EP4645366A2

EP4645366A2 - Method for operating an ion pump

Info

Publication number: EP4645366A2
Application number: EP25203447.5A
Authority: EP
Inventors: Tiziano Isoardi; Paolo Manassero; Chiara Paolini; Riccardo GARZELLA; Marco Marzot; Andrea BERTALLOT
Original assignee: Agilent Technologies Inc
Current assignee: Agilent Technologies Inc
Priority date: 2025-09-19
Filing date: 2025-09-19
Publication date: 2025-11-05

Abstract

The present invention relates a method for operating an ion pump allowing to accurately estimate the pressure even when the ion pump is fed with a high voltage, as well to a pumping system for implementing such method. The method provides for measuring a current absorbed by the ion pump at a first, lower voltage at which no leakage current is expected, estimating a current at a second, higher voltage starting from said measured current at said first voltage, measuring a current at said second voltage and comparing said estimated current at said second voltage and said measured current at sid second voltage. As a result, the presence of a leakage current and its contribution to the overall measured current when the pump is running at the second, higher voltage can be identified and a correction factor to be applied to the measured current at said second voltage for correctly estimating the pressure in the pump can be calculated.

Description

Technical field of the invention

The present invention relates to a method for operating an ion pump.
More in detail, the present invention relates to a method for operating an ion pump allowing the pressure inside the ion pump to be more accurately estimated.
The present invention also relates to a pumping system including an ion pump and suitable for implementing the above method.

