MeV UED MicroED

to_cbf.py

import os
import numpy as np
import sys
import fabio
from glob import glob
from scipy.signal import medfilt2d

header = {
    'beam_center_x': 512,
    'beam_center_y': 512,
    'bit_depth_readout': 32,
    'count_time': 0.099996,
    'countrate_correction_applied': 1,
    'description': 'UED',
    'detector_distance': 0.150,
    'detector_number': 'E-08-0102',
    'detector_readout_time': 3.78e-06,
    'efficiency_correction_applied': 1,
    'flatfield_correction_applied': 1,
    'frame_time': 0.1,
    'pixel_mask_applied': 0,
    'sensor_material': 'Si',
    'sensor_thickness': 0.00032,
    'threshold_energy': 4680000,
    'virtual_pixel_correction_applied': 1,
    'x_pixel_size': 9.5e-06,
    'y_pixel_size': 9.5e-06,
}


files = sorted(glob(sys.argv[-1]))
print(sys.argv[-1])
print("found %d files" % len(files))

for idx, fn in enumerate(files):

    out_fn = "img_%04i.cbf" % (idx+1,)
    print(idx, fn, '-->', out_fn)

    frame_data = np.genfromtxt(fn)

    frame_data[frame_data < 0.0]  = 0.0

    frame_data = medfilt2d(frame_data, 9)

    cbf = fabio.cbfimage.cbfimage(header=header, data=frame_data.astype(np.int32))
    cbf.write(out_fn)

    os.system('adxv -sa img_%04i.cbf img_%04i.img' % (idx+1,idx+1))

ued.ins

TITL ued1 in P 1 2(1) 1
CELL 0.002650  5.441     7.208    33.988  90.000  86.400  90.000
ZERR 2.00 0.25 0.25 0.25 0.0 3.0 0.0
LATT -1
SYMM -X,Y+1/2,-Z
SFAC C  0.208  0.922  0.702  5.018  1.144 16.523  0.425 47.300  0.030 =
  0.0 0.0 0.000 0.770 12.011
SFAC H  0.042  0.991  0.103  3.924  0.205 11.597  0.170 31.984  0.006 =
  0.0 0.0 0.000 0.310 1.008
SFAC O  0.308  1.088  0.733  4.593  0.710 13.052  0.185 33.768  0.048 =
  0.0 0.0 0.000 0.660 15.999
SFAC N  0.216  0.843  0.649  4.055  0.968 12.693  0.345 36.058  0.032 =
  0.0 0.0 0.000 0.700 14.007
SFAC S  0.610  1.173  1.367  6.371  2.268 18.858  0.808 49.703  0.108 =
  0.0 0.0 0.000 1.030 32.060
UNIT 38 42 0 0 0
HKLF 4
END

XDS.INP

!*****************************************************************************
!
! XDS input file created by TJL Feb 2020 for solving MeV UED structure
! - 4.2 MeV e-
! - yaw geometry
! - median filter spots
!
!*****************************************************************************

!====================== DETECTOR PARAMETERS ==================================
! DETECTOR=UED
 MINIMUM_VALID_PIXEL_VALUE=0
 OVERLOAD= 32768  
 GAIN=1.0  ! pixel contents is interpreted as X-ray counts
 QX=0.0095  QY=0.0095  ! [mm]
 NX= 1024  NY= 1024    !  1M 

 UNTRUSTED_ELLIPSE=     445 545    465  565
 TRUSTED_REGION=0.02 1.41 !Relative radii limiting trusted detector region

 DIRECTION_OF_DETECTOR_X-AXIS= 1.0 0.0 0.0
 DIRECTION_OF_DETECTOR_Y-AXIS= 0.0 1.0 0.0


!====================== JOB CONTROL PARAMETERS ===============================
 JOB= XYCORR INIT COLSPOT IDXREF DEFPIX INTEGRATE CORRECT
 !JOB= CORRECT


!====================== GEOMETRICAL PARAMETERS ===============================
 ORGX= 488  ORGY= 510      !Detector origin (pixels).  ORGX=NX/2; ORGY=NY/2
 DETECTOR_DISTANCE= 1500   !(mm)

 ROTATION_AXIS= 0.0 1.0 0.0

 OSCILLATION_RANGE=0.2

 X-RAY_WAVELENGTH=0.00265           !Angstroem
 INCIDENT_BEAM_DIRECTION=0.0 0.0 1.0
 FRACTION_OF_POLARIZATION=0.5 !default=0.5 for unpolarized beam
 POLARIZATION_PLANE_NORMAL= 0.0 1.0 0.0


!======================= CRYSTAL PARAMETERS =================================
! TJL : only use if known a priori
! SPACE_GROUP_NUMBER=1
! UNIT_CELL_CONSTANTS= 77.4   77.4   77.4  90.0  90.0  90.0


!==================== SELECTION OF DATA IMAGES ==============================
!Generic file name and format (optional) of data images
 NAME_TEMPLATE_OF_DATA_FRAMES=../data/corrected_yaw_data/img_????.img

! DATA_RANGE=1 5
 DATA_RANGE=1 225

 BACKGROUND_RANGE=1 225  !Numbers of first and last data image for background

 SPOT_RANGE=1 225       !First and last data image number for finding spots


!================= PARAMETERS CONTROLLING REFINEMENTS =======================
 REFINE(IDXREF)=BEAM AXIS ORIENTATION CELL  ! POSITION
 REFINE(INTEGRATE)= ORIENTATION POSITION BEAM ! CELL AXIS
 REFINE(CORRECT)=POSITION BEAM ORIENTATION CELL AXIS


!================== CRITERIA FOR ACCEPTING REFLECTIONS ======================
 VALUE_RANGE_FOR_TRUSTED_DETECTOR_PIXELS= 10 30000 !Used by DEFPIX

 INCLUDE_RESOLUTION_RANGE=25 1.0 !Angstroem; used by DEFPIX,INTEGRATE,CORRECT

!MINIMUM_ZETA=0.05 !Defines width of 'blind region' (XPLAN,INTEGRATE,CORRECT)
!WFAC1=1.0  !This controls the number of rejected MISFITS in CORRECT;
        !a larger value leads to fewer rejections.


