On Strongly $m$-$\Delta$-clean ring

Abstract

Motivated by the recent study of strongly $\Delta$-clean rings, we introduce and study strongly $m$-$\Delta$-clean rings. We establish their fundamental properties and characterize strongly $m$-$\Delta$-clean rings via lifting $m$-potent elements modulo $\Delta(R)$. We prove that every $m$-$\Delta$-clean ring is clean and that the factor ring of a strongly $m$-$\Delta$-clean ring modulo its Jacobson radical is reduced. Finally, we investigate corner rings, Morita context rings, and the relationships of these rings with $\Delta$-clean, local, semipotent, and strongly $m$-nil clean rings.

Keywords : clean ring, $m$-clean ring, $\Delta$-clean ring, $m$-$\Delta$-clean ring.

AMS Subject Classification : $16U99$, $16Z05$.

1 Introduction

Throughout this paper, all rings are associative with identity. Let $R$ be a ring. We denote by $U(R)$, $Id(R)$, $J(R)$, and $Nil(R)$ the group of units, the set of idempotents, the Jacobson radical, and the set of nilpotent elements of $R$, respectively. Recall that

$$\Delta(R)=\{x\in R\mid x+u\in U(R)\text{ for all }u\in U(R)\}.$$

It is well known that this is equivalent to

$$\Delta(R) =\{x\in R\mid 1-ux\in U(R)\text{ for all }u\in U(R)\}$$

$$=\{x\in R\mid 1-xu\in U(R)\text{ for all }u\in U(R)\}.$$

The subset $\Delta(R)$ was considered by Lam [6, Exercise 4.24] and later investigated in detail by Leroy and Matczuk [7]. It is known that $\Delta(R)$ is a subring of $R$ satisfying $J(R)\subseteq\Delta(R)$, and it is the largest Jacobson radical subring of $R$ which is closed under multiplication by units. In general, $\Delta(R)$ need not be an ideal of $R$.

The study of decomposition properties of ring elements has played an important role in modern ring theory. The notion of a clean ring was introduced by Nicholson [8]. A ring $R$ is called clean if every element of $R$ can be expressed as the sum of an idempotent and a unit. Since then, numerous generalizations of clean rings have been investigated, including strongly clean rings, strongly nil-clean rings, $m$-clean rings, and strongly $m$-nil clean rings [5, 9, 2]. These classes have attracted considerable attention due to their rich structural properties and their close connections with exchange rings, lifting problems, and decomposition theory.

Motivated by the role of $\Delta(R)$, several authors have investigated decompositions involving elements of $\Delta(R)$. In particular, Karabacak, Koşan, Quynh, and Tai [4] studied $\Delta U$-rings, while Danchev, Hasanzadeh, Javan, and Moussavi [3] introduced strongly $\Delta$-clean rings, in which every element can be expressed as the sum of an idempotent and an element of $\Delta(R)$ that commute with each other. They established several fundamental properties of strongly $\Delta$-clean rings.

Motivated by these developments, in this paper we introduce and investigate strongly $m$-$\Delta$-clean rings by replacing idempotents with $m$-potent elements. These notions naturally generalize the notion of strongly $\Delta$-clean rings, and provide a unified framework for studying decompositions involving $m$-potent elements and the subring $\Delta(R)$.

The paper is organized as follows.

In Section $2$, we establish the fundamental properties of $m$-$\Delta$-clean and strongly $m$-$\Delta$-clean rings. We prove that every $m$-$\Delta$-clean ring is clean. We further prove that if $R$ is strongly $m$-$\Delta$-clean, then nilpotent elements belong to $\Delta(R)$ and the factor ring $R/J(R)$ is reduced. In addition, we investigate annihilator conditions and show that strong $m$-$\Delta$-cleanness is inherited by corner rings.

In Section $3$, we obtain several characterization theorems for strongly $m$-$\Delta$-clean rings. In particular, we characterize these rings in terms of lifting $m$-potent elements modulo $\Delta(R)$. We further investigate the relationships among $m$-$\Delta$-clean rings, local rings, semipotent rings, $\Delta$-clean rings, and strongly $m$-nil clean rings. Finally, we study the behavior of strongly $m$-$\Delta$-clean rings in the setting of Morita context rings.

2 $m$-$\Delta$-clean Rings

We begin by recalling some basic properties of $\Delta(R)$ from [7, Lemma 1.1].

Proposition 2.1.

[7, Lemma 1.1] For any ring $R$, we have :

$(i)$ $\Delta(R)=\{r\in R\mid\forall_{u\in U(R)},\,ru+1\in U(R)\}=\{r\in R\mid\forall_{u\in U(R)},\,ur+1\in U(R)\}$.

$(ii)$ For any $r\in\Delta(R)$ and $u\in U(R)$, we have $ur,\,ru\in\Delta(R)$.

$(iii)$ $\Delta(R)$ is a subring of $R$.

$(vi)$ $\Delta(R)$ is an ideal of $R$ if and only if $\Delta(R)=J(R)$.

$(v)$ If $R=\displaystyle\prod_{i\in I}R_{i}$, then $\Delta(R)=\displaystyle\prod_{i\in I}\Delta(R_{i})$.

