59.1

Problem 59.1

srednickiChapter 59

习题 59.1

来源: 第59章, PDF第354页


59.1 Compute T2\langle |\mathcal{T}|^2 \rangle for Compton scattering, eγeγe^- \gamma \rightarrow e^- \gamma. You should find that your result is the same as that for e+eγγe^+ e^- \rightarrow \gamma \gamma, but with sts \leftrightarrow t, and an extra overall minus sign. This is an example of crossing symmetry; there is an overall minus sign for each fermion that is moved from the initial to the final state.

习题 59.1 - 解答


对于康普顿散射 e(p)+γ(k)e(p)+γ(k)e^-(p) + \gamma(k) \rightarrow e^-(p') + \gamma(k'),我们定义 Mandelstam 变量为: s=(p+k)2,t=(pp)2,u=(pk)2s = (p+k)^2, \quad t = (p-p')^2, \quad u = (p-k')^2 它们满足 s+t+u=2m2s+t+u = 2m^2

该过程在树图阶包含 ss-道和 uu-道两个 Feynman 图。根据 Feynman 规则,散射振幅 T\mathcal{T} 为: T=e2ϵμ(k)ϵν(k)uˉ(p)[γμ(++m)γνsm2+γν(̸k+m)γμum2]u(p)\mathcal{T} = -e^2 \epsilon^*_\mu(k') \epsilon_\nu(k) \bar{u}(p') \left[ \frac{\gamma^\mu (\not{p}+\not{k}+m) \gamma^\nu}{s-m^2} + \frac{\gamma^\nu (\not{p}-\not{k}'+m) \gamma^\mu}{u-m^2} \right] u(p) 我们需要计算自旋和极化平均的振幅平方 T2=14spins, pol.T2\langle |\mathcal{T}|^2 \rangle = \frac{1}{4} \sum_{\text{spins, pol.}} |\mathcal{T}|^2。利用 Ward 恒等式,可以将光子的极化求和替换为 ϵμϵα=gμα\sum \epsilon^*_\mu \epsilon_\alpha = -g_{\mu\alpha}。因此: T2=e44(Tss+Tuu+Tsu+Tus)\langle |\mathcal{T}|^2 \rangle = \frac{e^4}{4} (T_{ss} + T_{uu} + T_{su} + T_{us}) 其中各项为狄拉克矩阵的迹: Tss=1(sm2)2Tr[(̸p+m)γμ(++m)γν(+m)γν(++m)γμ]T_{ss} = \frac{1}{(s-m^2)^2} \text{Tr} \left[ (\not{p}'+m) \gamma^\mu (\not{p}+\not{k}+m) \gamma^\nu (\not{p}+m) \gamma_\nu (\not{p}+\not{k}+m) \gamma_\mu \right] Tuu=1(um2)2Tr[(̸p+m)γν(̸k+m)γμ(+m)γμ(̸k+m)γν]T_{uu} = \frac{1}{(u-m^2)^2} \text{Tr} \left[ (\not{p}'+m) \gamma^\nu (\not{p}-\not{k}'+m) \gamma^\mu (\not{p}+m) \gamma_\mu (\not{p}-\not{k}'+m) \gamma_\nu \right] Tsu=Tus=1(sm2)(um2)Tr[(̸p+m)γμ(++m)γν(+m)γμ(̸k+m)γν]T_{su} = T_{us} = \frac{1}{(s-m^2)(u-m^2)} \text{Tr} \left[ (\not{p}'+m) \gamma^\mu (\not{p}+\not{k}+m) \gamma^\nu (\not{p}+m) \gamma_\mu (\not{p}-\not{k}'+m) \gamma_\nu \right]

