1.基本定义
—— 态矢与算符,是量子力学及量子信息中最重要的概念之二(即使一个系统没有哈密顿量也存在态矢与算符)
——态矢(state vector),代表量子态的状态,可标记为
∣
ψ
⟩
|\psi\rangle
∣ψ⟩(狄拉克符号表示);算子或算符(operator),定义为对态矢或算子的操作,可记为
O
^
\hat{O}
O^
——态矢和算子所在的空间,被称为希尔伯特空间(态矢/量子的希尔伯特空间)
——定义态矢和算子的是那些与基矢选择无关的性质:
a.内积:
⟨
ψ
∣
φ
⟩
\langle\psi|\varphi\rangle
⟨ψ∣φ⟩(保真度)
⟨
φ
∣
O
^
∣
φ
⟩
\langle\varphi|\hat{O}|\varphi\rangle
⟨φ∣O^∣φ⟩(均值或观测值)等
b.迹(trace)trace(
O
^
\hat{O}
O^);
c.对易子
[
O
^
,
P
^
]
=
O
^
P
^
−
P
^
O
^
[\hat{O},\hat{P}] = \hat{O}\hat{P}-\hat{P}\hat{O}
[O^,P^]=O^P^−P^O^
d.本征关系
O
^
∣
φ
⟩
=
O
∣
φ
⟩
\hat{O}|\varphi\rangle = O|\varphi\rangle
O^∣φ⟩=O∣φ⟩
注:右矢(|
⟩
\rangle
⟩)定义在希尔伯特空间,左矢(
⟨
∣
\langle|
⟨∣)定义在对偶希尔伯特空间
例:泡利算符
σ
^
x
\hat{\sigma}^x
σ^x,
σ
^
y
\hat{\sigma}^y
σ^y,
σ
^
z
\hat{\sigma}^z
σ^z
满足如下性质:
[
σ
^
a
,
σ
^
b
]
=
2
i
ε
a
b
c
σ
^
c
[\hat{\sigma}^a,\hat{\sigma}^b] = 2i\varepsilon_{abc}\hat{\sigma}^c
[σ^a,σ^b]=2iεabcσ^c,
(
σ
^
a
)
2
=
−
i
σ
^
x
σ
^
y
σ
^
z
=
I
(\hat{\sigma}^a)^2 = -i\hat{\sigma}^x\hat{\sigma}^y\hat{\sigma}^z = I
(σ^a)2=−iσ^xσ^yσ^z=I(力学基础)
自旋算符给出了SU(2)群的生成元(群论里的知识)
注:代数性质是定义算符的核心
态矢和算子在给定基矢下的展开系数,可由向量或矩阵表示
基矢:定义为一组态矢{
∣
i
⟩
|i\rangle
∣i⟩} ,满足正交完备性:
⟨
i
∣
i
′
⟩
=
δ
i
i
′
,
∑
i
∣
i
⟩
⟨
i
∣
=
I
\langle i|i^{'}\rangle = \delta_{ii{'}} , \sum_i|i\rangle\langle i| = I
⟨i∣i′⟩=δii′,i∑∣i⟩⟨i∣=I
注:(a)迪拉克矩阵
δ
i
i
′
=
1
当
i
=
i
′
\delta_{ii{'}} = 1当i= i^{'}
δii′=1当i=i′,否则
δ
i
i
′
=
0
\delta_{ii{'}} = 0
δii′=0
(b)在特殊情况下,基矢可以不正交,也可以不完备或过完备;
(c)
∣
i
⟩
⟨
i
∣
=
∣
i
⟩
⨂
⟨
i
∣
|i\rangle\langle i| = |i\rangle\bigotimes\langle i|
∣i⟩⟨i∣=∣i⟩⨂⟨i∣,
⨂
\bigotimes
⨂称为直积、张量积、外积或克伦内克积。
2.基矢与系数
例:
σ
^
z
\hat{\sigma}^z
σ^z的两个本征态,记为
∣
↑
⟩
|\uparrow\rangle
∣↑⟩与
∣
↓
⟩
|\downarrow\rangle
∣↓⟩(在量子信息与量子计算领域,常被记作为
∣
0
⟩
|0\rangle
∣0⟩与
∣
1
⟩
|1\rangle
∣1⟩,与经典比特的两个状态相对应),其本征值分别为1与-1.
