Fund Project:Project supported by the Scientific and Technologial Innovation Program of Higher Education Institutions in Shanxi, China (Grant Nos. 2019L0785, 2019L0813) and the Beijing Natural Science Foundation, China (Grant No. 1182009).
Received Date:17 May 2019
Accepted Date:06 August 2019
Available Online:01 October 2019
Published Online:20 October 2019
Abstract:The interferometry of two Kapitza-Dirac (KD) pulses acting on cold atoms in a harmonic oscillator potential well is investigated by Feynman path integral method. The wave function and density distribution function of the system at a given time are calculated analytically by using the propagator under the action of an external field. The first KD pulse acts on cold atoms to produce a large number of modes in the harmonic oscillator potential well. The maximum value of wave packet of mode 0 is larger than those of other modes. These modes evolve along different paths. The external field changes the phase of each mode and makes the evolution path of the mode deviate from that without the external field. When the second KD pulse is added, it splits the mode of the first KD pulse, and thus generates more modes. These modes will evolve along different paths under the action of external field and harmonic potential well, and interfere with each other. At the moment of measurement, all the wave packets are separated without overlapping. The effect of the external field does not change the magnitude of the density distribution at the time of measurement, but makes the wave packet of each mode shift. The phase difference between adjacent modes decreases linearly with the increase of field intensity. When the external field is a gravity field, we calculate the Fisher information and the Cramer-Rao lowér bound. The Fisher information is proportional to the mass of atoms and inversely to the third power of harmonic potential well frequency. We can improve the measurement accuracy of interferometer by reducing the frequency of harmonic potential well and increasing atomic mass. When the initial state is the ground state of the harmonic potential well, the accuracy of the gravity acceleration measured by the interference device can be obtained to be 10–9 by using the experimental parameters. The initial state is the ground state of the harmonic potential well and the external field, and the calculation result indicates that the measurement accuracy will decrease. At the same time, the enhancement of inter-atomic repulsion and attraction interaction will also lead the measurement accuracy to increase. Keywords:cold atoms/ harmonic oscillator potential/ propagator/ Kapitza-Dirac pulse
全文HTML
--> --> -->
2.简谐势阱中冷原子受两次KD脉冲的模型考虑一团冷原子处在频率为$\omega $简谐势阱的基态(如图1所示). 在$t = 0$时刻加第一次 KD脉冲, 初始波包被劈裂成许多动量不同的模式(模式的运动轨迹在图1中用不同的颜色标记). 这些模式在简谐势阱和外场$\beta $的共同作用下演化. 当t = $ {\text{π}}/(2\omega )$, 相邻模式之间的距离达到最大, 加第二次KD脉冲会劈裂$t = {\text{π}}/(2\omega )$时刻的波包, 产生出更多的模式. 在${t_{\rm{f}}} = {\text{π}}/\omega $, 不同模式相干叠加. 通过测量原子在每个模式的分布, 可以获得外场强度. 图 1 简谐势阱中冷原子受到两次KD脉冲的示意图 在测量时刻${t_{\rm{f}}} ={\text{π}}/\omega $, 不同模式相干叠加, $\beta $为外场 Figure1. Diagram of cold atoms with two KD pulses in harmonic oscillator potential. At the measurement time ${t_{\rm{f}}} = {\text{π}}/\omega $, the coherent superposition of different modes occurs. $\beta $ is the external field.