1. 计算 TssT_{ss} 利用缩并恒等式 γν(+m)γν=2+4m\gamma^\nu (\not{p}+m) \gamma_\nu = -2\not{p} + 4m,中间部分化简为: Y=(++m)(2+4m)(++m)=4m2+(6m22s)+2ms+2m3Y = (\not{p}+\not{k}+m) (-2\not{p}+4m) (\not{p}+\not{k}+m) = 4m^2 \not{p} + (6m^2 - 2s) \not{k} + 2ms + 2m^3 再次缩并 γμYγμ=8m2+(4s12m2)+8ms+8m3\gamma^\mu Y \gamma_\mu = -8m^2 \not{p} + (4s - 12m^2) \not{k} + 8ms + 8m^3。代入迹中并利用运动学关系 2pp=2m2t2p'\cdot p = 2m^2-t2pk=s+tm22p'\cdot k = s+t-m^2,得到: Tss=8(sm2)2(s2+stm2t+3m4)=8(sm2)2(su+3sm2+um2+m4)T_{ss} = \frac{8}{(s-m^2)^2} \left( s^2 + st - m^2 t + 3m^4 \right) = \frac{8}{(s-m^2)^2} \left( -su + 3sm^2 + um^2 + m^4 \right)

2. 计算 TuuT_{uu}sus \leftrightarrow u 的对称性,直接将 TssT_{ss} 中的 ss 替换为 uuTuu=8(um2)2(su+3um2+sm2+m4)T_{uu} = \frac{8}{(u-m^2)^2} \left( -su + 3um^2 + sm^2 + m^4 \right)

3. 计算交叉项 TsuT_{su} 利用恒等式 γμγνγμ=2γν\gamma^\mu \not{A} \gamma^\nu \not{B} \gamma_\mu = -2 \not{B} \gamma^\nu \not{A} 等,化简迹内的矩阵乘积,计算各项的迹并代入运动学关系 (p+k)(pk)=m2(p+k)\cdot(p-k') = m^2 等,得到: 2Tsu=16m2(2m2+s+u)(sm2)(um2)2T_{su} = \frac{16m^2 (2m^2 + s + u)}{(s-m^2)(u-m^2)}

4. 组合结果 将上述结果相加,并进行代数重组(将部分交叉项吸收到平方项中): T2=2e4[su+5sm2+um2m4(sm2)2+su+5um2+sm2m4(um2)2+8m4(sm2)(um2)]\langle |\mathcal{T}|^2 \rangle = 2e^4 \left[ \frac{-su + 5sm^2 + um^2 - m^4}{(s-m^2)^2} + \frac{-su + 5um^2 + sm^2 - m^4}{(u-m^2)^2} + \frac{8m^4}{(s-m^2)(u-m^2)} \right] 进一步化简为标准形式: T2=2e4[m2usm2+m2sum2+4m2(1sm2+1um2)+4m4(1sm2+1um2)2]\langle |\mathcal{T}|^2 \rangle = 2e^4 \left[ \frac{m^2-u}{s-m^2} + \frac{m^2-s}{u-m^2} + 4m^2 \left( \frac{1}{s-m^2} + \frac{1}{u-m^2} \right) + 4m^4 \left( \frac{1}{s-m^2} + \frac{1}{u-m^2} \right)^2 \right]

