× Linear Transformation Multiplicity Dual Space Minimal Polynomial \(T\)-Complement Cyclic Operator Indecomposable Operator Characteristic Polynomial Diagonalizable Smith Normal Form Normal Subgroup Isomorphism Center Centralizer Normalizer Stabilizer Orbit \(p\)-group Sylow \(p\)-Subgroup Sylow's Theorem Simple Group Solvable Group Field Integral Domain Ascending Chain Condition Principal Ideal Domain Unique Factorization Domain Polynomial Ring Division Algorithm in SageMath Prime and Irreducible Adjoining Element Splitting Field Minimal Polynomial Field Extension Separable Algebraic Closure Galois Group Galois Theory KU 2015 (January) KU 2015 (August) KU 2016 (January) KU 2016 (August) KU 2017 (January) KU 2017 (August) KU 2018 (January) KU 2018 (August) KU 2019 (January) KU 2019 (August) KU 2020 (January) KU 2020 (August) KU 2021 (January) KU 2021 (August) KU 2022 (January) KU 2022 (August) KU 2023 (January) KU 2023 (August) KU 2024 (January)
☰ Menu

KU 2020 (January) ALGEBRA QUALIFYING EXAM

My name is Hanzhang Yin, and I have developed this website as a resource to facilitate the review of key concepts in abstract algebra, and it is not fully completed. Feel free to email me at hanyin@ku.edu if there are any errors or suggestions for improvement.

2. Let \( T : \mathbb{R}^3 \rightarrow \mathbb{R}^3 \) be a linear operator.

(a) Suppose the matrix of \( T \) with respect to the standard basis of \( \mathbb{R}^3 \) equals \[ \begin{pmatrix} \alpha & 0 & 0 \\ 0 & \beta & -\gamma \\ 0 & \gamma & \beta \end{pmatrix}, \] with \( \alpha, \beta, \gamma \in \mathbb{R} \). Prove that \( T \) is a normal operator.

(b) Suppose the matrix of \( T \) with respect to the standard basis is \[ A = \begin{pmatrix} 1 & 1 & 0 \\ 0 & 1 & 1 \\ 1 & 0 & 1 \end{pmatrix}. \] Prove that \( T \) is a normal operator and find an orthonormal basis \( B \) for \( \mathbb{R}^3 \) such that the matrix of \( T \) with respect to \( B \) has the form in part (a).

Proof (a). Given that the matrix is the linear operator \(T\) respect to the standard basis. We have \[ \begin{align} T(e_1) = \alpha e_1, \\ T(e_2) = \beta e_2 + \gamma e_3, \\ T(e_3) = -\gamma e_2 + \beta e_3. \end{align} \] Then, we have the transpose \(\begin{pmatrix} \alpha & 0 & 0 \\ 0 & \beta & \gamma \\ 0 & -\gamma & \beta \end{pmatrix}\), which is the adjoint of the matrix of \(T\). Thus, we have \[ \begin{align} T^*(e_1) &= \alpha e_1, \\ T^*(e_2) &= \beta e_2 - \gamma e_3, \\ T^*(e_3) &= \gamma e_2 + \beta e_3. \end{align} \] Now, let \(v = a e_1 + b e_2 + c e_3\), then we have \[ \begin{align} T^*(T(v)) &= T^*(a\alpha e_1 + b\beta e_2 + b\gamma e_3 - c\gamma e_2 + c\beta e_3)\\ &= T^*(a\alpha e_1 + (b\beta - c\gamma) e_2 + (b\gamma + c\beta) e_3)\\ &= a\alpha T^*(e_1) + (b\beta - c\gamma) T^*(e_2) + (b\gamma + c\beta) T^*(e_3)\\ &= a\alpha \alpha e_1 + (b\beta - c\gamma)(\beta e_2 - \gamma e_3) + (b\gamma + c\beta)(\gamma e_2 + \beta e_3)\\ &= a\alpha^2 e_1 + b(\beta^2 + \gamma^2)e_2 + c(\beta^2 + \gamma^2)e_3\\ \end{align} \] For the other direction, \[ \begin{align} T(T^*(v)) &= T(a\alpha e_1 + b\beta e_2 - b\gamma e_3 + c\gamma e_2 + c\beta e_3)\\ &= T(a\alpha e_1 + (b\beta + c\gamma) e_2 + (-b\gamma + c\beta) e_3)\\ &= a\alpha T(e_1) + (b\beta + c\gamma) T(e_2) + (-b\gamma + c\beta) T(e_3)\\ &= a\alpha \alpha e_1 + (b\beta + c\gamma)(\beta e_2 + \gamma e_3) + (-b\gamma + c\beta)(-\gamma e_2 + \beta e_3)\\ &= a\alpha^2 e_1 + b(\beta^2 + \gamma^2)e_2 + c(\beta^2 + \gamma^2)e_3\\ \end{align} \] Hence, we can see that \(T^*(T(v)) = T(T^*(v))\) for any \(v\in \mathbb{R}^3\). Therefore, we have \(T\) is a normal operator. \(\blacksquare\)

