DMM

2-BM has a double crystal multi-layer monochromator (DMM) to change energy. The beamline x-ray energy change is managed by the energy cli python library.

Login into 2bmb@arcturus then:

[2bmb@arcturus,42,~]$ bash
[2bmb@arcturus,42,~]$ energy set --mode Mono --energy-value 20

for help:

energy -h

More detailed instructions are here the energy cli

Technical information about the DMM are available at the links below:

Station	Description	Images	Info
2-BM-A	DMM	,	drawings1, drawings2, crystals specs, documentation folder

Substrate Specifications (Si <100>)

Parameter	Value
Material	Si <100> flat
Quantity	2 pieces
Dimensions	145.57 × 101.60 × 34.04 mm³ ± 0.25 mm
Optical Surface	140 × 92 mm²
Spherical Radius	> 20 km
Meridional Slope Error	1.0 µrad (rms)
Sagittal Slope Error	1.0 µrad (rms)
Microroughness	≤ 0.3 nm (rms) HSFR
HSFR Spatial Sampling	0.004–1 µm
Manufacturing Note	Grooves per ANL drawing “X2-230001-00” (28 Jan 1998)

Coating Specifications (W–B₄C Multilayer)

Parameter	Value
Coating Type	W–B₄C multilayer on Si substrate
Adhesion Layer	5 nm Cr
Stripe Dimension	140 × 44 mm²
Distance Between Stripes	4 mm
Multilayer Period	13.8 Å and 24 Å ± <1%
Interface Roughness	2–3 Å rms
Number of Layer Pairs	200 / 150
Gamma (Γ)	0.5

Stripe-Free Multilayer

This section collects all information related to the Stripe-Free Multilayer project.

Reference documents

The documents below describe the 2-BM beamline layout before and after the APS-U upgrade, and are used as the geometric reference for the Stripe-Free Multilayer design.

• A342-RT1000-00 LAY.pdf — pre-APS-U beamline layout and ray-tracing drawing.

• 02-BM Beamline Component Reference Table.docx — pre-APS-U component reference table listing the elements along the beamline and their distances from the source.

• APSU_FDR_Summary_2-BM.docx — APS-U Final Design Review summary for 2-BM. The new bending-magnet source is repositioned so that components sit 1822 mm farther from the source than in the original lattice. To keep the existing enclosures usable, the beamline centerline is rotated 1.35 mrad inboard around the new source and offset 42.295 mm inboard laterally relative to the original APS lattice BM centerline, accepting a 2.7 mrad fan (1.85 mrad through the front end).

Post APS-U component Z positions

Z is measured along the beam direction from the center of the straight section. Post APS-U values are obtained from the pre APS-U Z (component reference table) plus the +1822 mm source repositioning specified in the FDR summary.

Component	Post APS-U Z [mm]
Y3-30 Mirror	27626.2
DMM — first mirror	29335.2
DMM — second mirror	29934.2

Vertical intensity modulation

This section quantifies the vertical intensity modulation currently observed in the monochromatic beam delivered downstream of M1 + DMM. The modulation is the main motivation for the Stripe-Free Multilayer project: it shows up as horizontal bands in the projection images and as ring/streak artifacts in the reconstructed volumes.

Two distinct contributions can be separated by the choice of exposure time:

• A slow, stationary modulation from the M1 mirror (figure error / coating), best characterised with the long-exposure flats_01 dataset.

• A fast, moving stripe pattern from the DMM (residual W–B₄C stripe structure), best characterised with the short-exposure, high-rate S11-AHU505_1000frms_99fps_001.h5 stream, which freezes the motion.

Measurement setup

The flat fields used for this analysis are the white fields collected during the Flat Field Stability Measurement (flats_01, Feb 22, 2026), available via Globus (flats_01). File-naming convention: flat_2x_2bin3.45um_momo20keV_NNNN.tif (4710 frames in 471 sets of 10, one set every 60 s over ~8 hours).

The beamline (energy, M1, DMM) and the imaging chain (scintillator, objective, effective pixel size) match the Vibration Frequency Measurement used earlier; what differs in flats_01 is the detector ROI (2048 × 1536 px instead of 1024 × 1024), the exposure time (0.1 s instead of 0.009999 s), and the cadence/format (10 frames every 60 s saved as TIFFs, rather than a continuous 99 fps HDF5 stream). The values below reflect the flats_01 configuration.