5. 交叉对称性 (Crossing Symmetry) 分析 对于电子-正电子对湮灭过程 e(p1)+e+(p2)γ(k1)+γ(k2)e^-(p_1) + e^+(p_2) \rightarrow \gamma(k_1) + \gamma(k_2),其 Mandelstam 变量定义为 spair=(p1+p2)2s_{\text{pair}} = (p_1+p_2)^2, tpair=(p1k1)2t_{\text{pair}} = (p_1-k_1)^2, upair=(p1k2)2u_{\text{pair}} = (p_1-k_2)^2。该过程的自旋平均振幅平方已知为: Tpair2=2e4[um2tm2+tm2um24m2(1tm2+1um2)4m4(1tm2+1um2)2]\langle |\mathcal{T}_{\text{pair}}|^2 \rangle = 2e^4 \left[ \frac{u-m^2}{t-m^2} + \frac{t-m^2}{u-m^2} - 4m^2 \left( \frac{1}{t-m^2} + \frac{1}{u-m^2} \right) - 4m^4 \left( \frac{1}{t-m^2} + \frac{1}{u-m^2} \right)^2 \right] 通过交叉对称性,康普顿散射可以通过将初态光子移至末态、末态电子移至初态(变为正电子)来联系到对湮灭过程。这在运动学上等价于动量代换 pp1p \rightarrow p_1, kk1k \rightarrow -k_1, pp2p' \rightarrow -p_2, kk2k' \rightarrow k_2。 此时变量的映射关系为: stpair,tspair,uupairs \rightarrow t_{\text{pair}}, \quad t \rightarrow s_{\text{pair}}, \quad u \rightarrow u_{\text{pair}} 即发生了 sts \leftrightarrow t 的交换。同时,由于将一个费米子从末态移到了初态,根据费米子反交换关系,振幅平方需要引入一个全局负号。应用这一规则: TCompton2(s,t,u)=Tpair2(t,s,u)\langle |\mathcal{T}_{\text{Compton}}|^2 \rangle(s,t,u) = - \langle |\mathcal{T}_{\text{pair}}|^2 \rangle(t,s,u)sts \leftrightarrow t 代入 Tpair2\langle |\mathcal{T}_{\text{pair}}|^2 \rangle 并乘以 1-1Tpair2(t,s,u)=2e4[um2sm2+sm2um24m2(1sm2+1um2)4m4(1sm2+1um2)2]- \langle |\mathcal{T}_{\text{pair}}|^2 \rangle(t,s,u) = -2e^4 \left[ \frac{u-m^2}{s-m^2} + \frac{s-m^2}{u-m^2} - 4m^2 \left( \frac{1}{s-m^2} + \frac{1}{u-m^2} \right) - 4m^4 \left( \frac{1}{s-m^2} + \frac{1}{u-m^2} \right)^2 \right] 将负号吸收到前两项的分子中(注意 um2sm2=m2usm2-\frac{u-m^2}{s-m^2} = \frac{m^2-u}{s-m^2}),我们完美复现了直接计算得到的康普顿散射结果。

T2=2e4[m2usm2+m2sum2+4m2(1sm2+1um2)+4m4(1sm2+1um2)2]\boxed{ \langle |\mathcal{T}|^2 \rangle = 2e^4 \left[ \frac{m^2-u}{s-m^2} + \frac{m^2-s}{u-m^2} + 4m^2 \left( \frac{1}{s-m^2} + \frac{1}{u-m^2} \right) + 4m^4 \left( \frac{1}{s-m^2} + \frac{1}{u-m^2} \right)^2 \right] }

59.2

Problem 59.2

srednickiChapter 59

习题 59.2

来源: 第59章, PDF第355页


59.2 Compute T2\langle|\mathcal{T}|^2\rangle for Bhabha scattering, e+ee+ee^+ e^- \rightarrow e^+ e^-.

习题 59.2 - 解答


Bhabha 散射 (e+ee+ee^+ e^- \rightarrow e^+ e^-) 的过程在树图阶(Tree-level)包含两个费曼图:ss-道(正负电子湮灭为虚光子再产生正负电子)和 tt-道(正负电子通过交换虚光子发生散射)。

设初态电子和正电子的动量分别为 p1p_1p2p_2,末态电子和正电子的动量分别为 p3p_3p4p_4。定义 Mandelstam 变量: s=(p1+p2)2=(p3+p4)2s = (p_1 + p_2)^2 = (p_3 + p_4)^2 t=(p1p3)2=(p2p4)2t = (p_1 - p_3)^2 = (p_2 - p_4)^2 u=(p1p4)2=(p2p3)2u = (p_1 - p_4)^2 = (p_2 - p_3)^2 它们满足动量守恒关系:s+t+u=4m2s + t + u = 4m^2

根据 QED 费曼规则,总散射矩阵元 T\mathcal{T}ss-道与 tt-道之差(由于费米子交换反对称性,两图之间存在相对负号): iT=iTsiTti\mathcal{T} = i\mathcal{T}_s - i\mathcal{T}_t 其中: iTs=(ie)2[vˉ(p2)γμu(p1)]igμνs[uˉ(p3)γνv(p4)]i\mathcal{T}_s = (-ie)^2 \left[ \bar{v}(p_2) \gamma^\mu u(p_1) \right] \frac{-i g_{\mu\nu}}{s} \left[ \bar{u}(p_3) \gamma^\nu v(p_4) \right] iTt=(ie)2[uˉ(p3)γμu(p1)]igμνt[vˉ(p2)γνv(p4)]i\mathcal{T}_t = (-ie)^2 \left[ \bar{u}(p_3) \gamma^\mu u(p_1) \right] \frac{-i g_{\mu\nu}}{t} \left[ \bar{v}(p_2) \gamma^\nu v(p_4) \right]