Proof (b). Similar to part \((a)\), we have \[ \begin{align} T(e_1) &= e_1 + e_3, \\ T(e_2) &= e_1 + e_2, \\ T(e_3) &= e_2 + e_3. \end{align} \] Then we have the adjoint of the matrix of \(T\) is \(\begin{pmatrix} 1 & 0 & 1 \\ 1 & 1 & 0 \\ 0 & 1 & 1 \end{pmatrix}\). Then we have \[ \begin{align} T^*(e_1) &= e_1 + e_2, \\ T^*(e_2) &= e_2 + e_3, \\ T^*(e_3) &= e_1 + e_3. \end{align} \] Now, set \(v = a e_1 + b e_2 + c e_3\), then we have \[ \begin{align} T^*(T(v)) &= T^*(a e_1 + a e_3 + b e_1 + b e_2 + c e_2 + c e_3)\\ &= T^*((a + b)e_1 + (b + c)e_2 + (a + c)e_3)\\ &= (a + b)T^*(e_1) + (b + c)T^*(e_2) + (a + c)T^*(e_3)\\ &= (a + b)(e_1 + e_2) + (b + c)(e_2 + e_3) + (a + c)(e_1 + e_3)\\ &= (2a + b + c)e_1 + (a + 2b + c)e_2 + (a + b + 2c)e_3. \end{align} \] For the other direction, \[ \begin{align} T(T^*(v)) &= T(a e_1 + a e_2 + b e_2 + b e_3 + c e_1 + c e_3)\\ &= T((a + c)e_1 + (a + b)e_2 + (b + c)e_3)\\ &= (a + c)T(e_1) + (a + b)T(e_2) + (b + c)T(e_3)\\ &= (a + c)(e_1 + e_3) + (a + b)(e_1 + e_2) + (b + c)(e_2 + e_3)\\ &= (2a + b + c)e_1 + (a + 2b + c)e_2 + (a + b + 2c)e_3. \end{align} \] Hence, we can see that \(T^*(T(v)) = T(T^*(v))\) for any \(v\in \mathbb{R}^3\). Therefore, we have \(T\) is a normal operator. When \(v = ae_1 + be_2 + ce_3\), we have \[ T(v) = (a + b)e_1 + (b + c)e_2 + (a + c)e_3. \] Thus, we have \(T(a, b, c) = (a + b, b + c, a + c)\). Since \(T(e_1) + T(e_2) - T(e_3) = e_1 + e_3 + e_1 + e_2 - e_2 - e_3 = 2e_1\), we have \(T(e_1 + e_2 - e_3) = 2e_1\). Moreover, we have \(T(e_1) - T(e_2) = -(e_1 + e_2) + e_1 + e_3 = e_3 - e_2\) and \(T(e_3) = e_2 + e_3\). Let \(\mathcal{B} = \{e_1 + e_2 - e_3, e_3, e_1 - e_2\} = \{[1, 1, -1]^T, [0, 0, 1]^T, [-1, 1, 0]^T\}\). We have the matrix of \(T\) with respect to \(\mathbb{B}\) is \[ \begin{pmatrix} 2 & 0 & 0 \\ 0 & 1 & -1 \\ 0 & 1 & 1 \end{pmatrix}. \] Now, we only need to show that \(\mathcal{B}\) is a basis. \[ \begin{align} a\cdot \begin{pmatrix} 1 \\ 1 \\ -1 \end{pmatrix} + b\cdot \begin{pmatrix} 1 \\ -1 \\ 0 \end{pmatrix} + c\cdot \begin{pmatrix} 0 \\ 0 \\ 1 \end{pmatrix} &= \begin{pmatrix} 0 \\ 0 \\ 0 \end{pmatrix}\\ \begin{pmatrix} a + b \\ a - b \\ -a + c \end{pmatrix} &= \begin{pmatrix} 0 \\ 0 \\ 0 \end{pmatrix}\\ \end{align} \] Hence, we have \(a = b = c = 0\), which implies that they are independent. Thus, we have \(\mathcal{B}\) is a basis.

