The present invention will be better understood with reference to the accompanying drawing, in which:
Turning now to
As is known to those of skill in the art, a change in the force on the proof mass (3) causes a displacement in the position of the proof mass and a change in the length of the supporting quartz elastic element (4). The displacement of the proof mass (3) is sensed by the capacitive displacement sensor/bridge (5) to obtain a measure of the force acting on the proof mass (3). In practice, a restoring force (usually electrostatic) is applied to bring the proof mass (3) back to a standard position, to ensure linearity of measurement.
The molecular sieve material within the container (6) in conjunction with the reduced gas pressure within the chamber (1) ensure that the vapour content in the chamber (1) is maintained at a very low level. Contrary to the prior belief that instrumental drift was the result of an intrinsic vicoplastic creep of the amorphous quartz used to form the quartz elastic element (4), it has been found that the interaction of water vapour with the surface of the quartz accounts for significant instrumental drift. The interaction of the water vapour with the surface of the quartz causes certain changes to the quartz elastic element (4) and also increases the proof mass (3). These changes combine to elongate the quartz elastic element (4), and may simply be regarded as an effective reduction in its elastic coefficient or modulus. When the interaction of the water vapour with the surface of the quartz elastic element (4) continues, it creates a creep or drift.
To confirm that the interaction of water vapour with the quartz elastic element (4) accounts for significant drift in quartz elastic element gravimeters such as that described above, a series of experiments were performed as will now be described.
The first experiment performed was to determine the effect of water vapour on a quartz elastic element gravimeter. In this experiment, a quartz elastic element gravimeter such as the CG3 #256 instrument produced by Scintrex Limited of Ontario, Canada was used. The sealed metal chamber of the quartz elastic element gravimeter was firstly evacuated and then backfilled with commercial grade “dry nitrogen” at 150 torr pressure. The residual water vapour content of the “dry nitrogen” is not known, but is believed to be of the order of a few parts per million. The drift rate with that atmosphere, namely about 280 μGals/diem, was regarded as base level for this experiment. The sealed metal chamber was then evacuated to a pressure of less than 10−3 torr and exposed to a reservoir of water which was allowed to evaporate into the chamber to saturate its atmosphere. The vapour pressure of water in the chamber at 23° C. would then be 21 torr. The chamber was then backfilled to the original 150 torr pressure, with the same dry nitrogen. The quantity of water vapour was in the order of 14% of the gas in the chamber.
Table 1 below shows the change in the drift rate of the quartz elastic element gravimeter resulting from the addition of a small quantity of water vapour to the gas in the chamber, with the other constituents of the gas and the total vapour remaining unchanged.
As Table 1 shows, the observed drift rate of the instrument increased from 280 to 1250 μGals/diem when the quartz elastic element was exposed to the water vapour. The interaction of water vapour with the quartz elastic element clearly contributes to the drift rate of the quartz elastic element gravimeter.
Processes involving the interaction of water with the surface of fused silica have been well investigated and are discussed in such references as “The Chemistry of Silica” authored by R. K. Iler, John Wiley and Sons, 1979, and “Adsorption on Solids”, by V. Ponetz et al., London Butterworths, 1974. According to these sources the surface of silica is composed of siloxanes (SiOSi), which, at ordinary temperatures, have an affinity for and interact with the hydroxyl ions in the water vapour to form silanols (SiOH groups). The silicon surface is then said to be hydroxylated. When this process occurs with the quartz elastic element in a gravimeter, atoms are added to the surface of the quartz elastic element and therefore, its mass increases. This process is not limited to the apparent surface of the quartz elastic element. Hydroxl ions also penetrate below the surface of the quartz elastic element (i.e. are adsorbed), at a slower rate, into micropores or micro-capillaries in the body of the quartz elastic element. In addition to the increase in mass, the absorption process induces physical changes and possibly decreases the surface tension of the quartz elastic element. Both effects of the interaction of water vapour with the surface of the quartz elastic element result in an extension of the quartz elastic element i.e. an increase of apparent gravity. While the water vapour/quartz reaction continues, at a rate determined by the concentration of water vapour in the chamber and the availability of un-hydroxylated siloxanes, the effect is to create a long term positive drift in the gravity measurement.