我们需要计算对初态自旋求平均、对末态自旋求和的模方 T2=14spinsTsTt2\langle|\mathcal{T}|^2\rangle = \frac{1}{4} \sum_{\text{spins}} |\mathcal{T}_s - \mathcal{T}_t|^2T2=e44(Tsss2+Tttt22Tstst)\langle|\mathcal{T}|^2\rangle = \frac{e^4}{4} \left( \frac{T_{ss}}{s^2} + \frac{T_{tt}}{t^2} - \frac{2 T_{st}}{st} \right) 其中 Tss,Ttt,TstT_{ss}, T_{tt}, T_{st} 分别为对应的自旋求和迹(Traces)。

1. 计算 ss-道项 TssT_{ss}

Tss=Tr[γμ(̸p1+m)γν(̸p2m)]Tr[γμ(̸p4m)γν(̸p3+m)]T_{ss} = \text{Tr}\left[ \gamma^\mu (\not{p}_1 + m) \gamma^\nu (\not{p}_2 - m) \right] \text{Tr}\left[ \gamma_\mu (\not{p}_4 - m) \gamma_\nu (\not{p}_3 + m) \right] 利用标准迹定理 Tr[γμγν]=4(aμbν+aνbμgμνab)\text{Tr}[\gamma^\mu \not{a} \gamma^\nu \not{b}] = 4(a^\mu b^\nu + a^\nu b^\mu - g^{\mu\nu} a \cdot b),展开并收缩洛伦兹指标: Tss=32[(p1p4)(p2p3)+(p1p3)(p2p4)+m2(p1p2+p3p4)+2m4]T_{ss} = 32 \left[ (p_1 \cdot p_4)(p_2 \cdot p_3) + (p_1 \cdot p_3)(p_2 \cdot p_4) + m^2(p_1 \cdot p_2 + p_3 \cdot p_4) + 2m^4 \right] 代入运动学关系 p1p4=p2p3=12(2m2u)p_1 \cdot p_4 = p_2 \cdot p_3 = \frac{1}{2}(2m^2 - u)p1p3=p2p4=12(2m2t)p_1 \cdot p_3 = p_2 \cdot p_4 = \frac{1}{2}(2m^2 - t),以及 p1p2=p3p4=12(s2m2)p_1 \cdot p_2 = p_3 \cdot p_4 = \frac{1}{2}(s - 2m^2),化简得到: Tss=8(t2+u2+8m2s8m4)T_{ss} = 8 \left( t^2 + u^2 + 8m^2 s - 8m^4 \right)

2. 计算 tt-道项 TttT_{tt}

Ttt=Tr[γμ(̸p1+m)γν(̸p3+m)]Tr[γμ(̸p4m)γν(̸p2m)]T_{tt} = \text{Tr}\left[ \gamma^\mu (\not{p}_1 + m) \gamma^\nu (\not{p}_3 + m) \right] \text{Tr}\left[ \gamma_\mu (\not{p}_4 - m) \gamma_\nu (\not{p}_2 - m) \right] 根据交叉对称性(Crossing Symmetry),tt-道的迹可以通过在 ss-道结果中交换 p2p3p_2 \leftrightarrow -p_3 得到,这等价于交换 Mandelstam 变量 sts \leftrightarrow tTtt=8(s2+u2+8m2t8m4)T_{tt} = 8 \left( s^2 + u^2 + 8m^2 t - 8m^4 \right)