Definition 2.2.

Let $m\geq 2$. An element $r\in R$ is called $m$-$\Delta$-clean if $r=f+d$, where $f^{m}=f$ and $d\in\Delta(R)$. A ring $R$ is called $m$-$\Delta$-clean if every element of $R$ is $m$-$\Delta$-clean.

Definition 2.3.

Let $m\geq 2$. An element $r\in R$ is called strongly $m$-$\Delta$-clean if $r=f+d$, where $f^{m}=f$, $d\in\Delta(R)$, and $fd=df$. The ring $R$ itself is said to be strongly $m$-$\Delta$-clean if every element of $R$ is strongly $m$-$\Delta$-clean.

Observe that (strongly) $2$-$\Delta$-clean rings are precisely (strongly) $\Delta$-clean rings. Furthermore, every (strongly) $\Delta$-clean ring is (strongly) $m$-$\Delta$-clean for every integer $m\geq 2$. However, the converse does not hold in general.

Example 2.4.

For any prime $p$, $\mathbb{Z}_{p}$ is a strongly $m$-$\Delta$-clean ring but it is not a strongly $\Delta$-clean ring.

Proposition 2.5.

Every $m$-$\Delta$-clean ring is clean.

Proof.

Let $r\in R$. Since $R$ is a $m$-$\Delta$-clean ring, we have $r=f+d$, where $f^{m}=f$ and $d\in\Delta(R)$. Then $r=(1-f^{m-1})+\big(f+f^{m-1}-1+d\big)$. Since $(1-f^{m-1})^{2}=1-f^{m-1}$, the element $1-f^{m-1}$ is an idempotent. Furthermore, $(f+f^{m-1}-1)(f^{m-2}-f^{m-1}+1)=1$. Hence, $f+f^{m-1}-1\in U(R)$.

Since $d\in\Delta(R)$ it follows that $f+f^{m-1}-1+d\in U(R)+\Delta(R)=U(R)$. Therefore, $r=e+u$, where $e=1-f^{m-1}$ is idempotent and $u=f+f^{m-1}-1+d$ is a unit. Thus, every element of $R$ is clean, and consequently $R$ is a clean ring. ∎

Proposition 2.6.

Let $R=\displaystyle\prod_{i\in I}R_{i}$. Then $R$ is (strongly) $m$-$\Delta$-clean if and only if each $R_{i}$ is (strongly) $m$-$\Delta$-clean.

Proof.

The proof is straightforward. ∎

Theorem 2.7.

[7, Proposition 1.6.] Let $I$ be an ideal of $R$ such that $I\subseteq J(R)$. Then $\Delta(R/I)=\Delta(R)/I$.

Proposition 2.8.

Let $R$ be a (strongly) $m$-$\Delta$-clean ring such that $I\subseteq J(R)$. Then $R/I$ is (strongly) $m$-$\Delta$-clean.

Proof.

let $x\in R$. Then $x=f+d$ for some $m$-potent $f$ and $d\in\Delta(R)$. Therefore $x+I=(f+I)+(d+I)$. Since $f^{m}=f$ that implies $f+I$ is also an $m$-potent in $R/I$ and $d\in\Delta(R)$ that implies $d+I\in\Delta(R/I)=\Delta(R)/I$. Thus $x+I$ is $\Delta$-clean in $R/I$. ∎

Recall that a ring $R$ is called a $\Delta U$-ring if $1+\Delta(R)=U(R)$. For further details on $\Delta U$-rings, we refer to [4].

Proposition 2.9.

Let $R$ be a $\Delta U$-ring and $f^{m}=f$, $a\in\Delta(R)$. Then:

$(i)$ $f-f^{m-1}+a\in\Delta(R)$,

$(ii)$ $f+f^{m-1}+a\in\Delta(R)$,

$(iii)$ $2f+a\in\Delta(R)$.

Proof.

Since $R$ is a $\Delta U$-ring, we have $U(R)=1+\Delta(R)$.

$(i)$ Let $u=f-f^{m-1}+1$. Since $u$ is a unit in $R$, there exists $d\in\Delta(R)$ such that $u=1+d$. Hence, $f-f^{m-1}=d\in\Delta(R)$. As $\Delta(R)$ is a subring, it follows that $f-f^{m-1}+a\in\Delta(R)$.

$(ii)$ Since $f^{m}=f$, the element $f^{m-1}$ is an idempotent. Consequently, $(2f^{m-1}-1)^{2}=1$ and $1-2f^{m-1}\in U(R)$. Therefore, there exist $d\in\Delta(R)$ such that $1-2f^{m-1}=1+d$, which yields $-2f^{m-1}\in\Delta(R)$. Since $\Delta(R)$ is a subring of $R$, $2f^{m-1}\in\Delta(R)$. Together with part $(i)$, we obtain

$$f+f^{m-1}=(f-f^{m-1})+2f^{m-1}\in\Delta(R).$$

Hence, $f+f^{m-1}+a\in\Delta(R)$.