Prior art

A sputter ion pump is a capture vacuum pump which operates by sputtering a getter material. A sputter ion pump is able to create and maintain high-vacuum conditions within a confined environment. More particularly, ion pumps are commonly used in ultra-high vacuum (UHV) systems, because they can reach pressures lower than 10^-11 mbar.
Nowadays, the most common design of sputter ion pumps is based on the provision of one or more Penning cells, also known as Penning "traps".
The basic structure of a Penning trap includes a cathode and an anode received in a vacuum tight housing. The cathode includes a pair of plates which are arranged parallel to each other and at a distance from each other. The anode is made as a cell, for instance as a cylindrical cell with its longitudinal axis arranged perpendicular to cathode plates.
A pair of magnets are placed on opposite sides of the Penning trap, externally to the cathode plates and preferably outside the confined housing in which the cathode and the anode are received.
One of the electrodes is grounded, while the other electrode is fed with a high voltage (either positive or negative), so that an intense electric field is generated between the electrodes. More specifically, in a first possible configuration the cathode plates are grounded, while the anode cell is fed with positive high voltage, typically 3 - 7 kV, so that an intense electric field is generated between said electrodes (so called "diode configuration").
With reference to Figure 1a, according to said diode configuration, a vacuum tight housing 110 is arranged between the poles of a magnet 112 and a pair of cathode plates 114, 114' of getter material, for example titanium, are provided inside the housing 110; an anode 116 formed of a plurality of cylindrical hollow cells 118 is secured between the pair of cathode plates 114 and 114'. The two cathode plates 114, 114' are grounded, while the anode cells 118 of the anode 116 are connected to the positive pole of a potential difference source 120. This configuration is mainly used for pumping getterable gases: some gases, such as nitrogen (N₂), oxygen (O₂), carbon monoxide (CO) or carbon dioxide (CO₂) can form chemical compounds with getter materials, such as titanium, and are therefore called "getterable gases".
In a second possible configuration, the anode cell is grounded (and the walls of the vacuum tight housing are grounded as well), while the cathode plates are fed with negative high voltage, typically - 3 - - 7 kV, so that an intense electric field is generated between said electrodes (so called "triode configuration").
With reference to Figure 1b, according to said triode configuration, a vacuum tight housing 210 is arranged between the poles of a magnet 212 and a pair of cathode plates 214, 214' of getter material, for example titanium, are provided inside the housing 210; an anode 216 formed of a plurality of cylindrical hollow cells 218 is secured between the pair of cathode plates 214 and 214'. The anode cells 218 of the anode 216, as well as the walls of the housing 210 are grounded, while the cathode plates 214, 214' are connected to the negative pole of a potential difference source 220.
This configuration is mainly used for pumping inert gases, such as noble gases.
In both configurations, by applying such voltage between the electrodes of the ion pump, free electrons are produced and these free electrons are trapped in the Penning cells. More specifically, the electrical field traps the electrons in the axial direction, while the magnetic field generated by the magnets confines the electrons radially.
By trapping the electrons, the aim of the Penning cell is to increase as much as possible the path of these electrons from the cathode to the anode, hence increasing the probability of a collision between the electrons and gas molecules of the residual gas inside the pump. Because of such collisions, the trapped electrons can ionize the gas molecules, thus creating positive ions and extracted electrons: the ions, positively charged, are attracted towards the cathodes, while the extracted electrons are trapped as well and are available for further ionization of the other gas molecules.
This mechanism can be considered as the basic principle for the operation of Penning cells and therefore, for ion pumps.
With respect to other vacuum pumps, ion pumps provide remarkable advantages.
First of all, they are closed pumps as an ion pump keeps inside all the pumped gas, which avoids the risk of venting the system to which the ion pump is connected, prevents any contamination which could come from the roughing line and does not require a backing pump during steady operation.
Furthermore, an ion pump is a static device as it has no moving parts, which avoids both vibrations and the need for any lubricant (which could be a source of contamination).
In addition, ion pumps are often used as pressure indicators, thanks to the well-known almost linear dependence of pressure on the ion current, as long as the cell geometry, the number of cells, the voltage and the magnetic field are taken as fixed parameters.
More in detail, the relation between ion current and pressure in an ion pump can be expressed as follows: $I = k \times P$ where I is the ion current, P is the pressure and k is a parameter (so-called "sensitivity" of the pump) depending upon the geometrical dimensions and distances of the Penning cell, the number of Penning cells, the magnetic field, and the applied voltage.
Therefore, for a given ion pump, all geometrical and magnetic parameters being fixed, the only parameter that can change is the voltage applied. If the applied voltage changes, the sensitivity k changes.
With respect to this theoretical approach, however, it is to be noted that some spurious currents (so-called "leakage current"), independent from pressure, can arise in the ion pump. Unfortunately, at low pressures such leakage current can be comparable to or much higher than the pressure-dependent ion current.
The leakage current does not actually affect the pumping efficiency at all, but when this phenomenon occurs, the ion pump becomes poorly reliable in pressure reading because the pressure is proportional to the ion current only, while the measured current always results from the sum of the ion current and the leakage current (not depending form the pressure). The leakage current can arise from both external sources (such as power supply, connecting cables, and so on) and internal sources (e.g., the metallization of ceramic insulators).
However, the main cause of leakage current is the field electron emission (FEE) from cathodes, by which free electrons are emitted from the titanium surface when strong electric field is applied.
A simplified, and far from complete, theory of the field electron emission is reported here below.
Such theory was originally formulated by Fowler and Nordheim.
Let N(W_x)dW_x be the electron supply function or the number of electrons within a metal having energy between W_x and W_x+dW_x and D(W_x) be the electron transmission coefficient, or the probability that an electron with an energy W_x escapes the potential barrier.
Then, the total number of electrons tunneling through the barrier is: $J = e \int D (Wx) N (Wx) dW$
At room temperature, this results in the basic Fowler-Nordheim equation: $J = 1.54 \cdot 10^{- 6} E^{2} Φ^{- 1} \exp (- 6.83 \cdot 10^{9} Φ^{1.5} E^{- 1} f (y))$ where J is in A/m², E is the surface electric field in Vm^-1 and Φ is the work function of the emitting surface (e.g. 3.87 eV for titanium).
y is related to E and Φ by: $y = 3.79 \cdot 10^{- 5} E^{0.5} Φ^{- 1}$ and f (y) is an elliptical function.
When E ~ 10⁹ V/m it can be fitted by $f (y) = 0.956 - 1.062 y^{2}$
Theoretically, an external field of the order of magnitude of E ~ 10⁹ V/m is in fact needed to be able to observe FEE current.
In practice, FEE is observed at gap fields more than two orders of magnitude lower than that (in an ion pump with, for example, an electrode spacing of 7 mm and powered at 7 kV, E =10⁶ V/m).
It has been recognized that the microscopic surface of an electrode is far from being flat (this phenomenon is dramatically enhanced in an ion pump cathode).
Microprotrusion or "whiskers" present on the electrodes are responsible for the FEE current due to the high electric field that is generated locally at the tip.
The microscopic value of E at the tip of a protrusion is given by: $E = β {Vd}^{- 1}$ where V is the applied voltage, d is the electrode spacing and β is a "shape factor" depending on the height and width of the protrusion.
If the field acts over an emitting area A, the FEE current is given by $I = A \times J$
The presence of the leakage current can limit the use of an ion pump as a pressure indicator, especially in low pressure applications (lower than 10^-9 mbar), in which it is more likely to have a not negligible component of leakage current compared to the ion current-
Indeed, since the pressure inside the pump is a function of the ion current while the measured current is the overall current (i.e., the sum of ion current and leakage current) absorbed by the pump, an inaccurate calculation of the pressure is obtained.
In view of the above, an object of the present invention is to provide a method for operating an ion pump allowing to provide an accurate estimation of the pressure inside the pump, even when the leakage current is not negligible.
A further object of the invention is to provide a pumping system including an ion pump and suitable for implementing such method.
This and other objects are achieved by the method and pumping system as claimed in the appended claims.