——
∣
↑
⟩
|\uparrow\rangle
∣↑⟩与
∣
↓
⟩
|\downarrow\rangle
∣↓⟩构成一组正交完备基矢,满足
⟨
i
∣
i
′
⟩
=
δ
i
i
′
,
∣
↑
⟩
⟨
↑
∣
+
∣
↓
⟩
⟨
↓
∣
=
I
\langle i|i^{'}\rangle = \delta_{ii{'}},|\uparrow\rangle\langle\uparrow|+|\downarrow\rangle\langle\downarrow| =I
⟨i∣i′⟩=δii′,∣↑⟩⟨↑∣+∣↓⟩⟨↓∣=I
——设定该组基矢的矢量表示为:
∣
1
⟩
=
∑
s
=
0
1
ϕ
s
∣
s
⟩
,
|1\rangle=\sum_{s=0}^{1} \phi_{s}|s\rangle, \quad
∣1⟩=∑s=01ϕs∣s⟩, 系数
ϕ
=
[
0
1
]
T
\phi=\left[\begin{array}{cc}0 & 1\end{array}\right]^{T}
ϕ=[01]T
∣
1
⟩
=
ϕ
0
∣
0
⟩
+
ϕ
1
∣
1
⟩
|1\rangle= \phi_{0}|0\rangle+\phi_{1}|1\rangle
∣1⟩=ϕ0∣0⟩+ϕ1∣1⟩,即只有当
ϕ
0
=
0
,
ϕ
1
=
1
\phi_{0} = 0,\phi_{1} = 1
ϕ0=0,ϕ1=1时等式成立
∣
0
⟩
=
∑
s
=
0
1
ϕ
s
′
∣
s
⟩
,
|0\rangle=\sum_{s=0}^{1} \phi_{s}^{\prime}|s\rangle, \quad
∣0⟩=∑s=01ϕs′∣s⟩, 系数
ϕ
′
=
[
1
0
]
T
\phi^{\prime}=\left[\begin{array}{cc}1 & 0\end{array}\right]^{T}
ϕ′=[10]T
注:此后都标记
∣
↑
⟩
=
∣
1
⟩
,
∣
↓
⟩
=
∣
0
⟩
|\uparrow\rangle=|1\rangle,|\downarrow\rangle=|0\rangle
∣↑⟩=∣1⟩,∣↓⟩=∣0⟩
——任意单个自旋的量子态可写成基矢的线性叠加:
∣
φ
⟩
=
φ
0
∣
0
⟩
+
φ
1
∣
1
⟩
|\varphi\rangle = \varphi_{0}|0\rangle +\varphi_{1}|1\rangle
∣φ⟩=φ0∣0⟩+φ1∣1⟩可见,系数𝜑为一个二维向量。
——态矢和算子的定义是独立于基矢的,因此,态并不等价于某个向量,算子也并不等价于某个矩阵;但在不引起误解的情况下,可直接可认为态即为对应的系数向量,算子即为对应的系数矩阵,如
∣
1
⟩
=
[
0
1
]
T
|1\rangle=\left[\begin{array}{cc}0 & 1 \end{array}\right]^{T}
∣1⟩=[01]T(真正的含义是:1态在给定基底下的系数为0,1)
——规定:左矢(bra)
⟨
φ
∣
\langle\varphi|
⟨φ∣ 对应于行向量,右矢(ket)
∣
φ
⟩
|\varphi\rangle
∣φ⟩对应于列向量;
⟨
φ
∣
\langle\varphi|
⟨φ∣为
∣
φ
⟩
|\varphi\rangle
∣φ⟩的转置共轭。(复数中需要取共轭)
例:量子态内积对应于系数向量的内积:
⟨
φ
∣
ψ
⟩
=
[
φ
0
∗
φ
1
∗
]
[
ψ
0
ψ
1
]
T
\quad\langle\varphi \mid \psi\rangle=\left[\begin{array}{ccc}\varphi_{0}^{*} & \left.\varphi_{1}^{*}\right]\left[\psi_{0}\right. & \psi_{1}\end{array}\right]^{T}
⟨φ∣ψ⟩=[φ0∗φ1∗][ψ0ψ1]T
证明: 由
∣
ψ
⟩
=
∑
i
ψ
i
∣
i
⟩
,
∣
φ
⟩
=
∑
j
φ
j
∣
j
⟩
