Homework Archive¶
Week 1¶
Question
(i) Write out an addition table for \(\mathbf{Z}_2 \oplus \mathbf{Z}_2\). \(\mathbf{Z}_2 \oplus \mathbf{Z}_2\) is called the Klein four group.
\(\star\)(ii) Proof the following conditions on a group \(G\) are equivalent:
(a) \(G\) is Abelian;
(b) \((ab)^2=a^2b^2, \forall a, b \in G\);
(c) \((ab)^{-1}=a^{-1}b^{-1}, \forall a, b \in G\);
(d) \((ab)^n=a^nb^n, \forall n \in \mathbf{Z}, a, b \in G\);
(e) \((ab)^n=a^nb^n\) for three consectuive integers \(n\) and all \(a, b \in G\).
And show that \((e) \Rightarrow (a)\) is false if "three" is replaced by "two".
(iii) Let \(Q_8\) be the group (under ordinary matrix multiplication) generated by the complex matrices \(A = \begin{pmatrix} 0 & 1 \\ -1 & 0 \end{pmatrix}\) and \(B = \begin{pmatrix} 0 & i \\ i & 0 \end{pmatrix}\), where \(i^2=-1\). Show that \(Q_8\) is a nonabelian group of order 8. \(Q_8\) is called the quaternion group.[Hint: Observe that \(BA = A^3B\), whence every element of \(Q_8\) is of the form \(A^iB^j\). Note also that \(A^4=B^4=I\), where \(I = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix}\) is the identity element of \(Q_8\).]
\(\star\)(iv) List all subgroups of \(\mathbf{Z}_2 \oplus \mathbf{Z}_2\). Is \(\mathbf{Z}_2 \oplus \mathbf{Z}_2\) isomorphic to \(\mathbf{Z}_4\)?
\(\star\)(v) Let \(G\) be an abelian group containing elements \(a\) and \(b\) of orders \(m\) and \(n\) respectively. Show that \(G\) contains an element whose order is the least common multiple of \(m\) and \(n\). [Hint: First try the case when \((m,n) = 1\).]
(vi) Let \(G\) be the multiplicative group of all nonsingular \(2 \times 2\) matrices with rational entries. Show that \(a = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix}\) has order 4 and \(b = \begin{pmatrix} 0 & 1 \\ -1 & -1 \end{pmatrix}\) has order 3, but \(ab\) has infinite order. Conversely, show that the additive group \(\mathbf{Z}_2 \oplus \mathbf{Z}\) contains nonzero elements \(a, b\) of infinite order such that \(a+b\) has finite order.
Answer
(i) The addition table is shown as below.
\(+\) | \((\overline{0}, \overline{0})\) | \((\overline{0}, \overline{1})\) | \((\overline{1}, \overline{0})\) | \((\overline{1}, \overline{1})\) |
\((\overline{0}, \overline{0})\) | \((\overline{0}, \overline{0})\) | \((\overline{0}, \overline{1})\) | \((\overline{1}, \overline{0})\) | \((\overline{1}, \overline{1})\) |
\((\overline{0}, \overline{1})\) | \((\overline{0}, \overline{1})\) | \((\overline{0}, \overline{0})\) | \((\overline{1}, \overline{1})\) | \((\overline{1}, \overline{0})\) |
\((\overline{1}, \overline{0})\) | \((\overline{1}, \overline{0})\) | \((\overline{1}, \overline{1})\) | \((\overline{0}, \overline{0})\) | \((\overline{0}, \overline{1})\) |
\((\overline{1}, \overline{1})\) | \((\overline{1}, \overline{1})\) | \((\overline{1}, \overline{0})\) | \((\overline{0}, \overline{1})\) | \((\overline{0}, \overline{0})\) |
(ii) (a) \(\Leftrightarrow\) (b): If \(G\) is Abelian, then \((ab)^2 = (ab)(ab) = a(ba)b = a(ab)b = a^2b^2, \forall a, b \in G\); We know that \((ab)^2 = abab\). If \((ab)^2 = a^2b^2\), then we have \(a^2b^2 = aabb = abab\). Using the cancellation law, we have \(ab = ba, \forall a, b \in G\), which means \(G\) is Abelian.
(a) \(\Leftrightarrow\) (c): If \(G\) is Abelian, then \((ab)^{-1} = b^{-1}a^{-1} = a^{-1}b^{-1}, \forall a, b \in G\); If \((ab)^{-1} = a^{-1}b^{-1} = b^{-1}a^{-1}\), then \((a^{-1})^{-1}(b^{-1})^{-1} = ab = ba = (b^{-1})^{-1}(a^{-1})^{-1}, \forall a, b \in G\).
(a) \(\Leftrightarrow\) (d): Proof by induction.If \(G\) is Abelian, then \(ab = ba, \forall a, b \in G\). Suppose that \((ab)^n = a^nb^n, \forall n \in \mathbf{N}^*\). So \((ab)^{n+1} = (ab)(ab)^n = (ab)a^nb^n = a(ba^n)b^n = aa^nbb^n = a^{n+1}b^{n+1}, \forall n \in \mathbf{N}^*, a, b \in G\). We also know that \((ab)^{-1} = b^{-1}a^{-1} = a^{-1}b^{-1}, \forall a, b \in G\). So suppose \((ab)^{-n} = a^{-n}b^{-n}, \forall n \in \mathbf{N}^*\). Then \((ab)^{-(n+1)} = (ab)^{-1}(ab)^{-n} = a^{-1}b^{-1}a^{-n}b^{-n} = a^{-1}a^{-n}b^{-1}b^{-n} = a^{-(n+1)}b^{-(n+1)}, \forall n \in \mathbf{N}^*, a, b \in G\). And \((ab)^0 = e = e^2 = a^0b^0\). So \((ab)^n=a^nb^n, \forall n \in \mathbf{Z}, a, b \in G\); If \((ab)^n=a^nb^n, \forall n \in \mathbf{Z}, a, b \in G\), then let \(n = 2\). From (b) we can easily proof that \(G\) is Abelian.
(a) \(\Leftrightarrow\) (e) If \(G\) is Abelian, it's obvious(from (d)). If \((ab)^n=a^nb^n\) for three consectuive integers \(n\) and all \(a, b \in G\), which means \((ab)^k = a^kb^k, (ab)^{k+1} = a^{k+1}b^{k+1}, (ab)^{k+2} = a^{k+2}b^{k+2}, \exists k \in \mathbf{Z}\). So \((ab)^{k+1} = (ab)(ab)^k = aba^kb^k = a^{k+1}b^{k+1}\). Using the cancellation law, we have \(ba^k = a^kb\). And \((ab)^{k+2} = (ab)^2(ab)^k = ababa^kb^k = a^{k+2}b^{k+2}\). Using the cancellation law, we have \(ba = a^{k+1}b^2a^{-k}b^{-1} = a(a^kb)ba^{-k}b^{-1} = ab(a^kb)a^{-k}b^{-1} = abba^ka^{-k}b^{-1} = ab, \forall a, b \in G\). So \(G\) is Abelian.
It's obvious that \((e) \Rightarrow (a)\) is false if "three" is replaced by "two", for \((ab)^0 = a^0b^0\) and \((ab)^1 = a^1b^1\) are always correct.