4. Given a field \( K \), consider the group \( GL_n(K) \) of invertible \( n \times n \) matrices with entries in \( K \).

(a) Show that the center of \( GL_n(K) \) is \( \{ \alpha I_n \mid \alpha \in K^\times \} \), where \( I_n \) is the \( n \times n \) identity matrix. Hint: Consider the matrix \( I_n + A_{ij} \), where \( A_{ij} \) is the matrix whose \((i,j)\)-th entry is 1, and all other entries are 0.

(b) Show that \( |SL_2(\mathbb{F}_3)| = 24 \), where \( \mathbb{F}_3 \) is the field with three elements, and \( SL_2(\mathbb{F}_3) \) is the subgroup of \( GL_2(\mathbb{F}_3) \) of \( 2 \times 2 \) matrices with determinant 1.

(c) Use (a) to prove that \( SL_2(\mathbb{F}_3) \) is not isomorphic to the symmetric group \( S_4 \).

Proof (a). Let us denote \(S\) as \( \{ \alpha I_n \mid \alpha \in K^\times \} \). Suppose that \(s\in S\), then we can have \(s = \alpha I_n\), for some \(\alpha\in K\setminus\{0\}\). Let \(g\in GL_n(K)\), then we have \[ g\cdot s\cdot g^{-1} = g\cdot \alpha I_n\cdot g^{-1} = \alpha g\cdot I_n\cdot g^{-1} = \alpha I_n. \] Hence, we showed that \(s\in Z(GL_n(K))\). Now, suppose that \(z\in Z(GL_n(K))\), then we have \[ g\cdot z\cdot g^{-1} = z, \] for any \(g\in GL_n(K)\). Then, we have

Proof (b). For any \(g\in SL_2(\mathbb{F}_3)\), we can know that \[ g = \begin{pmatrix} a & b \\ c & d \end{pmatrix}, \] where \(ad - bc = 1\). Hence, we can count the number of combinations such that \(x - y = 1\) where \(x, y\in \mathbb{F}_3\). We have the following combinations: \[ \begin{align} 0 - 2 &= 1,\\ 1 - 0 &= 1,\\ 2 - 1 &= 1 \end{align} \] Then, we want to count the number of combinations of \(xy = 0, 1, 2\) where \(x, y\in \mathbb{F}_3\). When \(xy = 0\), we have \(5\) combinations, which are \[ 0\cdot 0, 0\cdot 1, 1\cdot 0, 2\cdot 0, 0\cdot 2. \] When \(xy = 1\), we have \(2\) combinations, which are \[ 1\cdot 1, 2\cdot 2. \] When \(xy = 2\), we have \(2\) combinations, which are \[ 1\cdot 2, 2\cdot 1. \] Hence, we can know that there are \(5\cdot 2 = 10\) combinations such that we will have \(0 - 2 = 1\). And there are \(2\cdot 5 = 10\) combinations such that we will have \(1 - 0 = 1\). And there are \(2\cdot 2 = 4\) combinations such that we will have \(2 - 1 = 1\). Therefore, there are \(10 + 10 + 4 = 24\) such elements in \(SL_2(\mathbb{F}_3)\). \(\blacksquare\)