3. 计算干涉项 TstT_{st}

干涉项的迹包含在一个大迹中: Tst=Tr[γμ(̸p1+m)γν(̸p3+m)γμ(̸p4m)γν(̸p2m)]T_{st} = \text{Tr}\left[ \gamma^\mu (\not{p}_1 + m) \gamma^\nu (\not{p}_3 + m) \gamma_\mu (\not{p}_4 - m) \gamma_\nu (\not{p}_2 - m) \right] 利用 Gamma 矩阵的收缩恒等式 γμγμ=4\gamma^\mu \gamma_\mu = 4, γμγμ=2\gamma^\mu \not{a} \gamma_\mu = -2\not{a}, γμγμ=4ab\gamma^\mu \not{a} \not{b} \gamma_\mu = 4 a \cdot b 等,将 γμ\gamma^\muγμ\gamma_\mu 收缩: Tst=Tr[{2̸p3γν̸p1+4m(p1ν+p3ν)2m2γν}(̸p4m)γν(̸p2m)]T_{st} = \text{Tr}\left[ \{-2\not{p}_3 \gamma^\nu \not{p}_1 + 4m(p_1^\nu + p_3^\nu) - 2m^2 \gamma^\nu\} (\not{p}_4 - m) \gamma_\nu (\not{p}_2 - m) \right] 展开并保留偶数个 Gamma 矩阵的项,计算迹后代入 Mandelstam 变量: Tst=32(p1p4)(p2p3)+16m2(p1p3+p2p4)16m2(su)32m4T_{st} = -32(p_1 \cdot p_4)(p_2 \cdot p_3) + 16m^2(p_1 \cdot p_3 + p_2 \cdot p_4) - 16m^2(s - u) - 32m^4 进一步化简可得: Tst=8(u28m2u+12m4)T_{st} = -8 \left( u^2 - 8m^2 u + 12m^4 \right)

4. 组合得到精确结果

Tss,Ttt,TstT_{ss}, T_{tt}, T_{st} 代入 T2\langle|\mathcal{T}|^2\rangle 的表达式中: T2=e44[8(t2+u2+8m2s8m4)s2+8(s2+u2+8m2t8m4)t22[8(u28m2u+12m4)]st]\langle|\mathcal{T}|^2\rangle = \frac{e^4}{4} \left[ \frac{8(t^2 + u^2 + 8m^2 s - 8m^4)}{s^2} + \frac{8(s^2 + u^2 + 8m^2 t - 8m^4)}{t^2} - \frac{2[-8(u^2 - 8m^2 u + 12m^4)]}{st} \right] 提取公因子 2e42e^4,得到包含电子质量 mm 的精确自旋平均矩阵元模方:

T2=2e4[t2+u2+8m2s8m4s2+s2+u2+8m2t8m4t2+2(u28m2u+12m4)st]\boxed{ \langle|\mathcal{T}|^2\rangle = 2e^4 \left[ \frac{t^2 + u^2 + 8m^2 s - 8m^4}{s^2} + \frac{s^2 + u^2 + 8m^2 t - 8m^4}{t^2} + \frac{2(u^2 - 8m^2 u + 12m^4)}{st} \right] }

5. 高能极限(无质量近似)

在极高能散射条件下(s,t,um2s, -t, -u \gg m^2),可以忽略电子质量取 m0m \to 0。此时 s+t+u0s + t + u \approx 0,精确公式大幅简化为: T22e4(t2+u2s2+s2+u2t2+2u2st)\langle|\mathcal{T}|^2\rangle \approx 2e^4 \left( \frac{t^2 + u^2}{s^2} + \frac{s^2 + u^2}{t^2} + \frac{2u^2}{st} \right) 利用 u2(1s2+1t2+2st)=u2(s+t)2s2t2=u4s2t2u^2 \left( \frac{1}{s^2} + \frac{1}{t^2} + \frac{2}{st} \right) = u^2 \frac{(s+t)^2}{s^2 t^2} = \frac{u^4}{s^2 t^2},该结果可以写成高度对称的紧凑形式:

T22e4s4+t4+u4s2t2\boxed{ \langle|\mathcal{T}|^2\rangle \approx 2e^4 \frac{s^4 + t^4 + u^4}{s^2 t^2} }
59.3

Problem 59.3

srednickiChapter 59

习题 59.3

来源: 第59章, PDF第355页


59.3 Compute T2\langle|\mathcal{T}|^2\rangle for Møller scattering, eeeee^- e^- \rightarrow e^- e^-. You should find that your result is the same as that for e+ee+ee^+ e^- \rightarrow e^+ e^-, but with sus \leftrightarrow u. This is another example of crossing symmetry.