(iii) \(A = \begin{pmatrix} 0 & 1 \\ -1 & 0 \end{pmatrix}\), \(A^2 = \begin{pmatrix} -1 & 0 \\ 0 & -1 \end{pmatrix} = -I\), \(A^3 = -A\), \(A^4 = I\).
\(B = \begin{pmatrix} 0 & i \\ i & 0 \end{pmatrix}\), \(B^2 = \begin{pmatrix} -1 & 0 \\ 0 & -1 \end{pmatrix} = -I\), \(B^3 = -B\), \(B^4 = I\).
\(BA = \begin{pmatrix} -i & 0 \\ 0 & i \end{pmatrix}\), \(AB = \begin{pmatrix} i & 0 \\ 0 & -i \end{pmatrix}\). \(BA = -AB = (-I)AB = A^2AB = A^3B\).
So \(B^iA^j = A^kB^i, \forall i, j, \exists k \in \mathbf{Z}\), whence every element of \(Q_8\) is of the form \(A^iB^j\). From this, we have \(Q_8 = \{I, A, A^2 = -I = B^2, A^3 = -A, B, AB, A^2B = -B = B^3, A^3B = -AB\}\). So \(Q_8\) is a nonabelian group of order 8.
(iv) 列出子群由 (i) 可知 . 二者不同构从元素的阶考虑 .
(v) 对 \(m, n\) 进行素分解,设 \(m = \prod_i p_i^{\alpha_i}\), \(n = \prod_i p_i^{\beta_i}\). 这里将二者素分解所使用的素数保持一致,实际上未起作用的素数的指数记为 \(0\). 进而记 \(m' = \prod_{i: \alpha_i \geqslant \beta_i} p_i^{\alpha_i}\), \(n' = \prod_{i: \alpha_i < \beta_i} p_i^{\beta_i}\). \(a' = a^{m/m'}\), \(b' = b^{n/n'}\). 则 \(m', n'\) 是互素的 , 且 \(m'n'\) 是 \(m, n\) 的最小公倍数 . 下面证明 \(a', b'\) 的阶是分别是 \(m', n'\).
设 \(k\) 是 \(a'\) 的阶, 则 \((a')^k = (a^{m/m'})^k = a^{mk/m'} = e\), 所以 \(m \mid mk/m'\), 也就有 \(m' \mid k\). 而 \(a'^{m'} = (a^{m/m'})^{m'} = e\), 也就有 \(k \mid m'\), 所以 \(k = m'\), \(a'\) 的阶是 \(m'\). 同理可证 \(b'\) 的阶是 \(n'\). 下面证明 \(a'b'\) 的阶是 \(m'n'\).
类似地, 设 \(a'b'\) 的阶是 \(r\), 因为 \((a'b')^{m'n'} = (a^{m/m'})^{m'n'}(b^{n/n'})^{m'n'} = a^{mn'}b^{nm'} = e\), 所以 \(r \mid m'n'\). 而 \(e = (a'b')^{r} = (a'b')^{rm'} = a'^{rm'}b'^{rm'} = b'^{rm'}\), 所以 \(n' \mid rm'\), \(m', n'\) 互素, 有 \(n' \mid r\). 同理可证 \(m' \mid r\). 所以 \(m'n' \mid r\), \(r = m'n'\). 也就是说 \(a'b'\) 的阶是 \(m'n'\), \(m, n\) 的最小公倍数.
(vi) \(a = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix}\), \(a^2 = \begin{pmatrix} -1 & 0 \\ 0 & -1 \end{pmatrix} = -I\), \(a^3 = -a\), \(a^4 = I\). \(b = \begin{pmatrix} 0 & 1 \\ -1 & -1 \end{pmatrix}\), \(b^2 = \begin{pmatrix} -1 & -1 \\ 1 & 0 \end{pmatrix}\), \(b^3 = I\). 而 \(ab = \begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix} \begin{pmatrix} 0 & 1 \\ -1 & -1 \end{pmatrix} = \begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix}\). 易知 \((ab)^k = \begin{pmatrix} 1 & k \\ 0 & 1 \end{pmatrix} \neq I\), \(\forall k \in \mathbf{Z}\). 所以 \(ab\) 的阶为无穷大 .
考虑 \(\mathbf{Z} \oplus \mathbf{Z}_2\) 中的 \(a = (\overline{1}, \overline{1}), b = (\overline{0}, \overline{-1})\), 有 \(a + b = (\overline{1}, \overline{0})\) 的阶是 2, 且 \(a, b\) 的阶都是无穷大.
Week 2¶
Question
(i) (Euler-Fermat) Let \(a\) be a integer and \(p\) a prime such that \(p \nmid a\). Then \(a^{p-1} \equiv 1 \pmod p\).[Hint: Consider \(\overline{a} \in \mathbf{Z}_p\) and the multiplicative group of nonzero elements of \(\mathbf{Z}_p\).] It follows that \(a^p \equiv a \pmod p\) for any integer \(a\).
(ii) Let \(G\) be a group of order \(2n\); then \(G\) contains an element of order 2. If \(n\) is odd and \(G\) abelian, there is only one element of order 2.
(iii) (a) If \(G\) is a group, then the center of \(G\) is a normal subgroup of \(G\).(Center: \(C=\{a \in G \mid ax=xa, \forall x \in G\}\).)
(b) The center of \(S_n\) is the identity subgroup for all \(n>2\).
(iv) If \(f: G \rightarrow H\) is a homomorphism with kernel \(N\) and \(K<G\), then prove that \(f^{-1}(f(K))=KN\). Hence \(f^{-1}(f(K))=K\) iff \(N<K\).
(v) Show that \(N=\{(1), (12)(34), (13)(24), (14)(23)\}\) is a normal subgroup of \(S_4\) contained in \(A_4\) such that \(S_4/N \cong S_3\) and \(A_4/N \cong Z_3\).
(vi) Find all normal subgroups of \(D_n\).
Answer
(i) 考虑模 \(p\) 剩余类构成的乘法群 , 因为 \(p\) 是素数 , 所以该乘法群的阶为 \(p - 1\). 且对于任意 \(a \in \mathbf{Z}_p\), 考虑 \(c_1, c_2 \in \mathbf{Z}_p\), 则必有 \((c_1 - c_2)a \not \equiv 0 \pmod p\). 也就有 \(c_1a \not \equiv c_2a \pmod p\). 所以由 \(a\) 诱导的映射 \(f_a: \mathbf{Z}_p \rightarrow \mathbf{Z}_p\) 是一个内自同构 . 也就是 \(a, 2a, \ldots, (p - 1)a\) 是 \(\mathbf{Z}_p\) 的一个置换 . 从而 \(\prod_{i = 1}^{p - 1} ia = (p - 1)!a^{p - 1} \equiv (p - 1)! \pmod p\). 由于 \(p \nmid a\), 所以 \(a^{p - 1} \equiv 1 \pmod p\).
(ii) 利用反证法 .
(a) 设 \(G\) 中没有阶为 \(2\) 的元素, 即 \(\forall a \neq e \in G\), \(a^2 \neq e\). 也就是说 \(\forall a (\neq e) \in G\), \(a^{-1} \neq a\). 而考虑 \(e\) 的逆元就是本身, 剩下的 \(2n-1\) 个元素无法进行逆元配对, 所以 \(G\) 中必有阶为 \(2\) 的元素.