采用简谐势阱的特征长度$\sigma $作为长度量纲, 简谐势阱的特征时间$1/\omega $作为时间量纲对系统进行无量纲化. 系统在两次KD脉冲后的态密度分布函数含时演化规律如图2、图3和图4所示. 本节所有图中和文中的参数都已无量纲化. 图2中, 两次KD脉冲的强度都是$V = 1$. 第一次KD脉冲加在时刻$t = 0$. 由于$V = 1$, 所以初始的波包被劈裂成3个模式($l = 0, \pm 1$). l来自(1)式中第一类贝塞尔的下角标, 用来标记不同模式. 这三个模式在简谐势阱中分别沿不同的路径演化. 从图中可以看出, 上下模式($l = \pm 1$)的态密度分布是对称的. 中间模式($l = 0$)的态密度分布的最大值要比其他两个的模式大一些, 也就是说大部分冷原子集中在中间的模式上. 当系统演化到时刻$t = {\text{π}}/2$, 上、下两个模式离中间模式的距离达到最大$\Delta L = (M - 1)k$. 此时加第二次KD脉冲, 之前每个模式又被劈裂成3个模式($l' = 0, \;\pm 1$). 为了与l区别, l'来自第二次KD脉冲中第一类贝塞尔的下角标. 这些不同的模式在简谐势阱的作用下演化, 同时发生相干叠加. 从图中我们可以看出在测量时刻$t = {\text{π}}$, ($l = 1,\; l' = 1$)、($l = 0,\; l' = 1$)和($l = - 1,\; l' = 1$)这三个模式相干叠加, 形成了上面的波包. 同理, 中间的波包是由$l = - 1,\; 0,\; 1$三个模式的$l' = 0$分量相干叠加形成的. 下面的波包是由$l = - 1,\; 0,\; 1$三个模式的$l' = - 1$分量相干叠加形成的. 相对$t = 3{\text{π}}/4$时刻的态密度分布, $t = {\text{π}}$时刻的波包分布形式上更加简洁, 有助于我们在下一节来计算系统的测量精度. 图 2 没有外场的情况下($\beta = 0$), 系统态密度分布函数的演化规律 无量纲参数为$V = 1,\; k = 10$, 测量时刻在$t = {t_{\rm{f}}} = {\text{π}}$, 两次KD脉冲都用红色箭头表示, 态密度演化图右边的彩色条表示态密度分布函数值由低(蓝色)变到高(红色) Figure2. In the absence of an external field ($\beta = 0$), the density distribution function of system varies with time. The dimensionless parameter are $V = 1,\; k = 10$. The measurement time is at $t = {t_{\rm{f}}} ={\text{π}}$. Both KD pulses are represented by red arrows. The color bar on the right side of the density of states evolution diagram indicates that the value of the density distribution function changes from low (blue) to high (red).
图 3 在外场$\beta = 1$的情况下, 系统态密度分布函数的演化规律 图中除$\beta $之外的其他无量纲参数与图2相同, 红色箭头为KD脉冲, 测量时刻在$t = {t_{\rm{f}}} = {\text{π}}$ Figure3. In the case of external field $\beta = 1$, the evolution of density distribution function of the system is obtained. The dimensionless parameters in this figure are the same as those in Figure 2. The red arrow are KD pulses. The measurement time is at $t = {t_{\rm{f}}} = {\text{π}}$.
图 4 在外场$\beta = 2$的情况下, 系统态密度分布函数的演化规律 图中除$\beta $之外的其他无量纲参数与图2相同, 红色的箭头为KD脉冲, 测量时刻在$t = {t_{\rm{f}}} = {\text{π}}$ Figure4. In the case of external field $\beta = 2$, the evolution of density distribution function of the system is obtained. The dimensionless parameters in this figure are the same as those in Figure 2. The red arrow are KD pulses. The measurement time is at $t = {t_{\rm{f}}} = {\text{π}}$.
图3显示了外场强度$\beta = 1$的情况下, 系统的整体态密度演化. 相对于图2没有外场的情况, 第一次KD脉冲后, $l = 0, \pm 1$模式由于外场的作用向X轴负方向发生偏移. 在$t = {\text{π}}/2$时刻, 偏移的距离为$\left| { - \beta } \right| = 1$. 在第二次KD脉冲后, 所有的模式继续在外场的作用下发生整体偏移, 但这并不影响测量时刻上中下三个波包的形成. 波包中心偏移$x = 0$的距离为$\left| { - 2\beta } \right| = 2$. 这个偏移的距离从图5中也可以得到. 这说明测量时刻的波包包含了外场的信息. 我们可以通过测量时刻不同模式的分布来反向计算外场的强度. 图 5图2、图3、图4中测量时刻${t_{\rm{f}}} = {\text{π}}$态密度的分布规律 Figure5. The density distribution functions at measuring time${t_{\rm{f}}} = {\text{π}}$ for figure 2, 3, and 4.