习题 59.3 - 解答


对于 Møller 散射 e(p1)e(p2)e(p3)e(p4)e^-(p_1) e^-(p_2) \rightarrow e^-(p_3) e^-(p_4),最低阶的费曼图包含 tt-通道和 uu-通道的光子交换。根据费米-狄拉克统计,交换末态两个全同费米子会引入一个相对负号。散射振幅 T\mathcal{T} 可以写为: T=TtTu\mathcal{T} = \mathcal{T}_t - \mathcal{T}_u 其中,两个通道的振幅分别为: Tt=e2t[uˉ(p3)γμu(p1)][uˉ(p4)γμu(p2)]\mathcal{T}_t = \frac{e^2}{t} [\bar{u}(p_3) \gamma^\mu u(p_1)] [\bar{u}(p_4) \gamma_\mu u(p_2)] Tu=e2u[uˉ(p4)γνu(p1)][uˉ(p3)γνu(p2)]\mathcal{T}_u = \frac{e^2}{u} [\bar{u}(p_4) \gamma^\nu u(p_1)] [\bar{u}(p_3) \gamma_\nu u(p_2)] 我们需要计算对初态自旋求平均、对末态自旋求和的振幅平方: T2=14spinsTtTu2=14spins(Tt2+Tu22Re(TtTu))\langle|\mathcal{T}|^2\rangle = \frac{1}{4} \sum_{\text{spins}} |\mathcal{T}_t - \mathcal{T}_u|^2 = \frac{1}{4} \sum_{\text{spins}} \left( |\mathcal{T}_t|^2 + |\mathcal{T}_u|^2 - 2\text{Re}(\mathcal{T}_t \mathcal{T}_u^*) \right)

在计算迹之前,先列出曼德尔施塔姆变量(Mandelstam variables)与动量内积的关系(利用 pi2=m2p_i^2 = m^2): s=(p1+p2)2=(p3+p4)2    p1p2=p3p4=s2m2s = (p_1 + p_2)^2 = (p_3 + p_4)^2 \implies p_1 \cdot p_2 = p_3 \cdot p_4 = \frac{s}{2} - m^2 t=(p1p3)2=(p2p4)2    p1p3=p2p4=m2t2t = (p_1 - p_3)^2 = (p_2 - p_4)^2 \implies p_1 \cdot p_3 = p_2 \cdot p_4 = m^2 - \frac{t}{2} u=(p1p4)2=(p2p3)2    p1p4=p2p3=m2u2u = (p_1 - p_4)^2 = (p_2 - p_3)^2 \implies p_1 \cdot p_4 = p_2 \cdot p_3 = m^2 - \frac{u}{2}

1. 计算 tt-通道平方项 14Tt2\frac{1}{4} \sum |\mathcal{T}_t|^2

利用自旋求和公式 u(p)uˉ(p)=+m\sum u(p)\bar{u}(p) = \not{p} + m,可得: 14Tt2=e44t2Tr[γμ(̸p1+m)γν(̸p3+m)]Tr[γμ(̸p2+m)γν(̸p4+m)]\frac{1}{4} \sum |\mathcal{T}_t|^2 = \frac{e^4}{4t^2} \text{Tr}[\gamma^\mu (\not{p}_1 + m) \gamma^\nu (\not{p}_3 + m)] \text{Tr}[\gamma_\mu (\not{p}_2 + m) \gamma_\nu (\not{p}_4 + m)] 计算狄拉克矩阵的迹: Tr[γμ(̸p1+m)γν(̸p3+m)]=4(p1μp3ν+p1νp3μgμν(p1p3m2))\text{Tr}[\gamma^\mu (\not{p}_1 + m) \gamma^\nu (\not{p}_3 + m)] = 4\left( p_1^\mu p_3^\nu + p_1^\nu p_3^\mu - g^{\mu\nu}(p_1 \cdot p_3 - m^2) \right) 将两个迹相乘并收缩洛伦兹指标: 14Tt2=4e4t2[2(p1p2)(p3p4)+2(p1p4)(p2p3)2m2(p1p3+p2p4)+4m4]=8e4t2[(s2m2)2+(u2m2)2m2(2m2t)+2m4]=2e4t2[(s2m2)2+(u2m2)2+4m2t]\begin{aligned} \frac{1}{4} \sum |\mathcal{T}_t|^2 &= \frac{4e^4}{t^2} \left[ 2(p_1 \cdot p_2)(p_3 \cdot p_4) + 2(p_1 \cdot p_4)(p_2 \cdot p_3) - 2m^2(p_1 \cdot p_3 + p_2 \cdot p_4) + 4m^4 \right] \\ &= \frac{8e^4}{t^2} \left[ \left(\frac{s}{2} - m^2\right)^2 + \left(\frac{u}{2} - m^2\right)^2 - m^2\left(2m^2 - t\right) + 2m^4 \right] \\ &= \frac{2e^4}{t^2} \left[ (s - 2m^2)^2 + (u - 2m^2)^2 + 4m^2 t \right] \end{aligned}