(b) 设 \(G\) 中有至少两个元素的阶为 \(2\), 设为 \(a, b\), 则由于 \(G\) 的可交换性, \(H = \{e, a, b, ab\}\) 构成了 \(G\) 的一个子群, \(\lvert G \rvert = [G:H]\lvert H \rvert\). 从而 \(4 \mid 2n\), 但 \(n\) 是奇数, 矛盾. 所以 \(G\) 中阶为 \(2\) 的元素只有一个.
(iii) (a) \(\forall c \in C\), 由中心的定义有 \(gc = cg\), \(\forall g \in G\). 所以 \(gC = Cg\), \(\forall g \in G\). 也就是说 \(C\) 是 \(G\) 的一个正规子群 .
(b) 反证法. 设 \(S_n\) 的中心包含不止 \(e\) 一个元素, 即包含一个置换 \(\varphi\), 存在 \(i\) 使得 \(\varphi(i) = j \neq i\). 因为 \(n > 2\), 所以存在 \(k \neq i, k \neq j\), 并且存在一个置换 \(\psi\) 使得 \(\psi(i) = k, \psi(k) = i\). 从而 \(\varphi \psi(i) = \varphi(k) \neq j\), \(\psi \varphi(i) = \psi(j) = j\). 也就是说 \(\varphi \psi \neq \psi \varphi\), 矛盾. 所以 \(S_n\) 的中心只有 \(e\) 一个元素.
(iv) \(f^{-1}(f(K)) = \{g \in G \mid f(g) = f(k), k \in K\}\). 而 \(f(g) = f(k)\) 有 \(e_H = f(g)f^{-1}(k) = f(gk^{-1})\), 所以 \(gk^{-1} \in N\). 也就是说 \(g \in kN \subset KN\), 从而 \(f^{-1}(f(K)) \subset KN\). 另一方面 , 设 \(kn \in KN\), \(f(kn) = f(k)f(n) = f(k)\), 所以 \(f^{-1}(f(kn)) = f^{-1}(f(k)) \in f^{-1}(f(K))\). 从而 \(KN \subset f^{-1}(f(K))\). 所以 \(f^{-1}(f(K)) = KN\).
若 \(N < K\) 等价于 \(KN = K\), 显然成立.
(v) 首先证明 \(N\) 是 \(S_4\) 和 \(A_4\) 的正规子群 . 注意到对两两互异的 \(i, j, k, l \in \{1, 2, 3, 4\}\) 有 :
(a) \(e = (1) \in N\);
(b) \(((ij)(kl))^2 = e\), 也就是说 \(((ij)(kl))^{-1} = ((ij)(kl)) \in N\);
(c) \(((ij)(kl)) \circ ((ik)(jl)) = (il)(jk)\).
设 \(\delta \in N, \sigma \in S_4\).
(a) 若 \(\delta = e\), 则 \(\sigma \delta \sigma^{-1} = e \in N\);
(b) 若 \(\delta \neq e\), 则 \(\delta\) 可以分解为两不交对换的乘积.
若 \(\sigma\) 是偶的, 则其可被分解为若干个两两不交的对换的乘积, 而任意对换 \(\epsilon\) 都有 \(\epsilon^{2n} = e\), \(\epsilon^{2n + 1} = \epsilon\), 根据前面给出的三条关于对换运算的性质, 可以得出 \(\sigma \delta \sigma^{-1} \in N\). \(N\) 是 \(A_4\) 的正规子群.
若 \(\sigma\) 是奇的, 则此处有 \(\sigma = (ijk)\), \(i, j, k \in \{1, 2, 3, 4\}\) 且两两互异. 而 \((ijk) = (ik)(ij)\). 故 \(\sigma \delta \sigma^{-1} \in N\). \(N\) 是 \(S_4\) 的正规子群.
下面证明同构. 由 Lagrange 定理, \(\lvert S_4/N \rvert = 6\), 所以 \(S_4/N\) 同构于循环群 \(Z_6\) 或 \(S_3\). 事实上只要证明元素的阶即可. \((1)N = N\), \((12)N, (13)N, (23)N\) 的阶均为 \(2\), \((123)N, (132)N\) 的阶均为 \(3\). 所以 \(S_4/N \cong S_3\).
同理由 Lagrange 定理, \(\lvert A_4/N \rvert = 3\), 所以 \(A_4/N\) 同构于循环群 \(Z_3\).
(vi) 见 Dihedral Group Theorem 3.8.
感觉是看到的讲二面体群最清晰的一篇文章了.
Tip
元素 \(r^is\) 实际上也是一个对称操作 , 可以考虑几何上的 1 号顶点对称轴的变换 . 对称操作阶均为 \(2\).
Week 3¶
Question
(i) A subset \(X\) of an abelian group \(F\) is said to be linearly independent if \(n_1x_1 + \cdots + n_kx_k = 0\) always implies \(n_i = 0\) for all \(i\) (where \(n_i \in \mathbf{Z}\) and \(x_1, \ldots ,x_k\) are distinct elements of \(X\)).
(a) \(X\) is linearly independent if and only if every nonzero element of the subgroup \(\langle X \rangle\) may be written uniquely in the form \(n_1x_1 + \cdots + n_kx_k\) (\(n_i \in \mathbf{Z}, n_i \neq 0, x_1, \ldots ,x_k\) distinct elements of \(X\)).
(b) If \(F\) is free abelian of finite rank \(n\), it is not true that every linearly independent subset of \(n\) elements is a basis [Hint: consider \(F = \mathbf{Z}\)].
(c) If \(F\) is free abelian, it is not true that every linearly independent subset of \(F\) may be extended to a basis of \(F\).
(d) If \(F\) is free abelian, it is not true that every generating set of \(F\) contains a basis of \(F\). However, if \(F\) is also finitely generated by \(n\) elements, \(F\) has rank \(m \leqslant n\).
\(\star\)(ii) (a) Show that the additive group of rationals \(\mathbf{Q}\) is not finitely generated.
(b) Show that \(\mathbf{Q}\) is not free.
(c) Conclude that Exercise 9 is false if the hypothesis "finitely generated" is omitted.
Exercise 9: Let \(G\) be a finitely generated abelian group in which no element (except 0) has finite order. Then \(G\) is a free abelian group.
(iii) Let \(k, m \in N^*\). If \((k,m) = 1\), then \(k\mathbf{Z}_m = \mathbf{Z}_m\) and \(\mathbf{Z}_m[k] = 0\). If \(k \mid m\), say \(m = kd\), then \(k\mathbf{Z}_m \cong \mathbf{Z}_d\) and \(\mathbf{Z}_m[k] \cong \mathbf{Z}_k\)
\(\star\)(iv) (a) What are the elementary divisors of the group \(\mathbf{Z}_2 \oplus \mathbf{Z}_9 \oplus \mathbf{Z}_{35}\); what are its invariant factors? Do the same for \(\mathbf{Z}_{26} \oplus \mathbf{Z}_{42} \oplus \mathbf{Z}_{49} \oplus \mathbf{Z}_{200} \oplus \mathbf{Z}_{1000}\).
(b) Determine up to isomorphism all abelian groups of order 64; do the same for order 96.
(c) Determine all abelian groups of order \(n\) for \(n \leqslant 20\).