|\psi\rangle=\sum_{i} \psi_{i}|i\rangle, \quad|\varphi\rangle=\sum_{j} \varphi_{j}|j\rangle
∣ψ⟩=∑iψi∣i⟩,∣φ⟩=∑jφj∣j⟩
有
⟨
φ
∣
ψ
⟩
=
∑
i
j
φ
j
∗
ψ
i
⟨
j
∣
i
⟩
\langle\varphi \mid \psi\rangle=\sum_{i j} \varphi_{j}^{*} \psi_{i}\langle j \mid i\rangle
⟨φ∣ψ⟩=∑ijφj∗ψi⟨j∣i⟩
由基矢的正交归一性
⟨
j
∣
i
⟩
=
δ
j
i
,
\langle j \mid i\rangle=\delta_{j i}, \quad
⟨j∣i⟩=δji, 有
⟨
φ
∣
ψ
⟩
=
∑
i
j
φ
j
∗
ψ
i
δ
j
i
=
∑
i
φ
i
∗
ψ
i
\langle\varphi \mid \psi\rangle=\sum_{i j} \varphi_{j}^{*} \psi_{i} \delta_{j i}=\sum_{i} \varphi_{i}^{*} \psi_{i}
⟨φ∣ψ⟩=∑ijφj∗ψiδji=∑iφi∗ψi
证毕
基矢的矢量表示确定之后,可用这组基矢对算符做展开,得到算符的系数。例如,泡利算符的展开系数为2 × 2的矩阵,满足:
σ
^
α
=
∑
i
j
=
0
1
σ
i
j
a
∣
i
⟩
⟨
j
∣
\hat{\sigma}^\alpha = \sum_{ij = 0}^1\sigma_{ij}^a|i\rangle\langle j|
σ^α=ij=0∑1σija∣i⟩⟨j∣(再次强调:
σ
^
α
\hat{\sigma}^\alpha
σ^α是希尔伯特空间中的算子,
σ
i
j
a
\sigma_{ij}^a
σija是一个二阶张量(矩阵),一定要注意二者的区别和联系;在不引起误解的情况下,算符与其系数可混用)
由
σ
^
α
=
∑
i
j
=
0
1
σ
i
j
a
∣
i
⟩
⟨
j
∣
\hat{\sigma}^\alpha = \sum_{ij = 0}^1\sigma_{ij}^a|i\rangle\langle j|
σ^α=∑ij=01σija∣i⟩⟨j∣以及基矢的正交归一性,易得算符与其系数之间满足:
σ
i
j
a
=
⟨
i
∣
σ
^
α
∣
j
⟩
\sigma_{ij}^a=\langle i|\hat{\sigma}^\alpha|j\rangle
σija=⟨i∣σ^α∣j⟩证明如下:
σ
^
α
=
σ
00
a
∣
0
⟩
⟨
0
∣
+
σ
10
a
∣
1
⟩
⟨
0
∣
+
σ
01
a
∣
0
⟩
⟨
1
∣
+
σ
11
a
∣
1
⟩
⟨
1
∣
\hat{\sigma}^\alpha = \sigma_{00}^a|0\rangle\langle 0| + \sigma_{10}^a|1\rangle\langle 0|+ \sigma_{01}^a|0\rangle\langle 1|+ \sigma_{11}^a|1\rangle\langle 1|
σ^α=σ00a∣0⟩⟨0∣+σ10a∣1⟩⟨0∣+σ01a∣0⟩⟨1∣+σ11a∣1⟩⟨1∣
假如两边同时乘以
∣
0
⟩
⟨
0
∣
|0\rangle\langle 0|
∣0⟩⟨0∣,即:
⟨
0
∣
σ
^
α
∣
0
⟩
=
σ
00
a
⟨
0
∣
0
⟩
⟨
0
∣
0
⟩
+
σ
10
a
⟨
0
∣
1
⟩
⟨
0
∣
0
⟩
+
σ
01
a
⟨
0
∣
0
⟩
⟨
1
∣
0
⟩
+
σ
11
a
⟨
0
∣
1
⟩
⟨
1
∣
0
⟩
\langle 0|\hat{\sigma}^\alpha|0\rangle = \sigma_{00}^a\langle 0|0\rangle\langle 0|0\rangle + \sigma_{10}^a\langle 0|1\rangle\langle 0|0\rangle+ \sigma_{01}^a\langle 0|0\rangle\langle 1|0\rangle+ \sigma_{11}^a\langle 0|1\rangle\langle 1|0\rangle