$(iii)$ By parts $(i)$ and $(ii)$,

$$2f=(f+f^{m-1})+(f-f^{m-1})\in\Delta(R).$$

Since $a\in\Delta(R)$, it follows that $2f+a\in\Delta(R)$. This completes the proof. ∎

Proposition 2.10.

If $f\in R$ satisfies $f^{m}=f$ and $d\in\Delta(R)$, then the following hold:

$(i)$ $(f\pm f^{m-1})d$, $d(f\pm f^{m-1})\in\Delta(R)$

$(ii)$ For every positive integer $k$, $2f^{k}d\in\Delta(R)$. In particular, if $2$ is invertible in $R$, then $\Delta(R)$ is closed under multiplication by $m$-potent elements.

Proof.

$(i)$ Since $f$ is an $m$-potent element of $R$, we have

$$(1-f^{m-1}-f)(1-f^{m-1}-f^{m-2})=1=(1-f^{m-1}+f)(1-f^{m-1}+f^{m-2})$$

Hence, $f+f^{m-1}-1$ and $f-f^{m-1}+1$ are units in $R$. Since $\Delta(R)$ is closed under multiplication by units, it follows that $(f+f^{m-1}-1)d\in\Delta(R)$ and $(f-f^{m-1}+1)d\in\Delta(R)$.

Since $\Delta(R)$ is a subring, it is closed under addition and $d\in\Delta(R)$. Therefore, $(f\pm f^{m-1})d\in\Delta(R)$. The proof that $d(f\pm f^{m-1})\in\Delta(R)$ is similar.

$(ii)$ From part $(i)$, we have $(f+f^{m-1})d\in\Delta(R)$ and $(f-f^{m-1})d\in\Delta(R)$, then it follows that

$$2fd=(f+f^{m-1})d+(f-f^{m-1})d\in\Delta(R).$$

Now let $k$ be any positive integer. Then $f^{k}$ is again an $m$-potent element. Applying the above argument to $f^{k}$, we conclude that $2f^{k}d\in\Delta(R)$.

Finally, if $2\in U(R)$. Then $f^{k}d=2^{-1}(2f^{k}d)\in\Delta(R)$. In particular, taking $k=1$, we obtain $fd\in\Delta(R)$. Therefore, $\Delta(R)$ is closed under multiplication by $m$-potent elements. ∎

As an immediate consequence of the above Proposition 2.10, we can prove the [3, Proposition 2.6] by taking $m=2$ in the above result which is as follows :

Corollary 2.11.

[3, Proposition 2.6] For every $e\in\mathrm{Id}(R)$ and $d\in\Delta(R)$, the containment $2ed\in\Delta(R)$ holds.

Proposition 2.12.

Let $R$ be strongly $m$-$\Delta$-clean. If $a^{2}\in\Delta(R)$, then $a\in\Delta(R)$.

Proof.

Let $a\in R$. Since $R$ is strongly $m$-$\Delta$-clean, there exist $f,d\in R$ such that $a=f+d$, where $f^{m}=f$, $d\in\Delta(R)$, and $fd=df$. Then

$$a^{2}=(f+d)^{2}=f^{2}+2fd+d^{2}.$$

By assumption, $a^{2}\in\Delta(R)$. Moreover, since $d\in\Delta(R)$, we have $d^{2}\in\Delta(R)$. Also, since $f$ is $m$-potent and $d\in\Delta(R)$, it follows from the Proposition 2.10 $(ii)$ that $2fd\in\Delta(R)$. Therefore, $f^{2}=a^{2}-2fd-d^{2}\in\Delta(R)$.

Next, $f^{2}=f\big(f-f^{m-1}+1\big)$ and $f-f^{m-1}+1$ is a unit in $R$. Hence, multiplying by the inverse of $f-f^{m-1}+1$ implies

$$f=f^{2}(f-f^{m-1}+1)^{-1}\in\Delta(R)$$

Consequently, $a=f+d\in\Delta(R)$. This completes the proof. ∎

Corollary 2.13.

Let $R$ be a strongly $m$-$\Delta$-clean ring. If $a^{2^{k}}\in\Delta(R)$ for some positive integer $k$, then $a\in\Delta(R)$.

Proof.

Suppose that $a^{2^{k}}\in\Delta(R)$ for some positive integer $k$. Since $a^{2^{k}}=\big(a^{2^{k-1}}\big)^{2}$, Proposition 2.12 implies that $a^{2^{k-1}}\in\Delta(R)$. Repeating this argument successively yields

$$a^{2^{k-2}},\,a^{2^{k-3}},\,\ldots,\,a^{2},\,a\in\Delta(R).$$

Hence, $a\in\Delta(R)$. ∎

Corollary 2.14.

Let $R$ be a strongly $m$-$\Delta$-clean ring. Then every nilpotent element of $R$ belongs to $\Delta(R)$, that is, $\mathrm{Nil}(R)\subseteq\Delta(R)$.

Proof.

Let $a\in\mathrm{Nil}(R)$. Then, there exists a positive integer $n$ such that $a^{n}=0$. Choose a positive integer $k$ such that $2^{k}\geq n$. Since $a^{n}=0$, we have $a^{2^{k}}=0\in\Delta(R)$. Therefore, Corollary 2.13 implies that $a\in\Delta(R)$. Hence, $\mathrm{Nil}(R)\subseteq\Delta(R)$. ∎

Proposition 2.15.