Answer
(i) (a) 因为 \(F\) 是交换的 , 所以 \(\langle X \rangle\) 也是交换的 , 设 \(X\) 中元素的下标组成指标集 \(I\), 则 \(\langle X \rangle\) 中的元素都能写作 \(\sum_{i \in I} n_ix_i\) 的形式 .
假如存在元素 \(\alpha\) 满足 \(\alpha = \sum_{i \in I} n_ix_i = \sum_{i \in I} m_ix_i\), 则 \(\sum_{i \in I} (n_i - m_i)x_i = 0\).
(\(\Rightarrow\)) 若 \(X\) 线性无关, 则 \(n_i - m_i = 0\), \(\forall i \in I\), 也就是说 \(n_i = m_i\), \(\forall i \in I\). 即 \(\langle X \rangle\) 中的元素的表示方式唯一.
(\(\Leftarrow\)) 若 \(X\) 中所有元素均可被唯一标识, 由于 \(0 = \sum_{i \in I} 0x_i\), 所以 \(X\) 线性无关.
(b) 考虑整数集 \(\mathbf{Z}\), 其是一秩为 \(1\) 的自由阿贝尔群, 而我们知道 \(2\) 是线性无关的, 但我们用 \(2\) 只能生成偶数集 \(2 \mathbf{Z}\), 即 \(2\) 不是 \(\mathbf{Z}\) 的基.
(c) 继续考虑 (b) 中例子.
(d) 继续考虑整数集 \(\mathbf{Z}\). 因为 \(n = n \cdot 1 = n \cdot ((-2) + 3) = n \cdot(-2) + n \cdot 3\). 但 \(\{-2, 3\}\) 并不是 \(\mathbf{Z}\) 的基, 所以 \(\mathbf{Z}\) 可由不是基的生成集有限生成.
若阿贝尔群 \(F\) 可由集合 \(X = \{x_1, \ldots, x_n\}\) 有限生成, 则 \(\forall \alpha \in F\), 可被写作 \(\alpha = \sum_{i = 1}^n k_ix_i\), \(k_i \in \mathbf{Z}\). \(X\) 若线性无关, 则 \(F\) 的秩为 \(n\). 若 \(X\) 线性相关, 则存在 \(m_i\) 不全为 \(0\) 使得 \(\sum_{i = 1}^n k_ix_i = 0\). 若 \(F\) 有秩 \(m > n\) 则其有基 \(Y = \{y_1, \ldots, y-M\}\), 设 \(\sum_{i = 1}^m l_iy_i = 0\), 则 \(\sum_{i = 1}^m l_i \sum_{j = 1}^n r_j x_j = \sum_{j = 1}^n (\sum_{i = 1}^m l_i r_j) x_j = 0\). 令 \(\sum_{i = 1}^m l_i r_j = m_j, j = 1, \ldots, n\), 该方程组有非零解, 所以 \(Y\) 线性相关, 矛盾. 所以 \(F\) 的秩不会大于 \(n\).
(ii) (a) 设 \(\mathbf{Q}\) 是有限生成的 . 设其生成集为 \(X = \{\dfrac{n_1}{d_1}, \ldots, \dfrac{n_k}{d_k}\}\). 设素数 \(p \nmid \prod_{i = 1}^k d_i\), 则 \(X\) 无法生成 \(\dfrac{1}{p}\), 矛盾 .
(b) 任取 \(x, y \in \mathbf{Q}\), \(x, y \neq 0\), 存在 \(\lambda, \mu \in \mathbf{Z}\) 使得 \(\lambda x + \mu y = 0\). 所以 \(x, y\) 线性相关, 考虑到 \(\mathbf{Q}\) 不是循环群, 所以 \(\mathbf{Q}\) 不可能有一组基, 也就不是自由阿贝尔群.
(c) \(\mathbf{Q}\) 就是反例.
(iii) (a) \((k, m) = 1\) 时 , \(k\mathbf{Z}_m = \{kn \mid n \in \mathbf{Z}_m\}\). 而 \(\mathbf{Z}_m = \{\overline{0}, \overline{1}, \ldots, \overline{m-1}\}\). 若 \(k\overline{a} = k\overline{b} \pmod m\), 因为 \((k, m) = 1\), 所以有 \(\overline{a} = \overline{b} \pmod m\). 所以由 \(k\) 给出的同态 \(f_k\) 实际上是一个内自同构 , \(k\mathbf{Z}_m = \mathbf{Z}_m\). 而 \(\mathbf{Z}_m[k] = \{n \in \mathbf{Z} \mid kn \equiv 0 \pmod m\}\), 由于 \(k\overline{0} \equiv 0 \pmod m\), 且 \(f_k\) 是内自同构 , 所以 \(\mathbf{Z}_m[k] = 0\).
(b) \(k \mid m\), 设 \(m = kd\). 因为 \(k(id) = i(kd) = im \equiv 0 \pmod m\), \(i = 0, 1, \ldots, k-1\). 而当 \(k \leqslant i \leqslant m\) 时, 由带余除法有 \(i = l \cdot k + r\), \(0 \leqslant r < k\), 所以 \(k(id) = k(l \cdot k + r)d = k^2ld + krd = kml + krd \equiv krd \pmod m\), \(0 \leqslant krd < m\). 所以只用考虑 \(i = 0, 1, \ldots, k-1\) 的情况, \(\lvert \mathbf{Z}_m[k] \rvert = \lvert \mathbf{Z}_k \rvert = k\). 考虑同态 \(\varphi: \mathbf{Z}_m[k] \to \mathbf{Z}_k\), \(\varphi(id) = i\). 因为 \(i = 0\) 时, \(id = 0\), 所以 \(\varphi\) 是单射, 进而 \(\mathbf{Z}_m[k] \cong \mathbf{Z}_k\). 考虑 \(k\mathbf{Z}_m = \{kn \mid n \in \mathbf{Z}_m\}\). 由带余除法 \(n = jd + r\), \(j = 0, 1, \ldots, k-1\), \(0 \leqslant r < d\), 所以 \(kn = kjd + kr \equiv kr \pmod m\). 定义 \(\psi: k\mathbf{Z}_m \to \mathbf{Z}_d\), \(\psi(kr) = r\), 这是一个同构映射, 所以 \(k\mathbf{Z}_m \cong \mathbf{Z}_d\).
(iv) (a) \(\mathbf{Z}_2 \oplus \mathbf{Z}_9 \oplus \mathbf{Z}_{35}\) 的初等因子为 \(2, 3^2, 5, 7\), 不变因子为 \(3, 210\); \(\mathbf{Z}_{26} \oplus \mathbf{Z}_{42} \oplus \mathbf{Z}_{49} \oplus \mathbf{Z}_{200} \oplus \mathbf{Z}_{1000}\) 的初等因子为 \(2, 2, 2^3, 2^3, 3, 5^2, 5^3, 7, 7^2, 13\), 排列为
不变因子为每行乘积 , 即 \(2, 2, 1400, 1911000\).