⟨0∣σ^α∣0⟩=σ00a⟨0∣0⟩⟨0∣0⟩+σ10a⟨0∣1⟩⟨0∣0⟩+σ01a⟨0∣0⟩⟨1∣0⟩+σ11a⟨0∣1⟩⟨1∣0⟩
因为
⟨
1
∣
0
⟩
\langle 1|0\rangle
⟨1∣0⟩和
⟨
0
∣
1
⟩
\langle 0|1\rangle
⟨0∣1⟩都等于0,
⟨
0
∣
0
⟩
\langle 0|0\rangle
⟨0∣0⟩等于1
所以原式等于:
⟨
0
∣
σ
^
α
∣
0
⟩
=
σ
00
a
\langle 0|\hat{\sigma}^\alpha|0\rangle = \sigma_{00}^a
⟨0∣σ^α∣0⟩=σ00a,同理其它条件情况下也成立。
通过
σ
^
z
\hat{\sigma}^z
σ^z的本征方程及三个泡利算符的对易关系,可以求得其在该组基矢下的系数矩阵(在不引起误解的情况下,提及某张量时,可省略其下标,如
σ
i
j
a
\sigma_{ij}^a
σija,写为
σ
x
\sigma^x
σx等)
σ
x
=
[
0
1
1
0
]
,
σ
y
=
[
0
i
−
i
0
]
,
σ
z
=
[
1
0
0
−
1
]
\sigma^{x}=\left[\begin{array}{ll} 0 & 1 \\ 1 & 0 \end{array}\right], \quad \sigma^{y}=\left[\begin{array}{cc} 0 & i \\ -i & 0 \end{array}\right], \quad \sigma^{z}=\left[\begin{array}{cc} 1 & 0 \\ 0 & -1 \end{array}\right]
σx=[0110],σy=[0−ii0],σz=[100−1]
(注:自旋算符与泡利算符相差因子2, 即
s
^
α
=
σ
^
α
/
2
)
\left.\hat{s}^{\alpha}=\hat{\sigma}^{\alpha} / 2\right)
s^α=σ^α/2)
2.1 代码实例(生成自旋算符)
import numpy as np
#定义函数
def spin_operator_one_half():
op = dict()
op['i'] = np.eye(2)#identity
op['x'] = np.zeros((2,2))
op['x'][0, 1] = 1 / 2
op['x'][1, 0] = 1 / 2
op['y'] = np.zeros((2, 2), dtype=np.complex)
op['y'][0, 1] = 1j / 2
op['y'][1, 0] = -1j / 2
op['z'] = np.zeros((2, 2))
op['z'][0, 0] = 1 / 2
op['z'][1, 1] = -1 / 2
return op
#输出
spin = spin_operator_one_half()
print('Spin-x operator = ')
print(spin['x'])
print('\nSpin-y operator = ')
print(spin['y'])
print('\nSpin-z operator = ')
print(spin['z'])
3.系数运算
给定基矢,确定量子态与算子的向量与矩阵表示之后,相关的计算变为向量与矩阵的运算
例:定义上升算符
σ
^
+
\hat{\sigma}^{+}
σ^+ 和下降算符
σ
^
−
,
\hat{\sigma}^{-},
σ^−, 其在
σ
^
Z
\hat{\sigma}^{Z}
σ^Z 的本征基矢下的矩阵表示为
σ
+
=
[
0
1
0
0
]
,
σ
−
=
[
0
0
1
0
]
\sigma^{+}=\left[\begin{array}{ll} 0 & 1 \\ 0 & 0 \end{array}\right], \quad \sigma^{-}=\left[\begin{array}{ll} 0 & 0 \\ 1 & 0 \end{array}\right]