Let $R$ be a strongly $m$-$\Delta$-clean ring with $2\in U(R)$. Then for any $a\in R$, $a-a^{m}\in\Delta(R)$.

Proof.

Let $a\in R$. Then $a=f+d$, where $f^{m}=f$, $d\in\Delta(R)$, and $fd=df$. Since $2\in U(R)$, Proposition 2.10 implies that $f^{i}d^{j}\in\Delta(R)$ for all positive integers $i$ and $j$. Hence, $a^{m}=(f+d)^{m}=f+d^{\prime}$ for some $d^{\prime}\in\Delta(R)$. Therefore, $a-a^{m}=(f+d)-(f+d^{\prime})=d-d^{\prime}\in\Delta(R)$. ∎

Recall that a ring $R$ is called reduced if it contains no nonzero nilpotent elements; equivalently, $Nil(R)=(0)$.

Lemma 2.16.

Let $R$ be a strongly $m$-$\Delta$-clean ring. Then $R/J(R)$ is reduced.

Proof.

Suppose that $x^{2}\in J(R)$. Since $J(R)\subseteq\Delta(R)$, Proposition 2.12 implies that $x\in\Delta(R)$. We next show that $x\in J(R)$. Let $r\in R$ be arbitrary. Then

$$(1-rx)(1+xr)=1+xr-rx-rx^{2}r=(xr-rx)+(1-rx^{2}r).$$

Since $x^{2}\in J(R)$, that implies $1-rx^{2}r\in U(R)$. Therefore, it suffices to prove that $xr-rx\in\Delta(R)$. Since $R$ is strongly $m$-$\Delta$-clean, $r=f+d$, where $f^{m}=f\quad\text{and}\quad fd=df$. Then

$$xr-rx=x(f+d)-(f+d)x=(xf-fx)+(xd-dx).$$

Since $x,d\in\Delta(R)$, we have $xd-dx\in\Delta(R)$. Hence it remains to show that $xf-fx\in\Delta(R)$.

Now $[f^{m-1}x(1-f^{m-1})]^{2}=0\in\Delta(R)$, because $(1-f^{m-1})f^{m-1}=0$. Hence Proposition 2.12 implies that $f^{m-1}x-f^{m-1}xf^{m-1}\in\Delta(R)$. Similarly, $[(1-f^{m-1})xf^{m-1}]^{2}=0$ that implies $xf^{m-1}-f^{m-1}xf^{m-1}\in\Delta(R)$. Furthermore,

$$xf^{m-1}-f^{m-1}x=(xf^{m-1}-f^{m-1}xf^{m-1})-(f^{m-1}x-f^{m-1}xf^{m-1})\in\Delta(R).$$

Hence,

$$xf-fx=f^{m-1}x-xf^{m-1}+x(f+f^{m-1})-(f+f^{m-1})x\in\Delta(R).$$

Consequently, $xr-rx\in\Delta(R)$. Therefore,

$$(1-rx)(1+xr)\in U(R)+\Delta(R)\subseteq U(R).$$

Hence $1-rx$ is right invertible for every $r\in R$. This shows that $x\in J(R)$. Thus every nilpotent element of $R/J(R)$ is zero, and hence $R/J(R)$ is reduced. ∎

Let $R$ be a ring and let $a\in R$. Denote

$$ann_{l}(a)=\{r\in R:ra=0\}\quad\text{and}\quad ann_{r}(a)=\{r\in R:ar=0\}.$$

Lemma 2.17.

Let $R$ be a ring and let $a\in R$. Suppose that $a=f+d$ is a strongly $m$-$\Delta$-clean decomposition of $a$, where $f$ is an $m$-potent element, $d\in\Delta(R)$, and $fd=df$. Then

$$ann_{l}(a)\subseteq ann_{l}(f)\quad\text{and}\quad ann_{r}(a)\subseteq ann_{r}(f).$$

Proof.

We prove that $ann_{l}(a)\subseteq ann_{l}(f)$. The proof for the right annihilator is similar.

Let $r\in ann_{l}(a)$. Then $ra=0$, and hence $r(f+d)=0$, which implies that $rf=-rd$. Since $f^{m}=f$, multiplying the above equality on the right by $f^{m-1}$, we get $rf=-rdf^{m-1}$. Since $fd=df$, we obtain

$$rf=(-1)^{t}rd^{t}f^{\,m-t},\qquad 1\leq t\leq m-1.$$

In particular, taking $t=m-1$, we have $rf=(-1)^{m-1}rd^{\,m-1}f$.

Again, since $fd=df$, it follows that $rd^{\,m-1}f=rfd^{\,m-1}$, and therefore $rf=(-1)^{m-1}rfd^{\,m-1}$.

Hence,

$$rf\bigl(1+(-1)^{m}d^{\,m-1}\bigr)=0.$$

Since $d\in\Delta(R)$, we have $(-1)^{m}d^{\,m-1}\in\Delta(R)$, and therefore $1+(-1)^{m}d^{\,m-1}\in U(R)$. Therefore $rf=0$. Thus $r\in ann_{l}(f)$ and consequently $ann_{l}(a)\subseteq ann_{l}(f)$. ∎

Theorem 2.18.