(b)
(c)
\(n\) | \(1\) | \(2\) | \(3\) | \(4\) |
Group | \(Z_1\) | \(Z_2\) | \(Z_3\) | \(Z_2 \oplus Z_2, Z_4\) |
\(n\) | \(5\) | \(6\) | \(7\) | \(8\) |
Group | \(Z_5\) | \(Z_2 \oplus Z_3\) | \(Z_7\) | \(Z_2 \oplus Z_2 \oplus Z_2, Z_2 \oplus Z_4, Z_8\) |
\(n\) | \(9\) | \(10\) | \(11\) | \(12\) |
Group | \(Z_3 \oplus Z_3, Z_9\) | \(Z_2 \oplus Z_5\) | \(Z_{11}\) | \(Z_2 \oplus Z_2 \oplus Z_3, Z_4 \oplus Z_3\) |
\(n\) | \(13\) | \(14\) | \(15\) | \(16\) |
Group | \(Z_{13}\) | \(Z_2 \oplus Z_7\) | \(Z_3 \oplus Z_5\) | \(Z_2 \oplus Z_2 \oplus Z_2 \oplus Z_2, Z_2 \oplus Z_2 \oplus Z_4,\) \(Z_2 \oplus Z_8, Z_4 \oplus Z_4, Z_{16}\) |
\(n\) | \(17\) | \(18\) | \(19\) | \(20\) |
Group | \(Z_{17}\) | \(Z_2 \oplus Z_3 \oplus Z_3\), \(Z_2 \oplus Z_9\) |
\(Z_{19}\) | \(Z_2 \oplus Z_2 \oplus Z_5, Z_4 \oplus Z_5\) |
Week 4¶
Question
\(\star\) (i) If \(\lvert G \rvert = p^nq\), with \(p > q\) primes, then \(G\) contains a unique normal subgroup of index \(q\).
(ii) Every group \(G\) of order \(p^2\) (\(p\) prime) is abelian [Hint: Exercise 4.9 and Corollary 5.4].
Exercise 4.9: If \(G/C(G)\) is cyclic, then \(G\) is abelian.
Corollary 5.4: The center \(C(G)\) of a nontrivial finite \(p\)-group \(G\) contains more than one element.
(iii) Consider the set \(G = \{\pm 1, \pm i, \pm j, \pm k\}\) with multiplication given by \(i^2 = j^2 = k^2 = -1\); \(ij = k; jk = i, ki = j\); \(ji = -k, kj = -i, ik = -j\), and the usual rules for multiplying by \(\pm 1\). Show that \(G\) is a group isomorphic to the quaternion group \(Q_8\).
(iv) (a) Show that there is a nonabelian subgroup \(T\) of \(S_3 \times \mathbf{Z}_4\) of order 12 generated by elements \(a, b\) such that \(\lvert a \rvert = 6\), \(a^3 = b^2\), \(ba = a^{-1}b\).
(b) Any group of order 12 with generators \(a, b\) such that \(\lvert a \rvert = 6\), \(a^3 = b^2\), \(ba = a^{-1}b\) is isomorphic to \(T\).
Answer
(i) \(\lvert G \rvert = p^nq\), \(p, q\) 均为素数且 \((p, q) = 1\). 考虑 \(p\), 对 \(G\) 应用 Sylow- 第一定理 , 则 \(G\) 存在一个阶为 \(p^n\) 的子群 \(H\), 即为 \(G\) 的一个 Sylow-\(p\) 子群 , 其指标为 \([G:H] = \lvert G \rvert / \lvert H \rvert = q\). 由 Sylow- 第三定理 , \(G\) 中的 Sylow-\(p\) 子群的个数为 \(n_p\), \(n_p \equiv 1 \pmod p\), 且 \(n_p \mid \lvert G \rvert\), 所以 \(n_p \mid q\). 但若 \(n_p = q\), 因为 \(p > q\), 所以 \(n_p \not \equiv 1 \pmod p\), 矛盾 . 所以 \(n_p = 1\). 也就是说 \(H\) 是 \(G\) 的唯一的 Sylow-\(p\) 子群 . 又 \([G:H] = q\), \(q\) 是最小的整除 \(G\) 阶的素数 , 所以 \(H\) 是正规子群 .
(ii) 考虑 \(\lvert G \rvert = p^2\), 因为 \(p \mid \lvert C(G) \rvert \mid \lvert G \rvert\), 所以 \(\lvert C(G) \rvert = p\) 或 \(\lvert C(G) \rvert = p^2\). 若 \(\lvert C(G) \rvert = p^2\), 则 \(C(G) = G\), \(G\) 交换 . 若 \(\lvert C(G) \rvert = p\), 则 \(\lvert G/C(G) \rvert = p\), 但
Week 5¶
Question
(i) (a) \(A_4\) is not the direct product of its Sylow subgroups, but \(A_4\) does have the property: \(mn = 12\) and \((m, n) = 1\) imply there is a subgroup of order \(m\).
(b) \(S_3\) has subgroups of orders 1, 2, 3, and 6 but is not the direct product of its Sylow subgroups.
(ii) If \(D_n\) is the dihedral group with generators \(a\) of order \(n\) and \(b\) of order \(2\), then
(a) \(a^2 \in D_n'\).
(b) If \(n\) is odd, \(D_n' \cong Z_n\).
(c) If \(n\) is even, \(Dn' \cong Z_m\), where \(2m = n\).
(d) \(D_n\) is nilpotent if and only if \(n\) is a power of 2.
(iii) (a) Find a normal series of \(D_4\) consisting of 4 subgroups.
(b) Find all composition series of the group \(D_4\).
(c) Do part (b) for the group \(A_4\).
(d) Do part (b) for the group \(S_3 \times \mathbf{Z}_2\).
(e) Find all composition factors of \(S_4\) and \(D_6\).
(iv) (a) Let \(G\) be an (additive) abelian group. Define an operation of multiplication in \(G\) by \(ab = 0\) (for all \(a, b \in G\)). Then \(G\) is a ring.
(b) Let \(S\) be the set of all subsets of some fixed set \(U\). For \(A, B \in S\), define \(A + B = (A - B) \cup (B - A)\) and \(AB = A \cap B\). Then \(S\) is a ring. Is \(S\) commutative? Does it have an identity?
(v) Let \(R\) be the set of all \(2 \times 2\) matrices over the complex field \(\mathbf{C}\) of the form
where \(\overline{z}, \overline{w}\) are the complex conjugates of \(z\) and \(w\) respectively (that is, \(c = a+b\sqrt{-1} \Leftrightarrow \overline{c} = a-b\sqrt{-1}\)). Then \(R\) is a division ring that is isomorphic to the division ring \(K\) of real quaternions. Hint: Define an isomorphism $\varphi K \rightarrow R $ by letting the images of \(1, i, j, k \in K\) be respectively the matrices
(vi) (a) The subset \(G = \{1, -1, i, -i, j, -j, k, -k\}\) of the division ring \(K\) of real quaternions forms a group under multiplication.
(b) \(G\) is isomorphic to the quaternion group (Exercises I.4.14 and I.2.3).
(c) What is the difference between the ring \(K\) and the group ring \(R(G)\) (\(R\) the field of real numbers)?
证明
(iii) (a) \(D_4 = \langle a, b \mid a^4 = b^2 = e, ba = a^{-1}b \rangle\). 其中 \(\langle a \rangle = \{e, a, a^2, a^3\}\), \(\langle a^2 \rangle = \{e, a^2\}\). 所以 \(D_4 > \langle a \rangle > \langle a^2 \rangle > \langle e \rangle\) 是一个正规列 .
(b) 考虑 \(D_4\) 的极大正规子群,
Week 6¶
Question
\(\star\)(i) (a) The center of the ring \(S\) of all \(2 \times 2\) matrices over a field \(F\) consists of all matrices of the form \(\begin{pmatrix} a & 0 \\ 0 & a \end{pmatrix}\).
