The field of this invention is high-speed cylindrical ceramic sample spinners, especially for Nuclear Magnetic Resonance (NMR) Magic Angle Spinning (MAS) or for an ultra-centrifuge, employing hydrostatic gas bearings.
Because of the richness of the information available from NMR, it has often been argued that NMR is the most powerful analytical technique for molecular structure determination. However, NMR has been more successful with liquids or materials dissolved in solvents than with rigid solids. The basic problem in NMR of solids is that rapid molecular tumbling and diffusion are not naturally present to average out chemical shift anisotropy and dipolar couplings of abundant spin nuclides. Hence, the lines are normally broad and unresolved (often hundreds of ppm in width). A large number of techniques have been developed to improve the resolution in NMR of solids, but most modern techniques include extremely rapid spinning of the sample at the “Magic Angle” (the zero of the second Legendre polynomial, 54.7°) with respect to B0. If the rotational rate is fast compared to chemical shift anisotropies and dipolar couplings (in units of Hz), the resolution is dramatically improved—often by two or three orders of magnitude. Even when the spinning is not fast enough to satisfy the above conditions, substantial improvements in resolution are generally obtained from the combination of MAS and multiple-pulse methods.
Many MAS designs have been based somewhat on the classical work of J. W. Beams, and it is customary to refer to conical bearing/drives of the type detailed most clearly in U.S. Pat. No. 4,511,841 as Beams-type Bernoulli axial-bearing/drives. They comprise a rotor conical end (surface on the end of a rotor) of included angle ˜102°, a conical stator surface of included angle ˜91° engaging the rotor conical end, and a number of gas-feed holes through the conical stator surface at compound angles producing gas flow in the annular conical space between the stator and rotor surfaces that is characterized as rotational outward flow. Owing to the converging nature of this conical flow space, the radial component of the flow velocity may be sufficiently high at the periphery for a substantial Bernoulli effect, which, depending on various conditions, may exceed the hydrostatic effects nearer the center. As a result, a stable axial bearing may be formed over a rather wide range of spinning speeds, assuming sufficient space is available near the periphery of the conical surfaces for the gas to exhaust with very low back pressure. It is in fact this requirement which leads to the primary limitation of the utility of Beams-type drives in some important MAS applications, as, for example, when a gradient coil or dewar surrounding the spinner makes it impossible to adequately vent the Beams style bearing/drive.
High-speed NMR MAS spinners can be divided into two general classes: (1) designs that are inherently incompatible with automatic sample change because they require complex, high-precision disassembly/reassembly of the sample spinner for rotor changing, and (2) drop-in designs that are in principle compatible with simple automatic rotor changing. This invention belongs in the second of the above classes.
Some MAS designs in the first of the above categories include the following: In U.S. Pat. No. 4,254,373, Lippmaa discloses a double-ended drive design with no effective provision for either axial stability or high drive efficiency. In U.S. Pat. No. 4,456,882 I disclosed an MAS spinner with single-ended drive using cylindrical, ceramic sample containers with press-fit plastic turbines on hydrostatic air bearings that relies on carefully balanced back pressure from a front cover plate for axial stability against a point bearing. In U.S. Pat. No. 5,202,633 I disclosed a high temperature spinner with a hydrostatic axial bearing formed between the flat bottom end of the rotor and the inward flow exhausting from the radial bearing. Note that at the low flow velocities present here, there is no significant Bernoulli effect. In U.S. Pat. No. 5,508,615, I disclose a method of suppressing whirl instability in the radial bearings at very high surface speeds and improving the stability of balanced axial hydrostatic bearings, similar to the one used in U.S. Pat. No. 5,202,633. In U.S. Pat. No. 6,320,384 B1, an MAS spinner similar to that of U.S. Pat. No. 5,508,615 is used with novel methods of improving rf performance. In U.S. Pat. No. 6,803,764, Hioka discloses a design that incorporates many features of the above inventions. For example, it is worth noting that turbines with number of blades prime to the number of nozzles were in the Doty Scientific 4 mm production model XC4 in 1998, and those 4 mm rotors routinely spin at 25 kHz. The Doty Scientific HS5 production units in 1988 were utilizing blade profiles that resulted in a rotationally rearward velocity component in the turbine exhaust.
In U.S. Pat. No. 4,739,270, Daugaard and Langer (known to be the primary inventor here) disclose an outward-flow conical drive turbine at the top end of the rotor that at first glance seems compatible with automatic rotor change; but in practice it has not been, partially because the larger drive cap diameter complicates the flow requirements for the eject gas and partly because of the extreme sensitivity of this design to back pressure at the plug end (as would arise from a sample eject system) or below the drive turbine. It is worth noting that there is little Bernoulli effect in Langer's design. In U.S. Pat. No. 5,298,864, Muller discloses a laser-heated high-temperature spinner in which the axial bearing is formed on the back side of the drive turbine, again with outward flow. With regard to automation, it suffers the same deficiencies as Langer's design.
Some MAS designs in the second of the above two categories include the following: In U.S. Pat. No. 4,275,350, Hill discloses an attempt (which never succeeded) to achieve a spinner compatible with automatic sample change that includes a modified Beams-type conical drive surface. In U.S. Pat. No. 4,446,430, Stejskal discloses an outward-flow Bernoulli axial bearing formed on a flat end of the rotor and fed from a single, axial hole. In the aforementioned U.S. Pat. No. 4,511,841, Bartuska discloses a modified Beams-type drive; and in his later U.S. Pat. No. 4,940,942, he discloses a method of improving its axial stability and providing variable temperature operation for the sample. In U.S. Pat. No. 6,803,764, Hioka discloses a design of unclear novelty that might appear from
The axial air bearing of the instant invention utilizes inward flow with minimal rotational component over a rotor conical end to achieve improved stability and stiffness without the need for very-low-back-pressure venting. It is useable at surface speeds from zero to at least 80% of the speed of sound. Moreover, it is readily compatible with automatic sample change.