Let $e^{2}=e\in R$. If $a\in eRe$ is strongly $m$-$\Delta$-clean in $R$, then $a$ is strongly $m$-$\Delta$-clean in $eRe$.

Proof.

Since $a$ is strongly $m$-$\Delta$-clean in $R$, there exist elements $f,d\in R$ such that $a=f+d$, where $f^{m}=f$, $d\in\Delta(R)$, and $fd=df$.

As $a\in eRe$, we have $a=ea=ae$. Hence, $(1-e)a=a(1-e)=0$, that is, $1-e\in ann_{l}(a)\cap ann_{r}(a)$. By Lemma 2.17, it follows that $1-e\in ann_{l}(f)\cap ann_{r}(f)$. Therefore, $(1-e)f=f(1-e)=0$, which yields $f=ef=fe$.

Now, $a=f+d=efe+ede$. Since $f^{m}=f$ and $f=ef=fe$, we obtain $(efe)^{m}=efe$, so $efe$ is an $m$-potent element of $eRe$. So, it is enough to show that $ede\in\Delta(eRe)$.

Now we will prove that $eRe\cap\Delta(R)\subseteq\Delta(eRe)$. Let $r\in eRe\cap\Delta(R)$. Take any unit $u\in U(eRe)$. Then $(u+(1-e))(u^{-1}+(1-e))=1$, which implies that $u+(1-e)\in U(R)$. Since $r\in\Delta(R)$, it follows that $1-(u+(1-e))r\in U(R)$. Consequently, there exists an element $v\in R$ such that $(1-(u+(1-e))r)v=1$.

As $r\in eRe$, we have $(u+(1-e))r=ur$, and therefore $(1-ur)v=1$. Now multiply the above equality by $e$ on both sides, we get $e(1-ur)ve=e$. Since $u,r\in eRe$, we obtain $(e-ur)eve=e$. Observe that $eve\in eRe$. Thus $e-ur$ is left invertible in $eRe$ for every $u\in U(eRe)$. Consequently, it follows that $r\in\Delta(eRe)$. Hence, $eRe\cap\Delta(R)\subseteq\Delta(eRe)$.

Further,

$$ede=a-efe=a-f=d\in eRe\cap\Delta(R)\subseteq\Delta(eRe).$$

Therefore, $d\in\Delta(eRe)$. Moreover, $fd=df$ and $f,d\in eRe$. Hence, $a=efe+ede$ is a strongly $m$-$\Delta$-clean decomposition of $a$ in $eRe$, proving that $a$ is strongly $m$-$\Delta$-clean in $eRe$. ∎

Corollary 2.19.

Let $e^{2}=e\in R$. If $R$ is a strongly $m$-$\Delta$-clean ring, then $eRe$ is also strongly $m$-$\Delta$-clean.

3 Characterizations of $m$-$\Delta$-clean Rings

We say that $m$-potent elements lift modulo $\Delta(R)$ if, whenever $a\in R$ satisfies $a-a^{m}\in\Delta(R)$, there exists an $m$-potent element $f\in R$ such that $a-f\in\Delta(R)$. If, in addition, $af=fa$, then we say that $m$-potent elements lift strongly modulo $\Delta(R)$.

Theorem 3.1.

Let $R$ be a ring with $2\in U(R)$. Then the followings are equivalent:

$(i)$ $R$ is strongly $m$-$\Delta$-clean.

$(ii)$ For any every $x\in R$, $x-x^{m}\in\Delta(R)$ and each $m$-potent lifts (strongly) modulo $\Delta(R)$.

Proof.

$(i)\Rightarrow(ii)$ Assume that $R$ is strongly $m$-$\Delta$-clean and let $x\in R$. By Proposition 2.15, we have $x-x^{m}\in\Delta(R)$.

Now let $a\in R$ be such that $a-a^{m}\in\Delta(R)$. Since $R$ is strongly $m$-$\Delta$-clean, there exist an $m$-potent element $f\in R$ and an element $d\in\Delta(R)$ such that $a=f+d\quad\text{and}\quad af=fa$. It follows that $a-f=d\in\Delta(R)$, where $f$ is $m$-potent and commutes with $a$. Hence, every $m$-potent element lifts strongly modulo $\Delta(R)$.

$(ii)\Leftarrow(i)$ Let $x\in R$. By hypothesis, $x-x^{m}\in\Delta(R)$. Since $m$-potent elements lift strongly modulo $\Delta(R)$, there exists an $m$-potent element $f\in R$ such that $x-f\in\Delta(R)\quad\text{and}\quad xf=fx$. Set $d=x-f\in\Delta(R)$. Then $x=f+d$, where $f$ is $m$-potent and $d\in\Delta(R)$. Moreover, from $xf=fx$ it follows that $df=fd$. Hence, $R$ is strongly $m$-$\Delta$-clean. ∎

Proposition 3.2.

The following conditions are equivalent for a ring $R$:

$(i)$ $R$ is a $\Delta U$-ring.