(b) The center of \(S\) is not an ideal in \(S\).
(c) What is the center of the ring of all \(n \times n\) matrices over a division ring?
(ii) (a) Show that \(\mathbf{Z}\) is a principal ideal ring [see Theorem I.3.1].
(b) Every homomorphic image of a principal ideal ring is also a principal ideal ring.
(c) \(Z_m\) is a principal ideal ring for every \(m > 0\).
(iii) Determine all prime and maximal ideals in the ring \(\mathbf{Z}_m\).
(iv) An element \(e\) in a ring \(R\) is said to be idempotent if \(e^2 = e\). An element of the center of the ring \(R\) is said to be central. If \(e\) is a central idempotent in a ring \(R\) with identity, then
(a) \(1_R-e\) is a central idempotent;
(b) \(eR\) and \((1_R-e)R\) are ideals in \(R\) such that \(R = eR \times (1_R-e)R\).
\(\star\)(v) Let \(R\) be the subring \(\{a+b\sqrt{10} \mid a, b \in \mathbf{Z}\}\) of the field of real numbers.
(a) The map \(N: R \rightarrow \mathbf{Z}\) given by \(a+b\sqrt{10} \mapsto (a+b\sqrt{10})(a-b\sqrt{10}) = a^2-10b^2\) is such that \(N(uv) = N(u)N(v)\) for all \(u,v \in R\) and \(N(u) = 0\) if and only if \(u = 0\).
(b) \(u\) is a unit in \(R\) if and only if \(N(u) = \pm 1\).
(c) \(2, 3, 4+\sqrt{10}\) and \(4-\sqrt{10}\) are irreducible elements of \(R\).
(d) \(2, 3, 4+\sqrt{10}\) and \(4-\sqrt{10}\) are not prime elements of \(R\). [Hint: \(3 \cdot 2 = 6 = (4+\sqrt{10})(4-\sqrt{10})\)].
(vi) Show that in the integral domain of Exercise 3 every element can be factored into a product of irreducibles, but this factorization need not be unique (in the sense of Definition 3.5 (ii)).
Week 7¶
Question
(i) (a) If \(a\) and \(n\) are integers, \(n>0\), then there exist integers \(q\) and \(r\) such that \(a = qn+r\), where \(\lvert r \rvert < n/2\).
(b) The Gaussian integers \(\mathbf{Z}[\sqrt{-1}]\) form a Euclidean domain with \(\varphi(a+b\sqrt{-1}) = a^2 + b^2\) [Hint: To show that Definition 3.8(ii) holds, first let \(y = a+b\sqrt{-1}\) and assume \(x\) is a positive integer. By part (a) there are integers such that \(a = q_1x+r_1\) and \(b = q_2x+r_2\), with \(\lvert r_1 \rvert < x/2\), \(\lvert r_2 \rvert < x/2\). Let \(q = q_1 + q_2\sqrt{-1}\) and \(r = r_1+r_2\sqrt{-1}\); then \(y = qx+r\), with \(r = 0\) or \(\varphi(r) < \varphi(x)\). In the general case, observe that for \(x = c+d\sqrt{-1} \neq 0\) and \(\overline{x} = c-d\sqrt{-1}\), \(x\overline{x} > 0\). There are \(q, r_0 \in \mathbf{Z}[\sqrt{-1}]\) such that \(y\overline{x} = q(x\overline{x})+r_0\), with \(r_0 = 0\) or \(\varphi(r_0) < \varphi(x\overline{x})\). Let \(r = y-qx\); then \(y = qx+r\) and \(r = 0\) or \(\varphi(r) < \varphi(x)\).]
(ii) (a) The set \(E\) of positive even integers is a multiplicative subset of \(\mathbf{Z}\) such that \(E^{-1}\mathbf{Z}\) is the field of rational numbers.
(b) State and prove condition(s) on a multiplicative subset \(S\) of \(\mathbf{Z}\) which insure that \(S^{-1}\mathbf{Z}\) is the field of rationals.
(iii) If \(S = \{2, 4\}\) and \(R = \mathbf{Z}_6\), then \(S^{-1}R\) is isomorphic to the field \(\mathbf{Z}_3\). Consequently, the converse of Theorem 4.3(ii) is false.
(iv) (a) If \(R\) is the ring of all \(2 \times 2\) matrices over \(\mathrm{Z}\), then for any \(A in R\),
(b) There exist \(C, A \in R\) such that \((C + A)(C - A) \neq C^2 - A^2\). Therefore, Corollary 5.6 is false if the rings involved are not commutative.
\(\star\)(v) (a) If \(F\) is a field then every nonzero element of \(F[[x]]\) is of the form \(x^ku\) with \(u \in F[[x]]\) a unit.
(b) \(F[[x]]\) is a principal ideal domain whose only ideals are \(0, F[[x]] = (1_F) = (x^0)\) and \((x^k)\) for each \(k > 1\).
Week 8¶
Question
(i) (a) If \(D\) is an integral domain and \(c\) is an irreducible element in \(D\), then \(D[x]\) is not a principal ideal domain. [Hint: consider the ideal \((x,c)\) generated by $x4 and \(c\).]
(b) \(\mathrm{Z}[x]\) is not a principal ideal domain.
(c) If \(F\) is a field and \(n > 2\), then \(F[x_1, \ldots , x_n]\) is not a principal ideal domain. [Hint: show that \(x_1\) is irreducible in \(F[x_1, \ldots , x_{n-1}]\).]
(ii) Let \(f\) be a polynomial of positive degree over an integral domain \(D\).
(a) If \(\mathrm{char} D = 0\), then \(f' \neq 0\).
(b) If \(\mathrm{char} D = p \neq 0\), then \(f' = 0\) if and only if \(f\) is a polynomial in \(x^p\) (that is, \(f = a_o + a_px^p + a_{2p}x^{2p} + \cdots + a_{ip}x^{ip}\)).
(iii) (a) Let \(D\) be an integral domain and \(c \in D\). Let \(f(x) = \sum_{i=0}^n a_ix^i \in D[x]\) and \(f(x - c) = \sum_{i=0}^n a_i(x-c)^i \in D[x]\). Then \(f(x)\) is irreducible in \(D[x]\) iff \(f(x - c)\) is irreducible.
(b) For each prime \(p\), the cyclotomic polynomial(分圆多项式) \(f = x^{p-1} + x^{p-2} + \cdots + x + 1\) is irreducible in \(\mathrm{Z}[x]\). [Hint: observe that \(f = \dfrac{x^p-1}{x-1}\), whence \(f(x + 1) = \dfrac{(x + 1)^p - 1}{x}\). Use the Binomial Theorem 1 .6 and Eisenstein's Criterion to show that \(f(x + 1)\) is irreducible in \(\mathrm{Z}[x]\).]
(iv) Let \(I\) be a left ideal of a ring \(R\) and \(A\) an \(R\)-module.
(a) If \(S\) is a nonempty subset of \(A\), then \(IS = \{\sum_{i=1}^nr_ia_i \mid n \in N^*, r_i \in I, a_i \in S\}\) is a submodule of \(A\). Note that if \(S = \{a\}\) , then \(IS = Ia = \{ra \mid r \in I\}\).