Proof (c). Firstly, according to part (a), we have \(Z(GL_2(\mathbb{F}_3)) = \{ \alpha I_2 \mid \alpha\in \mathbb{F}_3\setminus\{0\} \}\). And we can know that \(Z(GL_2(\mathbb{F}_3))\subset Z(SL_2(\mathbb{F}_3)\). Hence, we have \(\{I_2, 2I_2\}\subset Z(SL_2(\mathbb{F}_3))\). In other words, \(|Z(SL_2(\mathbb{F}_3))|\geq 2\). Now, we want to look at the center of \(S_4\). We start from \(2\)-cycles of \(S_4\), Let \((a, b)\in S_4\) where \(a, b\in \{1, 2, 3, 4\}\). Hence, we can know that there exists \(c\in \{1, 2, 3, 4\}\) such that \(c\neq a\) and \(c\neq b\). Then, there exists \(\sigma\in S_4\) such that \(\sigma (a) = c\). Hence, we have \[ \sigma\cdot (a, b)\cdot \sigma^{-1} = (\sigma(a), \sigma(b)) = (c, \sigma(b))\neq (a, b). \] Hence, we can know that there is no \(2\)-cycle in the center of \(S_4\). Now, we want to look at the \(3\)-cycles of \(S_4\). Suppose that \((a, b, c)\in S_4\), where \(a, b, c\in \{1, 2, 3, 4\}\). Then, we can know that there exists \(d\in \{1, 2, 3, 4\}\) such that \(d\neq a\), \(d\neq b\), and \(d\neq c\). Then, there exists \(\sigma\in S_4\) such that \(\sigma(a) = d\). Hence, we have \[ \sigma\cdot (a, b, c)\cdot \sigma^{-1} = (\sigma(a), \sigma(b), \sigma(c)) = (d, \sigma(b), \sigma(c))\neq (a, b, c). \] Hence, there is no \(3\)-cycle in the center of \(S_4\). Now, we want to look at the \(4\)-cycles of \(S_4\). Suppose that \((a, b, c, d)\in S_4\), where \(a, b, c, d\in \{1, 2, 3, 4\}\). Then, we can know that there exists \(\sigma\in S_4\) where \(\sigma = (b, c)\). Hence, we can know that \[ \sigma\cdot (a, b, c, d)\cdot \sigma^{-1} = (a, c, b, d)\neq (a, b, c, d). \] Hence, there is no \(4\)-cycle in the center of \(S_4\). The last category is the disjoint product of two \(2\)-cycles. Suppose that \((a, b)(c, d)\in S_4\), where \(a, b, c, d\in \{1, 2, 3, 4\}\). Then, we can know that there exists \(\sigma\in S_4\) such that \(\sigma = (b, c)\). We hae \(\sigma\cdot (a, b)(c, d)\sigma^{-1} = (a, c)\neq (b, d)\). Hence, there is no disjoint product of two \(2\)-cycles in the center of \(S_4\). We can know that the center of \(S_4\) is trivial. Hence, we have \(Z(S_4) = \{I_4\}\), which implies that \(|Z(S_4)|\lt |Z(SL_2(\mathbb{F}_3))|\). Therefore, we can know that \(SL_2(\mathbb{F}_3)\not\cong S_4\). \(\blacksquare\)

5. Given a commutative ring \( R \) with 1, the Jacobson radical, \( J(R) \), is the intersection of all maximal ideals of \( R \).

(a) Find the Jacobson radical of \( \mathbb{Z}/12\mathbb{Z} \), and of \( \mathbb{Q}[x,y]/\langle x^2, y-5 \rangle \), with full justification.

(b) Fix an element \( r \) of a commutative ring \( R \) with 1. Prove that \( r \in J(R) \) if and only if for every \( s \in R \), the element \( rs - 1 \) is a unit of \( R \).