$(ii)$ All $m$-clean elements of $R$ are $\Delta$-clean.

Proof.

$(i)\Rightarrow(ii)$: Let $r\in R$ be an $m$-clean element. Let $r\in R$ be an $m$-clean element. Then $r=f+u$, where $f^{m}=f$ and $u\in U(R)$. Since $R$ is a $\Delta U$-ring, write $u=1+d$, where $d\in\Delta(R)$. Hence,

$$r=(1-f^{m-1})+(f+f^{m-1}+d).$$

Since $(1-f^{m-1})^{2}=1-f^{m-1}$, the element $1-f^{m-1}$ is idempotent. Moreover, Proposition 2.9 implies that $f+f^{m-1}+d\in\Delta(R)$. Thus, $r$ is $\Delta$-clean.

$(ii)\Rightarrow(i)$: Let $u\in U(R)$. Since $u$ is $m$-clean, it is $\Delta$-clean by hypothesis. Hence, $u=e+d$, where $e^{2}=e$ and $d\in\Delta(R)$. It follows that $eu^{-1}=1-du^{-1}$. Since $du^{-1}\in\Delta(R)$, we have $-du^{-1}\in\Delta(R)$, and hence $1-du^{-1}\in 1+\Delta(R)\subseteq U(R)$. Therefore, $eu^{-1}\in U(R)$, which implies that $e\in U(R)$.

Since $e$ is an idempotent and a unit, it follows that $e=1$. Consequently, $u=1+d\in 1+\Delta(R)$. Thus, $U(R)\subseteq 1+\Delta(R)$. The reverse inclusion is immediate, and hence $U(R)=1+\Delta(R)$. Therefore, $R$ is a $\Delta U$-ring. ∎

Recall that a ring $R$ is called an $m$-potent ring if every element of $R$ is $m$-potent, that is, $a^{m}=a$ for all $a\in R$.

Theorem 3.3.

A ring $R$ is an $m$-potent ring if and only if $R$ is strongly $m$-$\Delta$-clean and $\Delta(R)=0$.

Proof.

Suppose that $R$ is an $m$-potent ring, and let $x\in\Delta(R)$. Since $\Delta(R)$ is a subring, $x^{m-1}\in\Delta(R)$. Hence, $1-x^{m-1}$ is a unit in $R$. Moreover, $x^{m}=x$, and therefore $x(1-x^{m-1})=0$. Since $1-x^{m-1}$ is invertible, it follows that $x=0$. Thus, $\Delta(R)=0$. Since every element of $R$ is $m$-potent and $0\in\Delta(R)$, $R$ is strongly $m$-$\Delta$-clean.

Conversely, suppose that $R$ is strongly $m$-$\Delta$-clean and $\Delta(R)=0$. For any $a\in R$, let $a=f+d$, where $f^{m}=f$ and $d\in\Delta(R)$. Since $\Delta(R)=0$, we have $d=0$. Thus every element of $R$ is $m$-potent, and hence $R$ is an $m$-potent ring. ∎

Theorem 3.4.

Let $R$ be a ring with only trivial idempotents. Then $R$ is a semipotent ring if and only if $R$ is a local ring.

Proof.

$(\Rightarrow)$ Suppose that $R$ is semipotent. Let $x\in R$ be such that $x\notin U(R)$. We claim that $x\in J(R)$.

Assume, to the contrary, that $x\notin J(R)$. Then the principal ideal $(x)$ is not contained in $J(R)$. Since $R$ is semipotent, there exists a nonzero idempotent $e\in(x)$. By hypothesis, the only idempotents in $R$ are $0$ and $1$, and since $e\neq 0$, it follows that $e=1$. Hence $1\in(x)$, so there exists $t\in R$ such that $1=tx$. This shows that $x$ is a unit, contradicting the assumption that $x\notin U(R)$.

Therefore, $x\in J(R)$, and hence every non-unit of $R$ lies in $J(R)$. It follows that $R$ is local.

$(\Leftarrow)$ Conversely, suppose that $R$ is local. Then $R=U(R)\cup J(R)$. Let $I$ be an ideal of $R$ such that $I\nsubseteq J(R)$. Then there exists $x\in I$ such that $x\notin J(R)$. Since $R$ is local, this implies that $x\in U(R)$. Hence $1\in I$, and therefore $I=R$. In particular, $I$ contains the idempotent $1\neq 0$. Thus every ideal of $R$ not contained in $J(R)$ contains a nonzero idempotent. Therefore, $R$ is semipotent. ∎

Theorem 3.5.

Let $R$ be a ring with only trivial idempotents. If $R$ is a $m$-$\Delta$-clean ring, then it is local.

Proof.

Let $x\in R$. Since $R$ is $m$-$\Delta$-clean, then $x=f+d$, where $f^{m}=f$ and $d\in\Delta(R)$. Then $f^{m-1}$ is an idempotent. As $R$ has only trivial idempotents, either $f^{m-1}=0$ or $f^{m-1}=1$.

If $f^{m-1}=0$, then $f=f^{m}=f^{m-1}f=0$, and hence $x=d\in\Delta(R)$. Therefore, $1-x\in U(R)$.