(b) If \(I\) is a two-sided ideal, then \(A/IA\) is an \(R/I\)-module with the action of \(R/I\) given by \((r + I)(a + IA) = ra + IA\).
\(\star\)(v) (a) If \(A\) and \(B\) are \(R\)-modules, then the set \(\mathrm{Hom}_R(A,B)\) of all \(R\)-module homomorphisms \(A \rightarrow B\) is an abelian group with \(f + g\) given on \(a \in A\) by \((f + g)(a) = f(a) + g(a) \in B\). The identity element is the zero map.
(b) \(\mathrm{Hom}_R(A,A)\) is a ring with identity, where multiplication is composition of functions. \(\mathrm{Hom}_R(A,A)\) is called the endomorphism ring(自同态环) of \(A\).
(c) \(A\) is a left \(\mathrm{Hom}_R(A,A)\)-module with \(fa\) defined to be \(f(a)\) (\(a \in A, f \in \mathrm{Hom}_R(A,A)\)).
Week 9¶
Question
(Consider all the rings as communative rings with identity, and all division rings as fields.)
(i) (The five lemma) Let
(a) \(\varphi_1\) an epimorphism and \(\varphi_2, \varphi_4\) monomorphisms \(\Rightarrow\) \(\varphi_3\) a monomorphism;
(b) \(\varphi_5\) a monomorphism and \(\varphi_2, \varphi_4\) epimorphisms \(\Rightarrow\) \(\varphi_3\) an epimorphism.
(ii) (a) If \(0 \rightarrow A \rightarrow B \xrightarrow{f} C \rightarrow 0\) and \(0 \rightarrow C \xrightarrow{g} D \rightarrow E \rightarrow 0\) are exact sequences of \(R\)-modules, then the sequence \(0 \rightarrow A \rightarrow B \xrightarrow{gf} D \rightarrow E \rightarrow 0\) is exact.
(b) Show that every exact sequence may be obtained by splicing together suitable short exact sequences as in (a).
\(\star\)(iii) Let \(R\) be a principal ideal domain, \(A\) a unitary left R-module, and \(p \in R\) a prime (= irreducible). Let \(pA = \{pa \mid a \in A\}\) and \(A[p] = \{a \in A \mid pa = 0\}\).
(a) \(R/(p)\) is a field(Theorems III.2.20 and III.3.4);
(b) \(pA\) and \(A[p]\) are submodules of \(A\);
(c) \(A/pA\) is a vector space over \(R/(p)\), with scalar multiplication given by \((r + (p))(a + pA) = ra + pA\);
(d) \(A[p]\) is a vector space over \(R/(p)\), with scalar multiplication given by \((r + (p))(a) = ra\).
(iv) The following conditions on a ring \(R\) [with identity] are equivalent:
(a) Every [unitary] R-module is projective.
(b) Every short exact sequence of [unitary] R-modules is split exact.
(c) Every [unitary] R-module is injective.
answer
(i) 采取图追踪策略 .
(a) 设 \(a_3 \in A_3\),使得 \(\varphi_3(a_3) = 0\). 我们要证明 \(a_3 = 0\).
设 \(a_4 = f_3(a_3) \in A_4\),则 \(\varphi_4(a_4) = g_3(\varphi_3(a_3)) = g_3(0) = 0\). 因为 \(\varphi_4\) 是单同态,所以 \(a_4 = 0\).
考虑上行为正合列,\(f_3(a_3) = a_4 = 0\),所以 \(a_3 \in \mathrm{Ker} f_3 = \mathrm{Im} f_2\),即存在 \(a_2 \in A_2\),使得 \(f(a_2) = a_3\). 设 \(b_2 = \varphi_2(a_2) \in B_2\),所以 \(g_2(b_2) = \varphi_3(a_3) = 0\).
考虑下行为正合列,故 \(b_2 \in \mathrm{Ker} g_2 = \mathrm{Im} g_1\),即存在 \(b_1 \in B_1\),使得 \(g_1(b_1) = b_2\).
考虑 \(\varphi_1\) 是满同态,所以存在 \(a_1 \in A_1\),使得 \(b_1 = \varphi_1(a_1)\). 进而 \(b_2 = g_1(\varphi_1(a_1)) = \varphi_2(f_1(a_1)) = \varphi_2(a_2)\). 因为 \(\varphi_2\) 是单同态,所以 \(f_1(a_1) = a_2\).
考虑上行为正合列,所以 \(a_2 \in \mathrm{Im} f_1 = \mathrm{Ker} f_2\),所以 \(f_2(a_2) = a_3 = 0\). 故 \(\varphi_3\) 是单同态.
(b) 设 \(b_3 \in B_3\),我们要构造出 \(a \in A_3\),使得 \(\varphi_3(a) = b_3\).
设 \(b_4 = g_3(b_3) \in B_4\),因为 \(\varphi_4\) 是满同态,故存在 \(a_4 \in A_4\),使得 \(\varphi_4(a_4) = b_4\). 设 \(a_5 = f_4(a_4) \in A_5\),从而有 \(\varphi_5(a_5) = g_4(b_4)\).
考虑下行为正合列,所以 \(b_4 \in \mathrm{Im} g_3 = \mathrm{Ker} g_4\). 进而 \(\varphi_5(a_5) = 0\). 考虑 \(\varphi_5\) 是单同态,所以 \(a_5 = 0\).
考虑上行为正合列,所以 \(a_4 \in \mathrm{Ker} f_4 = \mathrm{Im} f_3\),即存在 \(a_3 \in A_3\),使得 \(f_3(a_3) = a_4\). 而 \(\varphi_4(f_3(a_3)) = \varphi_4(a_4) = b_4 = g_3(b_3)\),且 \(\varphi_4(f_3(a_3)) = g_3(\varphi_3(a_3))\),所以 \(g_3(b_3 - (\varphi_3(a_3))) = 0\).
考虑下行为正合列,所以 \(b_3 - \varphi_3(a_3) \in \mathrm{Ker} g_3 = \mathrm{Im} g_2\),也就是存在 \(b_2 \in B_2\),使得 \(g_2(b_2) = b_3 - \varphi_3(a_3)\).
考虑 \(\varphi_2\) 是满同态,所以存在 \(a_2 \in A_2\),使得 \(b_2 = \varphi_2(a_2)\). 而 \(f_2(a_2) + a_3 \in A_3\),且 \(\varphi_3(f_2(a_2) + a_3) = \varphi_3(f_2(a_2)) + \varphi_3(a_3) = g_2(\varphi_2(a_2)) + \varphi_3(a_3) = b_3 - \varphi_3(a_3) + \varphi_3(a_3) = b_3\). 所以 \(\varphi_3\) 是满同态.
Week 10¶
Question
(i) Without using Lemma 3.9 prove that:
(a) Every homomorphic image of a divisible abelian group is divisible.
(b) Every direct summand (Exercise I.8.12) of a divisible abelian group is divisible.
(c) A direct sum of divisible abelian groups is divisible.
(Lemma 3.9: An abelian group \(D\) is divisible if and only if \(D\) is an injective (unitary) \(\mathbf{Z}\)-module.
Exercise I.8.12: A normal subgroup \(H\) of a group \(G\) is said to be a direct factor (direct summand if \(G\) is additive abelian) if there exists a (normal) subgroup \(K\) of \(G\) such that \(G = H \times K\).)
