SAM85 has come to refer to one of the High Pressure facilities at the National Synchrotron Light Source, at Brookhaven National Laboratory. It is located at beamline X17B2, and uses multianvil high-pressure devices with a sample volume in the 1 mm2 range.. The X17B2 beamline is split into two beams within the X17B2 radiation hutch: the main beam normally uses white radiation, and the side beam incorporates a single-bounce monochromator producing monochromatic radiation in the 30-70 KeV range. Changing the monochromatic energy requires adjusting the monochromator 2q and moving the endstation into the adjusted beam.
In principle, the mutianvil devices which have been used include the DIA, the Deformation DIA (DDIA), the Tcup, the T-10, one or more Paris-Edenburgh (PE) cells, and a Rotational Drikimer Apparatus (RDA). At this writing (October, 2005), only the DDIA is installed, and it is installed on the main beam. Next year (2006) we plan to install a PE cell on the monochromatic side station, with a Tcup installed there later. A T-10 (a somewhat more robust version of the Tcup) can be installed on the main beam if needed, but hasn't been requested recently. The original high pressure device in SAM85 was a DIA, and the name has propogated to include all the facilities at this beamline.
The large volume high pressure device known as SAM85 employs an assembly in which the ram force applied along one vertical axis is converted by means of the DIA device (developed by Kobe Steel, Japan, in the 1960's) to the six equivalent components of force which act on six faces of the cubic sample assembly along the three orthogonal axes. The DIA device consists of the upper and lower pyramidal guide blocks (bolsters) installed on the heads of the press rams, four trapezoid end blocks (thrust blocks) and six anvil holders, as indicated in Figure 1. The inner surfaces of the guide blocks form a tetragonal pyramid. Two of the six anvils are along the center line of this pyramid and are fixed opposite to each other on each guide block. The other four anvils are horizontally located on the midpoints of the square edges of a bipyramid. This results in the formation of a cubic nest bounded by the flat faces of six anvils. The ram force is applied along the vertical axis. The six equivalent components of forces act along the three orthogonal axes to force advancement of the six anvils toward the center of the cube. The advance of all the anvils is automatically synchronized so that the cubic sample/furnace assembly is compressed hydrostatically. SAM85 (Six Anvil Machine designed in 1985) is similar to the MAX80 and 90 apparatus (Multi Anvil X-ray Apparatus designed in 1980 and 1990, respectively) at the Photon Factory, Tsukuba, Japan.
Figure 1 DIA apparatus
Uniaxial force on steel guide blocks applies horizontal force on steel bolsters and attached anvils onto cell assembly.
The Deformation DIA (D-DIA) is similar to the regular DIA, except that the top and bottom anvils can be individually controlled. Each one has its own hydraulic system, enabling the state of stress on the sample to be modified.
The maximum pressure and temperature depend on the size of the anvil and its material; with 6 mm truncation WC anvils, the cubic anvil apparatus is currently capable of obtaining pressures to 8 GPa at 2000K. Higher pressures can be reached with 4 mm WC (11 GPa at 1800K) or sintered diamond (14 GPa at 1800K) anvils. Tapering the anvils also helps (see figure 2)
Figure 2 Tapered Anvils
In 1995, we designed and tested a new high pressure module for use in the same press as SAM85. It is called the Teacup (or Tcup), because of its approximate size. (It could also be called SAM95). The apparatus is a two-stage multi-anvil system similar to the now-common 6-8 systems; the first stage is a steel cylinder split into six parts enclosing a cubic cavity (19.5 mm edge length with the [111] axis of the cube vertical) which contains the second stage anvil assembly. The second stage is assembled outside the press and consists of eight 10 mm edge WC or polycrystalline diamond cubes, separated by spacers. Each cube has one corner truncated into a triangular face; the eight truncations form an octahedral cavity in which the pressure medium is compressed. The cubes in the Tcup and T-10 are both 10mm size, but the T-10 has a somewhat larger containing ring.