Item	Value
X-ray energy	20.0 keV
Monochromator	2-BM-A double multilayer monochromator
DMM upstream arm angle (us_arm)	0.72579 ° (≈ 12.668 mrad)
DMM downstream arm angle (ds_arm)	0.73808 ° (≈ 12.882 mrad)
Mirror (M1)	2-BM Mirror, Pt stripe
M1 grazing-incidence angle	0.15 ° (≈ 2.618 mrad)
Scintillator	LuAG, 50 µm active thickness
Objective magnification / tube length	2.0× / 1.0 mm
Camera	FLIR Oryx ORX-10G-51S5M (s/n 19173710)
Camera pixel size (sensor)	3.45 µm
Effective image pixel size	3.45 µm (1.725 µm native × 2× binning)
ROI (X × Y) / binning	2048 × 1536 px / 2× binning
Field of view at detector (H × V)	≈ 7.07 × 5.30 mm
Exposure time	0.1 s
Acquisition cadence	10 frames / set, 1 set every 60 s, ~8 h total
Flat-field source files	flat_2x_2bin3.45um_momo20keV_NNNN.tif (4710 TIFFs in flats_01, Feb 22, 2026)
Detector Z from source (post APS-U)	≈ 54000 mm (54 m)
Distance DMM 2nd mirror → detector	≈ 22244 mm (54000 − 31756.2)

Quantitative metrics (mirror contribution, flats_01)

The following metrics are computed from a vertical line profile through a single flats_01 frame, averaged horizontally over the usable beam width. They characterise the mirror-induced modulation only; the DMM stripe contribution is averaged out by the 0.1 s exposure (see next subsection).

Metric	Value
Peak-to-valley modulation [%]	35.7
RMS modulation [%]	7.0
Dominant vertical period [µm]	≈ 130
Number of visible bands across beam	33 (in the ~3.5 mm analysis window)
Stability over time (drift) [%/h]	to be filled in

Modulation is defined as:

\[M_\mathrm{pv} = \frac{I_\mathrm{max} - I_\mathrm{min}}{I_\mathrm{max} + I_\mathrm{min}} \qquad M_\mathrm{rms} = \frac{\sigma_I}{\langle I \rangle}\]

computed on the flat-field profile after dark subtraction and normalization to the slowly varying envelope (low-pass filtered profile), so that only the high-frequency stripe contribution is retained.

Mirror-induced modulation (from flats_01)

The flats_01 exposures average out the fast DMM stripe motion (0.1 s exposure covers many cycles of the moving DMM pattern, see next subsection) and therefore reveal mainly the slowly varying, stationary modulation that originates at the M1 mirror. The ~130 µm vertical period quantified below is attributed to figure error on M1, not to the DMM coating.

Reference flat-field image from the flats_01 dataset (flat_2x_2bin3.45um_momo20keV_0001). The smooth horizontal banding visible here is the mirror contribution; the much faster DMM stripes are averaged out by the 0.1 s exposure.

Vertical line profile obtained from flat_2x_2bin3.45um_momo20keV_001.tif by averaging horizontally across the full 2048-pixel detector width. Top: raw profile (blue) with the low-pass envelope (red, 200 µm moving average) overlaid; the yellow band marks the analysis window where the beam intensity exceeds 50 % of its peak (~3.5 mm vertical extent). Bottom: residual after envelope removal — the high-frequency mirror contribution isolated by \((I - I_\mathrm{env}) / I_\mathrm{env}\). From this profile the peak-to-valley modulation is 35.7 % and the RMS modulation is 7.0 %, with a dominant vertical period of ≈ 130 µm (≈ 33 bright bands across the illuminated window).

DMM-induced stripes (from S11-AHU505_1000frms_99fps_001.h5)

The DMM contribution sits on top of the mirror-induced modulation and is best seen in the short-exposure, high-rate stream of the Vibration Frequency Measurement — file S11-AHU505_1000frms_99fps_001.h5 (1000 frames at 99 fps, 0.009999 s exposure), available via Globus (test_20251219_APS_PVs). Played back as a movie, the horizontal stripes shift vertically and rapidly frame-to-frame — a clear fingerprint of the DMM (the mirror pattern is static).

Single frame extracted from S11-AHU505_1000frms_99fps_001.h5 (one of the 1000 frames stored in the HDF5 file), representative of the DMM-induced horizontal stripes. Two stripe periodicities are visible: a fine spacing of ~83 px between adjacent bright stripes, and a coarser envelope with spacing of ~700 px between the strongest bands.

Two characteristic vertical spacings can be measured on the DMM stripe pattern (detector effective pixel size = 3.45 µm):

Feature	Distance [px]	Distance [µm]
Fine spacing (adjacent stripes)	≈ 83	≈ 286
Coarse spacing (envelope of strong bands)	≈ 700	≈ 2415 (≈ 2.4 mm)