If $f^{m-1}=1$, then $f\in U(R)$. Hence, $x=f+d\in U(R)+\Delta(R)=U(R)$

Thus, for every $x\in R$, either $x\in U(R)$ or $1-x\in U(R)$. Hence $R$ is local. ∎

Corollary 3.6.

Let $R$ have only trivial idempotents. If $R$ is an $m$-$\Delta$-clean ring, then $R$ is semipotent.

Proof.

The result follows immediately from Theorem 3.4. ∎

Remark 3.7.

The converse of Theorem 3.5 does not hold in general. Indeed, the ring $\mathbb{Z}_{3}$ has only trivial idempotents and is a local ring, but it is not a $2$-$\Delta$-clean ring.

For clarity, we use the notation $U_{m-1}$ to represent the set of $(m-1)$-th roots of unity which is defined as follows : $U_{m-1}(R)=\{u\in R:u^{m-1}=1\}$.

Theorem 3.8.

Let $U_{m-1}(R)\subseteq 1+\Delta(R)$. Then $R$ is strongly $m$-$\Delta$-clean if and only if strongly $\Delta$-clean.

Proof.

Suppose that $R$ is strongly $m$-$\Delta$-clean, and let $a=f+d$, where $f^{m}=f$, $d\in\Delta(R)$, and $fd=df$. Consider the element $u=f-f^{m-1}+1$. Then $u^{m-1}=1$, and hence $u\in U_{m-1}(R)$. By hypothesis, $U_{m-1}(R)\subseteq 1+\Delta(R)$, so $u=1+d^{\prime}$ for some $d^{\prime}\in\Delta(R)$. Therefore, $f=f^{m-1}+d^{\prime}$.

Consequently, $a=f^{m-1}+(d+d^{\prime})$. Since $d+d^{\prime}\in\Delta(R)$, $f^{m-1}$ is an idempotent, and $f^{m-1}(d+d^{\prime})=(d+d^{\prime})f^{m-1}$, it follows that $a$ is strongly $\Delta$-clean. Hence, $R$ is strongly $\Delta$-clean.

Conversely, every idempotent is $m$-potent. Therefore, every strongly $\Delta$-clean decomposition is also a strongly $m$-$\Delta$-clean decomposition. Hence, $R$ is strongly $m$-$\Delta$-clean. ∎

Lemma 3.9.

Let $R=\begin{pmatrix}A&M\\ N&B\end{pmatrix}$ be the Morita context. If $R$ is strongly $m$-$\Delta$-clean, then $A$ and $B$ are strongly $m$-$\Delta$-clean with $MN\subseteq J(A)$ and $NM\subseteq J(B)$.

Proof.

Consider the idempotent $e=\begin{pmatrix}1&0\\ 0&0\end{pmatrix}$. Then $eRe\cong A$ and $(1-e)R(1-e)\cong B$. Since $R$ is strongly $m$-$\Delta$-clean, Corollary 2.19 implies that the corner rings $eRe$ and $(1-e)R(1-e)$ are also strongly $m$-$\Delta$-clean. Hence $A$ and $B$ are strongly $m$-$\Delta$-clean rings.

Next, by Lemma 2.16 it follows that every nilpotent element of a strongly $m$-$\Delta$-clean ring belongs to $J(R)$. Therefore for every $m\in M$ and $n\in N$, we obtain

$$\begin{pmatrix}0&m\\ 0&0\end{pmatrix},\,\begin{pmatrix}0&0\\ n&0\end{pmatrix}\in J(R).$$

Therefore, it follows that

$$\begin{pmatrix}0&m\\ n&0\end{pmatrix}=\begin{pmatrix}0&m\\ 0&0\end{pmatrix}+\begin{pmatrix}0&0\\ n&0\end{pmatrix}\in J(R)$$

for every $m\in M$ and $n\in N$.

Now let $m_{1},m_{2}\in M$ and $n_{1},n_{2}\in N$. Then

$$\begin{pmatrix}m_{1}n_{1}&0\\ 0&n_{2}m_{2}\end{pmatrix}=\begin{pmatrix}0&m_{1}\\ n_{2}&0\end{pmatrix}\begin{pmatrix}0&m_{2}\\ n_{1}&0\end{pmatrix}\in J(R).$$

Consequently,

$$\begin{pmatrix}MN&M\\ N&NM\end{pmatrix}\subseteq J(R).$$

Finally, by [10, Theorem 1], it follows that $MN\subseteq J(A)$ and $NM\subseteq J(B)$.

This completes the proof. ∎

Recall that a ring $R$ is called strongly $m$-nil clean (see [2]) if every element $a\in R$ can be expressed as $a=f+n$, where $f^{m}=f$, $n\in Nil(R)$, and $fn=nf$.

Theorem 3.10.

Let $R$ be a ring and $m\geq 2$ be an integer with $m\not\equiv 1~(mod~3)$, $m\not\equiv 1~(mod~8)$. Then $R$ is strongly $m$-nil clean ring if and only if $R$ is strongly $m$-$\Delta$-clean and $\Delta(R)\subseteq Nil(R)$.

