PulseShape 0.1.2

A pulse shaping program for EPR!

PusleShape

PulseShape is an EasySpin pulse function clone written in python. The major purpose for rewriting pulse in Python is to free the function from the proprietary MATLAB universe and make it easier to use on Linux systems that often ship with e580 spectrometers.

PulseShape is built around the Pulse object which accepts arguments similar to those used by the easyspin pulse function.

Installation

PulseShape can be installed and updated using pip, the python package manager.

pip install PulseShape

PulseShape is tested to work on python 3.6-3.9. While one of the major purposes of PulseShape is to work on Linux systems, PulseShape works well on all systems (Windows, Mac and Linux) and only depends on numpy and scipy.

Alternatively, PulseShape can be installed by downloading or cloning the git repository.

git clone https://gitlab.com/mtessmer/PulseShape.git
cd PulseShape
python setup.py install

e580 Setup

Instructions for setting up python and PulseShape on the Linux system that usually ships with e580 spectrometers coming soon.

Example: sech\tanh pulse with resonator compensation

PulseShape

EasySpin

import numpy as np import matplotlib.pyplot as plt from PulseShape import Pulse profile = np.loadtxt('data/Transferfunction.dat') pulse = Pulse(pulse_time=0.150, time_step=0.000625, flip=np.pi, freq=[40, 120], type='sech/tanh', beta=10, profile=profile) plt.figure(figsize=(5, 5)) plt.plot(pulse.time * 1000, pulse.IQ.real, label='real') plt.plot(pulse.time * 1000, pulse.IQ.imag, label='imaginary') plt.xlabel('time (ns)') plt.ylabel('Amplitude') plt.legend() plt.show()

Par = struct Par.Type = 'sech/tanh'; Par.beta = 10; Par.tp = 0.150; Par.Phase = 0; Par.Flip = pi; Par.Frequency = [40 120] Par.TimeStep=0.000625 filename = 'Transferfunction.dat'; delimiter = ' '; formatSpec = '%f%f%[^\n\r]'; fileID = fopen(filename,'r'); dataArray = textscan(fileID, formatSpec, 'Delimiter', ... delimiter, 'MultipleDelimsAsOne', ... true, 'TextType', 'string'); fclose(fileID); Par.FrequencyResponse = [dataArray{:, 1}, dataArray{:, 2}]; [t, IQ] = pulse(Par) [t, IQ, modulation] = pulse(Par) figure(1) hold on plot(t, real(IQ)) plot(t, imag(IQ)) xlabel('time ns') ylabel('Amplitude') x0=10; y0=10; width=465; height=448; set(gcf,'position',[x0,y0,width,height])

Example: Working with multiple pulses

All time, IQ, other paramters and data are stored withing the Pulse object itself so it's easy to work with multiple pulses

import numpy as np
import matplotlib.pyplot as plt
from PulseShape import Pulse

profile = np.loadtxt('data/Transferfunction.dat')
st_pulse = Pulse(pulse_time=0.150,
                 time_step=0.000625,
                 flip=np.pi,
                 freq=[40, 120],
                 type='sech/tanh',
                 beta=10,
                 profile=profile)

g_pulse = Pulse(pulse_time=0.06,
                time_step=0.000625,
                flip=np.pi,
                type='gaussian',
                trunc=0.1)

offsets = np.linspace(-20, 140, 256)
st_pulse.exciteprofile(offsets)
g_pulse.exciteprofile(offsets)

fig, (ax1, ax2) = plt.subplots(2, figsize=(8, 10))
ax1.set_title('Pulse IQ')
ax1.plot(st_pulse.time * 1e3, st_pulse.IQ.real, label=r'sech/tanh $\Re$', color='C0')
ax1.plot(st_pulse.time * 1e3, st_pulse.IQ.imag, label=r'sech/tanh $\Im$', alpha=0.5, color='C0')
ax1.plot(g_pulse.time * 1e3, g_pulse.IQ.real, label='gaussian', color='C1')
ax1.set_ylabel('Amplitude')
ax1.set_ylabel("Time (ns)")
ax1.legend()

ax2.set_title('Excitation Profile')
ax2.plot(offsets, st_pulse.Mz)
ax2.plot(offsets, g_pulse.Mz)
ax2.set_xlabel('Frequency Offset (MHz)')
ax2.set_ylabel('Mz')
plt.show()