σ+=[0010],σ−=[0100]
有
σ
^
+
∣
0
⟩
=
∣
1
⟩
,
σ
^
−
∣
1
⟩
=
∣
0
⟩
,
σ
^
+
∣
1
⟩
=
σ
^
−
∣
0
⟩
=
0
,
\hat{\sigma}^{+}|0\rangle=|1\rangle, \quad \hat{\sigma}^{-}|1\rangle=|0\rangle, \quad \hat{\sigma}^{+}|1\rangle=\hat{\sigma}^{-}|0\rangle=0,
σ^+∣0⟩=∣1⟩,σ^−∣1⟩=∣0⟩,σ^+∣1⟩=σ^−∣0⟩=0, 分别对应于
[
0
1
0
0
]
[
0
1
]
=
[
1
0
]
,
[
0
0
1
0
]
[
1
0
]
=
[
0
1
]
,
[
0
1
0
0
]
[
1
0
]
=
[
0
0
1
0
]
[
0
1
]
=
[
0
0
]
\left[\begin{array}{ll} 0 & 1 \\ 0 & 0 \end{array}\right]\left[\begin{array}{l} 0 \\ 1 \end{array}\right]=\left[\begin{array}{l} 1 \\ 0 \end{array}\right], \quad\left[\begin{array}{ll} 0 & 0 \\ 1 & 0 \end{array}\right]\left[\begin{array}{l} 1 \\ 0 \end{array}\right]=\left[\begin{array}{l} 0 \\ 1 \end{array}\right], \quad\left[\begin{array}{ll} 0 & 1 \\ 0 & 0 \end{array}\right]\left[\begin{array}{l} 1 \\ 0 \end{array}\right]=\left[\begin{array}{ll} 0 & 0 \\ 1 & 0 \end{array}\right]\left[\begin{array}{l} 0 \\ 1 \end{array}\right]=\left[\begin{array}{l} 0 \\ 0 \end{array}\right]
[0010][01]=[10],[0100][10]=[01],[0010][10]=[0100][01]=[00]
算符的连乘对应于矩阵乘, 满足结合律。例如
σ
^
+
σ
^
−
∣
1
⟩
=
−
∣
0
⟩
,
\hat{\sigma}^{+} \hat{\sigma}^{-}|1\rangle=-|0\rangle,
σ^+σ^−∣1⟩=−∣0⟩, 对应
[
1
0
0
−
1
]
[
0
0
1
0
]
[
1
0
]
=
−
[
0
1
]
\left[\begin{array}{cc} 1 & 0 \\ 0 & -1 \end{array}\right]\left[\begin{array}{ll} 0 & 0 \\ 1 & 0 \end{array}\right]\left[\begin{array}{l} 1 \\ 0 \end{array}\right]=-\left[\begin{array}{l} 0 \\ 1 \end{array}\right]
[100−1][0100][10]=−[01]
注:矩阵积可写为求和的形式, 例如
σ
+
φ
⇔
Σ
j
σ
i
j
+
φ
j
,
\boldsymbol{\sigma}^{+} \boldsymbol{\varphi} \Leftrightarrow \Sigma_{j} \boldsymbol{\sigma}_{i j}^{+} \boldsymbol{\varphi}_{\boldsymbol{j}},
σ+φ⇔Σjσij+φj, 即进行相应的指标收缩(将一个算符作用在一个算符上,或者将一个算符作用在一个向量上)
注 :指标收缩是张量网络的核心