Proof.

We first show that every element of $\Delta(R)$ is nilpotent. Let $a\in\Delta(R)$. Since $R$ is strongly $m$-nil clean, there exist an $m$-potent element $f$ and a nilpotent element $n$ in $R$ such that $a=f+n$, with $fn=nf$. Consider the element $1-f^{m-1}=\bigl(f+(1-f^{m-1})\bigr)-f=v-f$, where $v=f+(1-f^{m-1})$ is a unit in $R$. Since $\bigl(f+f^{m-1}-1\bigr)\bigl(f^{m-2}-f^{m-1}+1\bigr)=1$. Hence,

$$1-f^{m-1}=v-f=v-(a-n)=(v-a)+n.$$

Since $a\in\Delta(R)$ and $v\in U(R)$, we have $u=v-a\in U(R)+\Delta(R)=U(R)$. Therefore, $1-f^{m-1}=u+n=u\bigl(1+u^{-1}n\bigr)\in U(R)$. On the other hand, $1-f^{m-1}$ is an idempotent. Hence, $1-f^{m-1}=1$, which implies $f^{m-1}=0$. Therefore, $f=f^{m-1}f=0$. Thus, $a=n\in Nil(R)$. Thus, each element of $\Delta(R)$ is nilpotent.

On the other hand, by [2, Theorem 1.10] $R/J(R)$ is $m$-potent ring. Then $Nil(R)\subseteq J(R)$. Since if $r\in Nil(R)$ that implies $r^{k}=0$ for some integer $k$. then $(r+J(R))^{k}=J(R)$ and again $R/J(R)$ is $m$-potent ring, that implies $r+J(R)=J(R)$ which shows that $r\in J(R)$.

Then it follows that $Nil(R)\subseteq J(R)\subseteq\Delta(R)$. Hence $Nil(R)=\Delta(R)$. Consequently $R$ is strongly $m$-$\Delta$-clean.

By [2, Theorem 1.10], the factor ring $R/J(R)$ is an $m$-potent ring. Let $r\in Nil(R)$. Then $r^{k}=0$ for some positive integer $k$, and hence $(r+J(R))^{k}=J(R)$. Since every nilpotent element in an $m$-potent ring is zero, it follows that $r+J(R)=J(R)$, and therefore $r\in J(R)$. Thus, $Nil(R)\subseteq J(R)\subseteq\Delta(R)$. Therefore we obtain $\Delta(R)=Nil(R)$. Hence every strongly $m$-nil clean decomposition is a strongly $m$-$\Delta$-clean decomposition. Therefore, $R$ is strongly $m$-$\Delta$-clean and $\Delta(R)$ is nil.

Conversely, let $a\in R$. Since $R$ is strongly $m$-$\Delta$-clean, there exist an $m$-potent element $f\in R$ and an element $d\in\Delta(R)$ such that $a=f+d$ and $fd=df$. As $\Delta(R)\subseteq Nil(R)$, $d$ is nilpotent. Hence $R$ is strongly $m$-nil clean. ∎

Theorem 3.11.

Let $R=\begin{pmatrix}A&M\\ N&B\end{pmatrix}$ be a Morita context, and let $m\geq 2$ be an integer satisfying $m\not\equiv 1~(mod~3)$ and $m\not\equiv 1~(mod~8)$. Suppose that $J(A)$ and $J(B)$ are nilpotent and that $\Delta(A/J(A))=\Delta(B/J(B))=0$. Then the following are equivalent:

$(i)$ $R$ is strongly $m$-$\Delta$-clean.

$(ii)$ $A$ and $B$ are strongly $m$-$\Delta$-clean with $MN\subseteq J(A)$ and $NM\subseteq J(B)$.

Proof.

$(i)\Rightarrow(ii)$: The implication follows immediately from Lemma 3.9.

$(ii)\Rightarrow(i)$: Assume that $A$ and $B$ are strongly $m$-$\Delta$-clean rings satisfying $MN\subseteq J(A)$ and $NM\subseteq J(B)$. By Proposition 2.8, the factor rings $A/J(A)$ and $B/J(B)$ are strongly $m$-$\Delta$-clean. Since $\Delta(A/J(A))=\Delta(B/J(B))=0$, Theorem 3.3 implies that $A/J(A)$ and $B/J(B)$ are $m$-potent rings. As $J(A)$ and $J(B)$ are nilpotent ideals, they are nil ideals. Hence, [2, Theorem 1.10] implies that both $A$ and $B$ are strongly $m$-nil clean.

Furthermore, since $MN\subseteq J(A)$ and $NM\subseteq J(B)$, and both $J(A)$ and $J(B)$ are nilpotent ideals, it follows that $MN$ and $NM$ are nilpotent. Applying [2, Theorem 1.10], we conclude that the Morita context ring $R=\begin{pmatrix}A&M\\ N&B\end{pmatrix}$ is strongly $m$-nil clean.

Finally, by Theorem 3.10, every strongly $m$-nil clean ring is strongly $m$-$\Delta$-clean. Therefore, $R$ is strongly $m$-$\Delta$-clean. This completes the proof. ∎