Non-patent literature
Patent Literature
An improved axial gas bearing for a gas-driven NMR MAS sample rotor is disclosed that utilizes inward flow with a low rotational component over a rotor conical end. A conical flow region is formed between the rotor conical end and a conical stator bearing surface such that the included angle defining the stator surface is not less than the included angle defining the rotor conical end. Gas is injected radially inward with a significant axial rearward component from a number of small holes at high velocity from the periphery into the conical flow region. Compared to the radial velocity components, the tangential flow components of the injected gas are small and preferably opposed to the direction of the rotor rotation. The high and accelerating negative radial velocities may result in significant Bernoulli effect, such that the mean static pressure over the conical rotor end may be less than atmospheric pressure for a given axial clearance, but as the clearance decreases, the hydrostatic effects exceed the Bernoulli effects and the mean axial pressure over the conical rotor end may then exceed atmospheric pressure by a substantial amount. Thus, a self-stabilizing axial bearing is formed with improved stability and stiffness for rotor surface speeds up to at least 80% of the speed of sound. Motive power required to spin the rotor may be provided by a radial-inflow microturbine at the opposite end of the rotor in a way that is readily compatible with automatic sample change.
The invention will be described with respect to a drawing in several figures.
The high-speed NMR sample spinner shown in
Note that the spinner axis is normally oriented at 54.7° with respect to vertical during operation. Hence, the right side or “top” of
The inflow-Bernoulli or “Doty” axial bearing is shown in more detail in
The axial nozzle 22 throat diameter d2 is typically greater than 50% of d1 and less than 1.8 times d1. The number of axial nozzles is typically not greater than the number of radial bearing orifices 11 at either end. One representative type of tip cap 19 is shown in
The axial nozzles 22 are normally supplied from the same manifold 14 as the radial bearing orifices, and the pressure in the bearing manifolds is normally maintained to be greater than the pressure in the drive manifold 15. The axial nozzles preferably are converging, and typically the exit diameter is about 15% less than the entrance diameter, for increased exit velocity and hence increased Bernoulli effect and greater tolerance of exhaust backpressure downstream from exhaust hole 23. The total axial nozzle 11 exit flow area, nπd22/4, where n is the number of axial nozzles, is typically less than one-third of the total radial bearing orifice flow area, mπd12/4, where m is the number of radial nozzles near each end.
The key to the bearing is obtaining converging flow of high radial velocity, and for this the stator axial bearing surface need not be a simple conical section but may have a convex curvature, as shown in
As is clear from
The differences between rotational inflow of the instant invention and the rotational outflow in the Beams drive/bearing are substantial. Any inward rotating flow is continuously accelerating, which results in a centrifugal pressure that adds to the static pressure and eventually may exceed the Bernoulli effect. For flow in a vaneless space with negligible friction, conservation of angular momentum requires rotational deceleration of outward flow and rotational acceleration of inward flow. Hence, the static pressure increase is much greater for inward flow than for outward flow for a given initial rotational rate. For inward flow, the static pressure increase from the centripetal acceleration is too great for a high-speed Bernoulli drive/bearing to work, while for outward flow, it often works well, as Beams, Andrew, Bartuska, etc. have shown. To permit the inflow Bernoulli bearing to work, it is necessary to avoid attempting to also use it for rotational drive. In fact, it is beneficial to inject the gas counter to the rotor's rotation to keep the mean rotational speed in the axial bearing space as low as practical. The increased drag on the rotor is of little consequence while the increased suction capacity of the bearing and the increased tolerance to bearing-exhaust back pressure are of considerable benefit in enhancing axial stiffness and stability.
The microturbine 17 may be of the self-shrouded radial inflow type comprising a plurality of radial flow passages and a central axial exhaust hole, according to the prior art. This microturbine type has been found to permit isentropic efficiencies above 35% in the larger NMR rotor sizes. Alternatively, an unshrouded radial inflow type comprising a plurality of radial blades, according to the prior art, may be used.
Although this invention has been described herein with reference to specific embodiments, it will be recognized that changes and modifications may be made without departing from the spirit of the present invention. For example, the cylindrical stator of
This application claims priority from U.S. application Ser. No. 60/522,621, filed Oct. 20, 2004, which application is incorporated herein by reference for all purposes.
Number | Name | Date | Kind |
---|---|---|---|
4201941 | Fyfe et al. | May 1980 | A |
4254373 | Lippmaa et al. | Mar 1981 | A |
4275350 | Hill et al. | Jun 1981 | A |
4446430 | Stejskal | May 1984 | A |
4456882 | Doty | Jun 1984 | A |
4511841 | Bartuska et al. | Apr 1985 | A |
4739270 | Daugaard et al. | Apr 1988 | A |
4940942 | Bartuska et al. | Jul 1990 | A |
5202633 | Doty et al. | Apr 1993 | A |
5298864 | Muller et al. | Mar 1994 | A |
5508615 | Doty et al. | Apr 1996 | A |
6320384 | Doty et al. | Nov 2001 | B1 |
6803764 | Hioka | Oct 2004 | B2 |
Number | Date | Country | |
---|---|---|---|
20060082371 A1 | Apr 2006 | US |
Number | Date | Country | |
---|---|---|---|
60522621 | Oct 2004 | US |