2. 计算 uu-通道平方项 14Tu2\frac{1}{4} \sum |\mathcal{T}_u|^2

由于 Tu\mathcal{T}_u 可以通过在 Tt\mathcal{T}_t 中交换 p3p4p_3 \leftrightarrow p_4 得到,这在运动学上等价于交换 tut \leftrightarrow u。因此直接利用对称性可得: 14Tu2=2e4u2[(s2m2)2+(t2m2)2+4m2u]\frac{1}{4} \sum |\mathcal{T}_u|^2 = \frac{2e^4}{u^2} \left[ (s - 2m^2)^2 + (t - 2m^2)^2 + 4m^2 u \right]

3. 计算干涉项 12Re(TtTu)-\frac{1}{2} \sum \text{Re}(\mathcal{T}_t \mathcal{T}_u^*)

干涉项的自旋求和可以写为一个大迹: TtTu=e4tuTr[γμ(̸p1+m)γν(̸p4+m)γμ(̸p2+m)γν(̸p3+m)]\sum \mathcal{T}_t \mathcal{T}_u^* = \frac{e^4}{tu} \text{Tr}[ \gamma^\mu (\not{p}_1 + m) \gamma^\nu (\not{p}_4 + m) \gamma_\mu (\not{p}_2 + m) \gamma_\nu (\not{p}_3 + m) ] 为了简化计算,我们引入一个非常有用的 Chisholm 恒等式推论: γμ(+m)γν(+m)γμ=2(+m)γν(+m)\gamma^\mu (\not{A} + m) \gamma^\nu (\not{B} + m) \gamma_\mu = -2 (\not{B} + m) \gamma^\nu (\not{A} + m) 应用此恒等式(取 A=p1,B=p4A=p_1, B=p_4),迹化简为: Tr[]=2Tr[(̸p4+m)γν(̸p1+m)(̸p2+m)γν(̸p3+m)]\text{Tr}[\dots] = -2 \text{Tr}[ (\not{p}_4 + m) \gamma^\nu (\not{p}_1 + m) (\not{p}_2 + m) \gamma_\nu (\not{p}_3 + m) ] 接着化简中间部分 γν(̸p1+m)(̸p2+m)γν\gamma^\nu (\not{p}_1 + m) (\not{p}_2 + m) \gamma_\nuγν(̸p1̸p2+m̸p1+m̸p2+m2)γν=4(p1p2)2m(̸p1+̸p2)+4m2=4(s2m2)+4m22m(̸p1+̸p2)=2s2m(̸p1+̸p2)\begin{aligned} \gamma^\nu (\not{p}_1 \not{p}_2 + m\not{p}_1 + m\not{p}_2 + m^2) \gamma_\nu &= 4(p_1 \cdot p_2) - 2m(\not{p}_1 + \not{p}_2) + 4m^2 \\ &= 4\left(\frac{s}{2} - m^2\right) + 4m^2 - 2m(\not{p}_1 + \not{p}_2) \\ &= 2s - 2m(\not{p}_1 + \not{p}_2) \end{aligned} 将其代回迹中,得到: Tr[]=2{2sTr[(̸p4+m)(̸p3+m)]2mTr[(̸p4+m)(̸p1+̸p2)(̸p3+m)]}\text{Tr}[\dots] = -2 \left\{ 2s \text{Tr}[(\not{p}_4 + m)(\not{p}_3 + m)] - 2m \text{Tr}[(\not{p}_4 + m)(\not{p}_1 + \not{p}_2)(\not{p}_3 + m)] \right\} 分别计算这两个迹(奇数个 γ\gamma 矩阵的迹为零):