Figure 3 Tcup
As the x-ray beam from the synchrotron is relatively difficult to move, alignment is achieved by moving the apparatus with respect to the beam. With installation in the new X17B2 hutch, the system chosen here is to mount various portions of the apparatus separately. The press on a pedestal, which can translated in all three directions (X, Y, and Z), as well as a rotational adjustment about the vertical axis. This pedestal is mounted on the base, which can be leveled with three vertical translations. The base also has a horizontal translation in the Y direction (perpendicular to the beam) and a rotational adjustment about a vertical axis, but these are no longer used.
The beam defining apparatus is a set of incident slits and is mounted on a separate table.
The detector, detector slits, and camera are all mounted on a third, fixed table on individual stages.
Figure 4 outline drawing of system
See Appendix 2 for more details on alignment.
Because all of these adjustments must be made with the beam on and the hutch sealed, they are done remotely with 5 phase stepper motors, driven by OMS VME modules or, for some of the earliest motors, E500 CAMAC modules. These are controlled by a computer.
The entire system runs under EPICS and is controlled by one or more PCs running Windows XP. The communication between the PCs and the VME crate is via ethernet., A CAMAC crate is controlled by the VME crate. A second CAMAC crate for controlling beam line systems such as the beam defining apertures and the systems of other users who wish to use it is connected to another PC running LINUX. This system is located at the X17B1 control area. (see Figure A1. in Appendix A1).
The stepper motor controllers control 64 motors, as described in the figure. 32 are assigned to the main beam, and another 32 are used for the monochromatic side station. The pressure is measured with a digital Heise gauge. Two Linear Voltage Displacement Transducers (LVDTs) are used to remotely measure the advance of the lower ram and the spacing between the two anvils. All analog signals, except for the LVDTs, are measured using a Keithley 22 channel DVM.
Sample temperatures are made using one or two tungsten-rhenium thermocouples. Each thermocouple is connected to the Keithley mentioned above, wich is connected to VME through a RS232 serial port. The EMF is stored in the computer, and a polynomial curve determined from the calibration curve is applied to calculate the temperature. Although an electronic ice point compensator is available, it appears to cause more problems that it is worth. Instead, we simply set the zero point approximately to the hutch temperature (30° in this case).
Two regulated DC power supplies are available to supply heater current; which one is used depends on the heater resistance.
The analog signals from the LVDTs are connected to a Voltage-to-Frequency converter which is read by a binary VME card.
The remainder of the analog signals (thermocouple EMFs, pressure readings, voltage and current for the power supplies, etc.) are all read by the Keithley.
One or more germanium detectors are used to collect the diffracted x-rays. When the x-ray source is white, energy dispersive detection is used. The detectors drive 4 multichannel analyzers which store the diffracted x-rays in proportion to their energy, for each detector.
Usually we use a 4-element detector combined with a conical slit (see section 1.10). The MCA is a 4-channel digital signal processor manufactured by XIA. The XIA digital signal processor is a CAMAC module, so the data is transmitted via the CAMAC-VME interface.
One or two ionization chambers may be mounted in the x-ray path, to monitor the beam intensity. The first one is mounted between the slits and the sample; and the second one is used in place of the detector during alignment.
Currently, we are in the process of installing a monochromatic side station, with a second press and table assembly combined with a MAR 345 imaging plate detector. The MAR detector is controlled by a Linux computer.
A translating imaging plate assembly is used for most monochromatic experiments on the main beam. The imaging plate itself is mounted on a translation stage. It can be moved synchronously with increases in temperature, pressure, or simply at a constant rate, so the diffraction pattern as a function of these variables can be measured.
A Bruker CCD detector is also available for making 2-d diffraction measurements. It has faster readout than the IP, and is also used in conjunction with the IP for sample centering, pressure calibration, etc.
An optical system is in place for making radiographs of the sample. This uses two cameras, a high-resolution digital CCD (Spot), and an analog camera with lower resolution. The images from the CCD are stored on another PC, while the analog images can be recorded on a VCR.
Figure 4 Imaging system
A multi-element SSD detector can be used to measure diffraction in different directions. This requires use of a conical slit. Because the conical slit covers the direct beam, the YAG and mirror for the imaging system must be placed inside it (see figure 5).
Figure 5 Conical slit and Imaging system for use with multielement detector