!============== DECISION CONSTANTS FOR FINDING CRYSTAL SYMMETRY =============
!Decision constants for detection of lattice symmetry (IDXREF, CORRECT)
! MAX_CELL_AXIS_ERROR=0.03 ! Maximum relative error in cell axes tolerated
! MAX_CELL_ANGLE_ERROR=2.0 ! Maximum cell angle error tolerated

!Decision constants for detection of space group symmetry (CORRECT).
!Resolution range for accepting reflections for space group determination in
!the CORRECT step. It should cover a sufficient number of strong reflections.
 TEST_RESOLUTION_RANGE=8.0 4.5
 MIN_RFL_Rmeas= 50 ! Minimum #reflections needed for calculation of Rmeas
 MAX_FAC_Rmeas=2.0 ! Sets an upper limit for acceptable Rmeas


!============== INTEGRATION AND PEAK PROFILE PARAMETERS =====================
!BEAM_DIVERGENCE=   0.80         !arctan(spot diameter/DETECTOR_DISTANCE)
!BEAM_DIVERGENCE_E.S.D.=   0.080 !half-width (Sigma) of BEAM_DIVERGENCE


!======= PARAMETERS CONTROLLING CORRECTION FACTORS (used by: CORRECT) =======
!MINIMUM_I/SIGMA=3.0 ! min(I/sigma) for reflections used in global refinement 
!SNRC=0.1         ! min(I/sigma) for reflections used for finding corrections
!BATCHSIZE=5.0    ! rotation range covered by a batch of consecutive images 
!CORRECTIONS= DECAY MODULATION ABSORPTION


!====================== INDEXING PARAMETERS =================================
!SEPMIN=4.0       ! default is 6 for other detectors
!LUSTER_RADIUS=2 ! default is 3 for other detectors
 MAXIMUM_ERROR_OF_SPOT_POSITION=6.0
 MAXIMUM_ERROR_OF_SPINDLE_POSITION=4.0
 MINIMUM_FRACTION_OF_INDEXED_SPOTS=0.1

!=========== PARAMETERS DEFINING BACKGROUND AND PEAK PIXELS =================
! TJL Note: the following are critical parameters for spotfinding
 STRONG_PIXEL=2.0                              !used by: COLSPOT
 MINIMUM_NUMBER_OF_PIXELS_IN_A_SPOT=25          !used by: COLSPOT
 NBX=15  NBY=15  !Define a rectangle of size (2*NBX+1)*(2*NBY+1)
 BACKGROUND_PIXEL=5.0                          !used by: COLSPOT,INTEGRATE

!SIGNAL_PIXEL=3.0                              !used by: INTEGRATE
!A pixel above the threshold contributes to the spot centroid
!FIXED_SCALE_FACTOR=TRUE  !Default is FALSE; used by : INIT,INTEGRATE

This documents TJL's attempt to solve the structure of 2,6‐Bis(4‐hexylphenyl)anthracene collected at the SLAC MeV UED source.

Feb 2020

Data Processing Steps

1. to_cbf.py :: Convert data into .img format. I loaded data via numpy, then used fabIO to add relevant metadata make CBFs, which somehow XDS did not like. So then I converted the CBF files to SMV using adxv, which seems to have worked. Then I did some basic corrections:

   • zero out negative intensities
   • size 9 median filter on all images

2. XDS :: Ran XDS to process the data. I had to tweak a lot of parameters to get the data to process, but the input file is included here. One note:

   • the final step of XDS crashed for me with an error saying it could not read the header of INTEGRATE.HKL. Cause: XDS saves INCIDENT_BEAM_DIRECTION in units of 1/wavelength, and the MeV wavelength is so small that it writes a number of order 100 for the z-axis value. This number is bigger than the white space allotted for that value, so it "merges" into the value immediately before it. Simply adding a space works.

3. pointless :: Identified an alternate space group as C 1 2 1, the similar reference structure we have is P 1 2(1) 1, so this is close.

4. shelxt :: The input file is included.

Results

SHELXT fails to converge. Phasing via direct methods failed.

Improvements

Are necessary. In order of importance.

1. We need a bigger wedge. Current data only span ~55 degress. We should think about how to get 180. A higher symmetry sample would help a lot. This is absolutely critical.

2. Sample quality. The spot quality is poor, indicating a poorly ordered or bent crystal.

3. Getting more knowledge about the geometry of the system: a reference sample so we know the exact detector distance, ensuring the axis of rotation is perfectly aligned to the y-axis. Also we should pay attention to the stage motion -- it should be good to 0.2 degree I think, and more precision is better. Continue cutting the data at 0.2 degree wedges or finer for future work.

4. Image background. The current "hitfinding" data can probably work, but background is still high. This degrades our results.

Software Used (that is generally useful)

Install these!

• adxv : https://www.scripps.edu/tainer/arvai/adxv.html
• XDS : http://xds.mpimf-heidelberg.mpg.de/
• SHELX : http://shelx.uni-ac.gwdg.de/
• and the fabIO python package: https://pythonhosted.org/fabio/