  1. Tr[(̸p4+m)(̸p3+m)]=4(p3p4+m2)=4(s/2)=2s\text{Tr}[(\not{p}_4 + m)(\not{p}_3 + m)] = 4(p_3 \cdot p_4 + m^2) = 4(s/2) = 2s
  2. Tr[(̸p4+m)(̸p1+̸p2)(̸p3+m)]=4m[p4(p1+p2)+p3(p1+p2)]=4m(p3+p4)(p1+p2)\text{Tr}[(\not{p}_4 + m)(\not{p}_1 + \not{p}_2)(\not{p}_3 + m)] = 4m [p_4 \cdot (p_1 + p_2) + p_3 \cdot (p_1 + p_2)] = 4m (p_3 + p_4) \cdot (p_1 + p_2) 由动量守恒 p3+p4=p1+p2p_3 + p_4 = p_1 + p_2,上式等于 4m(p1+p2)2=4ms4m (p_1 + p_2)^2 = 4ms

将结果组合起来: Tr[]=2[2s(2s)2m(4ms)]=8s(s2m2)\text{Tr}[\dots] = -2 \left[ 2s(2s) - 2m(4ms) \right] = -8s(s - 2m^2) 因此,干涉项为: 12Re(TtTu)=12e4tu[8s(s2m2)]=4e4tus(s2m2)-\frac{1}{2} \sum \text{Re}(\mathcal{T}_t \mathcal{T}_u^*) = -\frac{1}{2} \frac{e^4}{tu} \left[ -8s(s - 2m^2) \right] = \frac{4e^4}{tu} s(s - 2m^2)

4. 最终结果与交叉对称性 (Crossing Symmetry)

将三部分相加,得到 Møller 散射的自旋平均振幅平方: T2=2e4[(s2m2)2+(u2m2)2+4m2tt2+(s2m2)2+(t2m2)2+4m2uu2+2s(s2m2)tu]\boxed{ \langle|\mathcal{T}|^2\rangle = 2e^4 \left[ \frac{(s-2m^2)^2 + (u-2m^2)^2 + 4m^2 t}{t^2} + \frac{(s-2m^2)^2 + (t-2m^2)^2 + 4m^2 u}{u^2} + \frac{2s(s-2m^2)}{tu} \right] }

与 Bhabha 散射的比较: 对于 Bhabha 散射 (e+ee+ee^+ e^- \rightarrow e^+ e^-),其过程由 ss-通道和 tt-通道主导,标准的自旋平均振幅平方结果为: TBhabha2=2e4[(s2m2)2+(u2m2)2+4m2tt2+(t2m2)2+(u2m2)2+4m2ss2+2u(u2m2)st]\langle|\mathcal{T}_{\text{Bhabha}}|^2\rangle = 2e^4 \left[ \frac{(s-2m^2)^2 + (u-2m^2)^2 + 4m^2 t}{t^2} + \frac{(t-2m^2)^2 + (u-2m^2)^2 + 4m^2 s}{s^2} + \frac{2u(u-2m^2)}{st} \right] 根据交叉对称性,将 Bhabha 散射中的入射正电子与出射正电子的动量反号并交换,在曼德尔施塔姆变量上的体现即为 sus \leftrightarrow u 的替换。 若我们在 TBhabha2\langle|\mathcal{T}_{\text{Bhabha}}|^2\rangle 中执行 sus \leftrightarrow u 的代换,上式直接变为: 2e4[(u2m2)2+(s2m2)2+4m2tt2+(t2m2)2+(s2m2)2+4m2uu2+2s(s2m2)ut]2e^4 \left[ \frac{(u-2m^2)^2 + (s-2m^2)^2 + 4m^2 t}{t^2} + \frac{(t-2m^2)^2 + (s-2m^2)^2 + 4m^2 u}{u^2} + \frac{2s(s-2m^2)}{ut} \right] 这与我们刚刚严格推导出的 Møller 散射结果完全一致,完美验证了交叉对称性。