Solution (a). Firstly, we want to find all the maximal ideals of \(\mathbb{Z}/12\mathbb{Z}\). According to the \(\textbf{Correspondence Theorem of Rings}\), we can know the maximal ideals \(M\) of \(\mathbb{Z}\), which contains \(12\mathbb{Z}\) can be mapped to \(M/\langle 12 \rangle\) in \(\mathbb{Z}/12\mathbb{Z}\), which is a maximal ideal of \(\mathbb{Z}/12\mathbb{Z}\). Since \(\mathbb{Z}\) is a PID, any ideal \(I\) of \(\mathbb{Z}\) can be written as \(a\mathbb{Z}\) where \(a\in \mathbb{Z}\). If \(12\mathbb{Z}\subset a\mathbb{Z}\), we can know that \(a\mid 12\). Thus, we can know that \(\mathbb{Z}, 2\mathbb{Z}, 3\mathbb{Z}, 4\mathbb{Z}, 6\mathbb{Z}, 12\mathbb{Z}\) are ideals containing \(12\mathbb{Z}\). And \(a\mathbb{Z}\) is a maximal ideal if and only if \(a\mathbb{Z}\) is a prime ideal since \(\mathbb{Z}\) is a PID. And \(a\mathbb{Z}\) is a prime ideal if and only if \(a\) is a prime in \(\mathbb{Z}\). Hence, we can know that \(2\mathbb{Z}, 3\mathbb{Z}\) are the only maximal ideals containing \(12\mathbb{Z}\). Then, we have \(2\mathbb{Z}/12\mathbb{Z}\) and \(3\mathbb{Z}/12\mathbb{Z}\) are the maximal ideals of \(\mathbb{Z}/12\mathbb{Z}\). Since \(2\mathbb{Z}/12\mathbb{Z} = \{0, 2, 4, 6, 8, 10\}\) and \(3\mathbb{Z}/12\mathbb{Z} = \{0, 3, 6, 9\}\), we have \(J(\mathbb{Z}/12\mathbb{Z}) = 2\mathbb{Z}/12\mathbb{Z}\cap 3\mathbb{Z}/12\mathbb{Z} = 6\mathbb{Z}/12\mathbb{Z} = \{0, 6\}\).