The second experiment was to determine the effect reducing the gas pressure has on the drift rate of a quartz elastic element gravimeter. In this experiment, a CG3 #161 quartz elastic element gravimeter produced by Scintrex Limited used.
In this experiment, the drift rate of the quartz elastic element gravimeter was measured when the chamber of the quartz elastic element gravimeter was filled with commercial grade “dry nitrogen” at a pressure of 150 torr. A vacuum pump was then used to evacuate the chamber until the pressure in the chamber reached 1 torr. The drift rate of the quartz elastic element gravimeter was again measured.
Table 2 below shows the change in the drift rate of the quartz elastic element gravimeter resulting from the reduction in pressure of the gas in the chamber, the composition of the gas remaining the same.
As can be seen in Table 2, by decreasing the pressure in the chamber of the quartz elastic element gravimeter, drift is effectively eliminated. This clearly demonstrates that the drift depends on the continued availability of water vapour. The very small negative drift that remains is attributed to the removal of loosely bonded surface water being released under the low pressure. Tables 1 and 2 above confirm that, contrary to commonly accepted belief, the long term drift rate of quartz elastic element accelerometers, such as gravimeters, is not an intrinsic property of the quartz elastic element, but is predominantly determined by the presence and concentration of water vapour in the gaseous environment surrounding the quartz elastic element.
As Table 2 shows, one means of reducing the water vapour concentration in the chamber housing the quartz elastic element is to exhaust the gas in the chamber. Alternatively, a gas which is totally devoid of water vapour can be used. As will be appreciated, gas of this purity is difficult to obtain. Even if zero water vapour concentration can be established in the chamber at one particular time, steps must be taken to prevent small amounts of water vapor from entering the chamber after sealing. This requires very positive sealing, using metal-metal seals rather than permeable seals such as Viton. Even so, there may be outgassing from components of the gravimeter and, with time, water vapour may be introduced into the chamber resulting in instrumental drift.
Given these practical limitations to the long term reduction of the water vapour content of the gas in the chamber, one way of achieving effective and long term water vapour content reduction is to provide means to continuously remove water vapour from the chamber. Whereas this could be accomplished through the continuous use of a vacuum pump, this is a ponderous and inconvenient approach. Using a desiccant to remove water vapour is a much more practical approach. As is known, several basic types of desiccants exist, including silica gell, clay, carbon and molecular sieves. The latter offer the best advantage. Molecular sieves can selectively remove water vapour, even at very low levels, from a mixture of gases. Molecular sieves are crystalline metal aluminosilicates (or zeolites) having a three-dimensional interconnecting network of silica and alumina tetrahedra. Natural water of hydration is removed from this network by heating, to produce extremely uniform cavities which selectively adsorb molecules, up to a specific size, which are polar, and which are less efficient for any non-polar molecules or molecules of larger size.
Water is a polar molecule because of the way that the atoms bind in the molecule such that there is an excess of positive charges on the hydrogen side of the molecule and an excess of electrons on the oxygen side. Non-polar molecules have their electrons distributed more symmetrically about the molecule, so that there is no excess of charge at the opposite sides. Non-polar gases include nitrogen, oxygen and carbon dioxide, as well as all the noble gases such as helium, neon, krypton and xenon. Thus, a molecular sieve can selectively extract water vapour, even at very low concentrations, from an atmosphere with much higher concentrations of non-polar gases (e.g. nitrogen). Thus, providing that other constituents of the gas in contact with the quartz elastic element are non-polar (or, if polar, are of molecular sizes much larger than that of water), then an appropriate molecular sieve will be a highly selective and effective means for removing water vapour in the chamber.