(ii)
(a) For any abelian group \(A\) and positive integer \(m\), \(\operatorname{Hom}(\mathbf{Z}_m, A) \cong A[m] = \{a \in A \mid ma = 0\}\).
(b) \(\operatorname{Hom}(\mathbf{Z}_m, \mathbf{Z}_n) \cong \mathbf{Z}_{(m, n)}\).
(c) The \(\mathbf{Z}\)-module \(\mathbf{Z}_m\) has \(\mathbf{Z}_m^* = 0\).
(d) For each \(k > 1\), \(\mathbf{Z}_m\) is a \(\mathbf{Z}_{mk}\)-module (Exercise 1.1); as a \(\mathbf{Z}_{mk}\)-module, \(\mathbf{Z}_m^* \cong \mathbf{Z}_m\).
(iii) Let \(A\) and \(B\) be abelian groups.
(a) For each \(m > 0\), \(A \otimes \mathbf{Z}_m \cong A/mA\).
(b) \(\mathbf{Z}_m \otimes \mathbf{Z}_n \cong \mathbf{Z}_{(m, n)}\).
(c) Describe \(A \otimes B\), when \(A\) and \(B\) are finitely generated.
(iv) If \(A\) is a torsion abelian group and \(\mathbf{Q}\) the (additive) group of rationals, then
(a) \(A \otimes \mathbf{Q} = 0\).
(b) \(\mathbf{Q} \otimes \mathbf{Q} \cong \mathbf{Q}\).
Week 11¶
Question
(i) If \(R\) is a nonzero commutative ring with identity and every submodule of every free \(R\)-module is free, then \(R\) is a principal ideal domain. [Hint: Every ideal \(I\) of \(R\) is a free \(R\)-module. If \(u, v \in I\) (\(u \neq 0, v \neq 0\)), then \(uv + (-v)u = 0\), which implies that \(I\) has a basis of one element; that is, \(I\) is principal.]
(ii) Every free module over an arbitrary integral domain with identity is torsion-free. The converse is false (Exercise II.1.10).
\(\star\)(iii) Let \(A\) be a cyclic \(R\)-module of order \(r \in R\).
(a) If \(s \in R\) is relatively prime to \(r\), then \(sA = A\) and \(A[s] = 0\).
(b) If \(s\) divides \(r\), say \(sk = r\), then \(sA \cong R/(k)\) and \(A[s] \cong R/(s)\).
(iv) If \(A\) is a finitely generated torsion module, then \(\{r \in R \mid rA = 0\}\) is a nonzero ideal in \(R\), say \((r_1)\). \(r_1\) is called the minimal annihilator of \(A\). Let \(A\) be a finite abelian group with minimal annihilator \(m \in \mathbf{Z}\). Show that a cyclic subgroup of \(A\) of order properly dividing \(m\) need not be a direct summand of \(A\).
Week 12¶
Question
\(\star\)(i) (a) \([F : K] = 1\) if and only if \(F = K\).
(b) If \([F : K]\) is prime, then there are no intermediate fields between \(F\) and \(K\).
(c) lf \(u \in F\) has degree \(n\) over \(K\), then \(n\) divides \([F : K]\).
\(\star\)(ii) (a) Consider the extension \(\mathbf{Q}(u) of \mathbf{Q}\) generated by a real root \(u\) of \(x^3 - 6x^2+ 9x + 3\). (Why is this irreducible?) Express each of the following elements in terms of the basis \(\{1, u, u^2\}\): \(u^4\); \(u^5\); \(3u^5 - u^4 + 2\); \((u + 1)^{-1}\); \((u^2 - 6u + 8)^{-1}\).
(b) Do the same with respect to the basis \(\{1, u, u^2, u^3, u^4\}\) of \(\mathbf{Q}(u)\) where \(u\) is a real root of \(x^5 + 2x + 2\) and the elements in question are: \((u^2 + 2)(u^3 + 3u)\); \(u^{-1}\); \(u^4(u^4 + 3u^2 + 7u + 5)\); \((u + 2)(u^2 + 3)^{-1}\).
(iii) A complex number is said to be an algebraic number if it is algebraic over \(\mathbf{Q}\) and an algebraic integer if it is the root of a monic polynomial in \(\mathbf{Z}[x]\).
(a) If \(u\) is an algebraic number, there exists an integer \(n\) such that \(nu\) is an algebraic integer.
(b) If \(r \in \mathbf{Q}\) is an algebraic integer, then \(r \in \mathbf{Z}\).
(c) If \(u\) is an algebraic integer and \(n \in \mathbf{Z}\), then \(u + n\) and \(nu\) are algebraic integers.
(d) The sum and product of two algebraic integers are algebraic integers.
(iv) Let \(c,d\) be constructible real numbers.
(a) \(c + d\) and \(c - d\) are constructible.
(b) If \(d \neq 0\), then \(c/d\) is constructible. [Hint: If \((x, 0)\) is the intersection of the \(x\) axis and the straight line through \((0, 1)\) that is parallel the line through \((0, d)\) and \((c, 0)\), then \(x = c/d\).]
(c) \(cd\) is constructible [Hint: use (b)].
(d) The constructible real numbers form a subfield containing \(\mathbf{Q}\).
(e) If \(c > 0\), then \(\sqrt{c}\) is constructible. [Hint: If \(y\) is the length of the straight line segment perpendicular to the \(x\) axis that joins \((1,0)\) with the (upper half of the) circle with center \((\frac{c + 1}{2}, 0)\) and radius \(\frac{c + 1}{2}\) then \(y = \sqrt{c}\).]
answer
(iii) (d) 首先证明下面这个命题:
以下两个条件等价:
(1) \(z \in \mathbf{C}\) 是代数整数;
(2) 存在有限生成 \(\mathbf{Z}\)-模 \(M\),使得 \(zM \subset M\).
(1) \(\Rightarrow\) (2) 设 \(f(x) = \sum_{i = 1}^n a_ix^i, a_n = 1, f \in \mathbf{Z}[x]\). 则 \(f(z) = 0\). 因为 \(z^n = -a_{n - 1}z^{n - 1} - \cdots - a_1z - a_0\),即 \(z^n\) 可以由 \(\{1, z, \ldots, z^{n - 1}\}\) 整系数线性表示,所以令 \(M = \langle 1, z, \ldots, z^{n - 1} \rangle\). 显然 \(zM \subset M\).
(2) \(\Rightarrow\) (1) 设 \(M = \langle v_1, \ldots, v_n \rangle\). 因为 \(zM \subset M\),所以 \(zv_i, i = 1, \ldots, n\) 均可由 \(v_1, \ldots, v_n\) 整系数线性表出,设
方程组系数矩阵为 \(A\). 从而此方程组等价于 \((A - zI)X = 0\) 有非平凡解,所以 \(\operatorname{det} (A - zI) = 0\). 这是一个首一整系数方程,\(z\) 是它的根,所以 \(z\) 是一个代数整数 .
所以设 \(x, y\) 都是代数整数,则存在有限生成 \(\mathbf{Z}\)- 模 \(M, N\) 使得 \(xM \subset M\),\(yN \subset N\). 容易验证 \(MN\) 依然是有限生成 \(\mathbf{Z}\)- 模 ( 所有元素用基表出然后拆开即可 ),而且 \((x + y)MN \subset MN\),\((xy)MN \subset MN\),所以 \(x + y\),\(xy\) 均是代数整数 .