Now, we want to identity all the maximal ideals of \(\mathbb{Q}[x, y]/\langle x^2, y - 5 \rangle\). Firstly, we know that if \(a(x, y)\in \langle x^2, y - 5\rangle\), then \[ a(x, y) = f(x, y)\cdot x^2 + g(x, y)\cdot (y - 5) = x\cdot (x\cdot f(x, y)) + g(x, y)\cdot (y - 5) \in \langle x, y - 5\rangle. \] In that case, we can know that \(\langle x^2, y - 5\rangle \subset \langle x, y - 5\rangle\). Now, we want to show that \(\langle x, y - 5\rangle\) is a maximal ideal by showing that \(\mathbb{Q}[x, y]/\langle x, y - 5\rangle\) is a field. Out first step is to show that \(\mathbb{Q}[x, y]/(x, y - 5) \cong \mathbb{Q}[y](y - 5)\). Define a map \(\psi: \mathbb{Q}[x, y]/\langle x, y-5\rangle\to \mathbb{Q}[y](y - 5)\) such that \(\psi(f(x, y) + (x, y - 5)) = f(0, y) + (y - 5)\). Let \(g(x, y), f(x, y)\in \mathbb{Q}[x, y]\), we can have \[ \begin{align} \psi(g(x, y) + f(x, y) + \langle x, y-5\rangle) &= g(0, y) + f(0, y) + (y - 5),\\ \psi(g(x, y)\cdot f(x, y) + (x, y - 5)) &= g(0, y)\cdot f(0, y) + (y - 5). \end{align} \] Hence, we can know that \(\psi\) is a ring homomorphism. Since \(\ker(\psi) = \{f(x, y) + \langle x, y-5\rangle \mid f(0, y) \in \langle y-5\rangle\}\), we can know that \(f(0, y) = (y-5) g(y)\) for some \(g(y)\in \mathbb{Q}[y]\). It means that \(f(x, y) = x\cdot h(x, y) + (y - 5)\cdot g(y)\) for some \(h(x, y)\in \mathbb{Q}[x, y]\), which implies that \(f(x, y)\in \langle x, y - 5\rangle\). Hence, we have \(\ker(\psi) = \langle x, y - 5\rangle\), which is trivial and we are able to conclude that \(\psi\) is an injective ring homomorphism. Now, for any \(f(y) + \langle y-5\rangle\in \mathbb{Q}[y]\), we define \(f(x, y) = f(y) + x\). Then, we can have \[ \psi(f(x, y) + \langle x, y-5\rangle = f(0, y) + (y - 5) = f(y) + (y - 5). \] Hence, we have \(\psi\) is also a surjective ring homomorphism. And we can know that \(\psi\) is an isomorphism. Therefore, we have \[ \mathbb{Q}[x, y]/\langle x, y-5\rangle\cong \mathbb{Q}[y]/\langle y - 5\rangle . \] Then, we will use the \(\textbf{First Isomorphism Theorem}\) to show that \(\mathbb{Q}[y](y - 5)\) is a field. We define a map \(\varphi: \mathbb{Q}[y]\to \mathbb{Q}\) such that \(\varphi(f(x)) = f(5)\). Let \(f(y), g(y)\in \mathbb{Q}[y]\). We can get \[ \begin{align} \varphi(f(y) + g(y)) &= f(5) + g(5),\\ \varphi(f(y)\cdot g(y)) &= f(5)\cdot g(5). \end{align} \] Hence, we can know that \(\varphi\) is a ring homomorphism. For any \(q\in \mathbb{Q}\), we also have \(q\in \mathbb{Q}[y]\) and \(\varphi(q) = q\) for any \(q\in \mathbb{Q}\). Thus, we can see that \(\varphi\) is a surjective ring homomorphism. Now, we want to show that \(\ker(\varphi) = \langle y - 5\rangle\). Suppose that \(h(y)\in \langle y - 5\rangle\), we can know that \(h(y) = (y - 5)\cdot k(y)\) where \(k(y)\in \mathbb{Q}[y]\). Then, we have \(\varphi(h(y)) = h(5) = 0\), which implies that \(h(y)\in \ker(\varphi)\) and \(\langle y - 5\rangle \subset \ker(\varphi)\). For the other direction, suppose that \(f(y)\in \ker(\varphi)\), then we have \(f(5) = 0\), which implies that \(f(y) = (y - 5)\cdot g(y)\) where \(g(y)\in \mathbb{Q}[y]\). Hence, we can know that \(f(y)\in \langle y - 5\rangle\) and \(\ker(\varphi) \subset \langle y - 5\rangle\). Then, we have \(\ker(\varphi) = \langle y - 5\rangle\). According to the \(\textbf{First Isomorphism Theorem}\), we can know that \[ \mathbb{Q}[y]/\ker(\varphi) = \mathbb{Q}[y]/\langle y - 5\rangle \cong \mathbb{Q}. \] Hence, we have shown that \(\mathbb{Q}[x, y]/\langle x, y - 5\rangle \cong \mathbb{Q}/\langle y - 5\rangle \cong \mathbb{Q}\), which is a field. And we can conclude that \(\langle x, y - 5\rangle \) is a maximal ideal. Now, we want to show that if \((x^2, y-5)\subset M\) where \(M\) is a maximal ideal, we can have \(x\in M\) by contradiction. Suppose that \(x\notin M\) and \(\langle x^2, y-5\rangle \subset M\). Hence, we can know that \(M\neq M + \langle x\rangle\) since \(0\in M\) and \(x\in \langle x\rangle\) and \( 0 + x = x\in M + \langle x\rangle \). Given that \(M\) is a maximal ideal, we can conclude that \(M + \langle x \rangle = R = \mathbb{Q}[x, y]\). Since \(1\in \mathbb{Q}[x, y]\), there exists \(f(x, y)\in \mathbb{Q}[x, y]\) and \(m(x, y)\in M\) such that \[ \begin{align} m(x, y) + x\cdot f(x, y) &= 1, \\ m(x, y) &= 1 - x\cdot f(x, y), \\ x\cdot m(x, y) &= x - x^2\cdot f(x, y). \end{align} \] Since \(m(x, y)\in M\), we have \(x\cdot m(x, y)\in M\) and \(x - x^2 \cdot f(x, y)\in M\). Again, we know that \(x^2\in M\) by assumption, so we have \(x^2 \cdot f(x, y)\in M\). Hence, \(x - x^2\cdot f(x, y) + x^2\cdot f(x, y) = x\in M\), which is a contradiction. Therefore, we can know that \(x\in M\). In that case, we can know that for any maximal ideal containing \((x^2, y-5)\), it must contain \(x\) and \(y - 5\). Therefore, we can know that \(\langle x, y-5\rangle \subset M\) for all maximal ideals \(M\) containing \((x^2, y-5)\). And \(\langle x, y-5\rangle\) is a maximal ideal, which implies that there is only one maximal ideal containing \((x^2, y-5)\), which is \(\langle x, y-5\rangle\). According to the \(\textbf{Correspondence Theorem of Rings}\), we can know that \(J(\mathbb{Q}[x, y]/\langle x^2, y-5\rangle) = \langle x, y-5\rangle/\langle x^2, y-5\rangle \).