Changing the ratio of Si/Al can affect the size of molecules that can be adsorbed, typically over the range from 3 to 30 Angstroms (0.3 to 3 nm). Water molecules have a diameter of 3.2 Angstroms, which determines the selection of a molecular sieve that is best suited to adsorb water vapour. For example, a commercially available molecular sieve such as Type 4A, produced by Sigma-Aldrich of Missouri, U.S.A. may be used. Molecular sieves have several other characteristics which are advantageous. For instance, they have a much higher equilibrium capacity for water vapour under very low humidity conditions, and are very effective in reducing the water vapour content of gases well below the part-per-million level. In addition, they can continue to adsorb water vapour at temperatures in excess of 150° C., although with diminished adsorption capacity above 40° C.
The third experiment was to determine the effect a molecular sieve desiccant placed in the chamber of a quartz elastic element gravimeter has on its drift rate. In this experiment, a CG3 #256 quartz elastic element gravimeter produced by Scintrex Limited was used.
In this experiment, the drift rate of the quartz elastic element gravimeter was measured when the chamber of the quartz elastic element gravimeter was filled with moist nitrogen at a pressure of 150 torr with a molecular sieve present in the chamber. An ADCOA Type 4-4 molecular sieve produced by Signman-Aldrich of Missouri, U.S.A. was used.
Initially, the chamber was firstly evacuated for five minutes. The molecular sieve, which was present in the chamber, was regenerated by this process. Then, as for Experiment 1, the chamber was exposed to a reservoir of water which was allowed to evaporate into the chamber, to saturate its atmosphere. The vapour pressure of water in the chamber (at 23° C.) would then be 21 torr. The chamber was then backfilled, to the original 150 torr pressure, with the same dry nitrogen.
Table 3 below shows the drift rate of the quartz elastic element gravimeter with the molecular sieve in the chamber compared to the quartz elastic element gravimeter of Experiment 1 when the chamber was filled with moist nitrogen at a pressure of 150 torr.
As can be seem from Table 3, the observed drift rate of the quartz elastic element gravimeter with the molecular sieve was zero. As will be appreciated with reference to Table 1, with the same quartz elastic element gravimeter, when the chamber was filled with moist nitrogen at a pressure of 150 torr and no molecular sieve was present, the drift rate was found to be 1250 μGals/diem.
Since nitrogen is a non-polar gas, and since the effect of the particular molecular sieve is for the selective removal of water vapour, two conclusions from the experiments reported in Table 3 can be drawn. Firstly, the results strongly reinforce the conclusion that the drift rate of quartz elastic element accelerometers is due to the presence of water vapour in contact with the quartz elastic element, and, secondly, that a properly selected desiccant, such as a molecular sieve, is highly effective in reducing the drift rate.
In the above experiments, reducing drift in quartz elastic elements gravimeters by reducing the water vapour content in the chamber housing the quartz elastic elements is achieved by using a desiccant or by evacuating the chamber. Those of skill in the art will however appreciate that other techniques to reduce the interaction of water vapour with the surface of the quartz elastic element can be used such as for example the technique disclosed in U.S. Pat. No. 3,459,522 to Thomas H. Elmer and Martin E. Nordberg. In this patent, a procedure for producing a substantially water-free silica surface is suggested by the sequential steps of preheating the silicon surface to a temperature of about 850° C., treating the silicon surface in a stream of dry chlorine gas at a temperature in the range of from about 600 to 1000° C., and then consolidating the silicon surface at a temperature of about 1250° C. Applying this procedure to the quartz elastic element will also reduce the rate of reaction of the quartz surface with water vapour in the chamber and thus, will reduce instrument drift rate.
The discovery that the drift of quartz elastic element accelerometers is not an intrinsic property of quartz elastic elements, but is due to the interaction of water vapour with the elastic element surface, provides the basis for devising an effective means of reducing instrumental drift. As discussed above, reducing water vapour pressure in the chamber, using a desiccant or treating the quartz elastic element surface can reduce instrumental drift. If desired, the above techniques can be used in combination to achieve the desired reduction in water vapour interaction with the quartz elastic element. Of course, other water vapour interaction reduction techniques may be used.
Although embodiments have been described above with reference to