), then all these associators have order 2. If $Q$ is power-associative and $|Q|$ is finite and relatively prime to 6, then $Q$ is a group. If $Q$ is a finite non-associative extra loop, then $16 \mid |Q|$.

math.GR↗