Proof (b). We firstly showed that if \(r\in J(R)\), then we can know that \(rs - 1\) for any \(s\in R\). We know that if \(m\in M\) where \(M\) is a maximal ideal then we want to show that \(m - 1\notin M\) by contradiction. Suppose that \(m - 1\in M\). We have \(m - 1 + m = - 1\in M\), which implies that \(1\in M\). Since \(M\) is a proper ideal, we know it does not contain a unit. Hence, it is a contradiction. Thus, we can know that \(m - 1\) is not contained in \(M\). Suppose that \(a\in J(R)\), then we can know that \(a\) is in every maximal ideal of \(R\). Hence, we can know that \(a - 1\) is not in any maximal ideal of \(R\). Since \(a - 1\) is not in any maximal ideal, we can know that \(\langle a - 1\rangle \not\subset M\) for any maximal ideal \(M\). Hence, we can know that \(\langle a - 1\rangle\) is not a proper ideal of \(R\), which implies that \(\langle a - 1\rangle = R\). Hence, we can know that \(a - 1\) is a unit. Since \(r\in J(R)\) adn \(J(R)\) is an ideal, we can know that that \(rs \in J(R)\) for any \(s \in R\). Since we already showed that \(a - 1\) is a unit if \(a \in J(R)\), we can have \(rs - 1\) is a unit.

For the other direction, suppose that \(rs - 1\) is a unit for any \(s\in R\). Then, we can know that \(r\cdot 1 - 1 = r- 1\) is a unit. Then, we can know that \(r\) is not a unit by contradiction. Suppose that \(r\) is a unit then there exists \(a\in R\) such that \(ra = 1\). It means that \(ra - 1 = 0\) for some \(a\in R\), which implies that \(ra - 1\) is not a unit. Thus, it is a contradiction. Thus, \(r\) is not a unit. Then, we can know that \(\langle r\rangle\) is a proper ideal of \(R\). Now, we want to show that \(\langle r \rangle \subset M\) for any maximal ideal \(M\) by contradiction. Suppose that there exists a maximal ideal \(M\) such that \(\langle r \rangle \not\subset M\). Then, we can know that \(\langle r \rangle + M \) is an ideal strictly containing \(M\). Since \(M\) is a maximal ideal, we can know that \(\langle r \rangle + M = R\). Thus, we can know that there exists \(m\in M\) and \(ra\in \langle r \rangle\) such that \(mb + rab = 1\) for some \(b\in R\). It shows that \(rab - 1 = -mb\). Since \(rab - 1\) is a unit by assumption, we can know that \(-mb\) is a unit. However, \(-mb\in M\) and \(M\) is a proper ideal, which implies that \(-mb\) is not a unit. Thus, it is a contradiction. Therefore, \(\langle r \rangle \subset M\) for any maximal ideal \(M\). Hence, we can know that \(r\in J(R)\). \(\blacksquare\)