得到不同非线性相互作用下的基态, 然后采用这些基态作为初始态, 使用算符劈裂法[29]对系统进行了数值演化, 得到测量时刻的态密度分布函数, 最后数值计算(14)式和Cramer-Rao lowér bound. 在以上计算中, 采用3.3节的无量纲化. 以简谐势阱的特征长度$\sigma $为长度量纲, 特征时间$1/\omega $为时间量纲, 对(31)式子进行了无量纲化. 下文中所有的参数都是无量纲的. 图6和图7显示了系统在不同非线性相互作用下的基态, 从图中可以发现原子间的排斥相互作用使初始的波包变宽, 吸引相互作用使初始的波包变窄. 图 6 外场$\beta = 1$的情况下系统基态的态密度分布函数, 不同的颜色对应不同非线性相互作用下的态密度分布函数 Figure6. In the case of external field $\beta = 1$, the density distribution function of the ground state of the system. Different colors correspond to the density distribution function under different non-linear interactions.
图 7 在外场$\beta = 1$的情况下系统基态的态密度分布函数, 不同的颜色对应不同非线性相互作用下的态密度分布函数 Figure7. In the case of external field $\beta = 1$, the density distribution function of the ground state of the system. Different colors correspond to the density distribution function under different non-linear interactions.
从图8可以看出, 即使在存在非线性相互作用的情况下, 不同模式在测量时刻仍然满足非重叠条件. 随着排斥相互作用的增强, 每个模式波包上的振荡幅度变得更大. 图9是图8中0模式的放大图. 在这个图中, 排斥相互作用对测量时刻态密度分布的影响显示更加清晰. 图 8 在$V = 1,\;k = 10,\; \beta = {\rm{1}}$的情况下测量时刻的态密度分布函数, 不同的颜色对应不同非线性相互作用下的态密度分布函数 Figure8. In the case of $V = 1,\;k = 10,\; \beta = {\rm{1}}$, the density distribution functions at measuring time. Different colors correspond to the density distribution function under different non-linear interactions.
图 9图8中0模式的放大图 Figure9. Detailed diagram of 0 mode in Fig. 8.
图10显示了在吸引相互作用的情况下, 系统测量时刻态密度的分布规律. 首先可以看出态密度分布函数满足非重叠条件, 其次和排斥相互作用的情况类似, 吸引相互作用的增强使测量时刻态密度上每一个波包的振荡程度加剧. 图11是图10中0模式得放大图, 在图11中可以更加清楚地看到这一点. 图 10 测量时刻系统态密度的分布函数, 除了非线性参数以外的其他无量纲参数和图8相同 Figure10. The density distribution functions at measuring time, dimensionless parameters other than non-linear parameters are the same as those in Fig. 8.
图 11图10中0模式的放大图 Figure11. Detailed diagram of 0 mode in Fig. 10.
当非线性参数改变时, 系统测量精度的变化规律显示在图12中. 我们首先计算了线性情况下(${g_{\rm{1}}} = {\rm{0}}$)系统的测量精度, 然后通过改变非线性参数计算了测量精度且给出了和线性情况下测量精度的比值. 从图12中可以看出, ${g_{\rm{1}}} = {\rm{0}}$时数值计算和解析计算的结果符合得很好. 随着排斥相互作用和吸引相互作用的增强, 系统的测量精度存在上升的趋势((Δβ非线性/β)/(Δβ线性/β)的值越小,系统的测量精度越高), 且相对于非线性参数的变化具有较好的对称性. 当$\left| {{g_1}} \right|$增加到2时, 测量精度会提升一个数量级. 实验上可以通过Feshbach共振技术[30]来调节吸引和排斥相互作用, 进而改变非线性相互作用对测量精度的影响. 图 12 系统的测量精度随非线性参数的变化规律, 除了非线性参数以外的其他无量纲参数和图8相同 Figure12. The variation of measuring accuracy of the system with nonlinear parameters, dimensionless parameters other than non-linear parameters are the same as those in Fig. 8.