Summary of the invention

As mentioned above, the current due to field electron emission ("FEE current") represents the most relevant contribution to leakage current.
From the above theoretical considerations, it follows that two important parameters govern the phenomenon of field electron emission: the applied voltage and the electrode spacing. More particularly, the applied voltage strongly affects the FEE current and decreasing the applied voltage is an effective way to decrease the FEE current, so that the FEE current may be negligible if the applied voltage is low enough.
The voltage value at which the FEE current becomes negligible might depend, for instance, from the geometry of the ion pump.
This means that an ion pump could be effectively used as a reliable pressure indicator, even at low pressures, provided that the applied voltage is kept low enough.
However, users are willing to run the ion pump at high voltages, especially at low pressures. Indeed, when the ion pump reaches very low pressure conditions there are not enough residual gas molecules to be ionized with subsequent generation of free electrons, so that a reduction of free electrons occurs. As a result, there is the risk that the electron plasma is switched off.
This risk can be reduced by supplying the pump electrodes with a higher voltage.
The method of the invention allows to estimate the contribution of the leakage current to the overall current absorbed by the pump when the pump is supplied with higher voltages, so that the actual ion current can be more accurately estimated and, consequently, the pressure can be accurately estimated.
The method according to the invention relies on two assumptions:

if the voltage applied to the ion pump electrodes is sufficiently low, the leakage current (or, at last, the leakage current due to field electron emission) is substantially negligible;
once the constructional and geometrical parameters of the pump are set, a theoretical relationship exists between the applied voltage and the ion current (i.e. between the applied voltage and the overall current in the absence of leakage current).

In this respect, it should be considered that the pressure in the ion pump is substantially independent from the applied voltage.
Therefore, it can be stated that $P = I_{1} / k_{1} = I_{2} / k_{2}$ and therefore $I_{2} = I_{1} \times (k_{2} / k_{1})$
where P is the pressure, I₁ is the current at a first voltage, k₁ is the sensitivity of the ion pump at said first voltage, I₂ is the current at a second voltage and k₂ is the sensitivity of the ion pump at said second voltage.
As mentioned above, for a given ion pump, once the geometrical and magnetic parameters are fixed, the sensitivity depends on the applied voltage only.
Accordingly, the method according to embodiments of the invention comprises the following steps:

applying a first voltage, said first voltage being selected low enough so that the corresponding leakage current is substantially negligible;
measuring the current at said first voltage (i.e. while said first voltage is being applied);
calculating an estimated current at a second voltage, different from the irst voltage and more particularly higher than the first voltage, based on the theoretical relationship between the current at said first voltage and the current at said second voltage;
applying said second voltage;
measuring the current at said second voltage (i.e. while said second voltage is being applied);
comparing the estimated current at said second voltage with the measured current at said second voltage.

As the first voltage is so low that the leakage current at said first voltage is substantially negligible, the measured current at said first voltage substantially coincides with the ion current at said first voltage.
In other words, for a given ion pump, a voltage threshold at which the leakage current is substantially negligible is set and the first voltage is selected equal to or lower than said threshold. Such threshold might depend, for instance, from the constructional parameters of the ion pump (such as the geometry of the ion pump) and/or from the operating parameters of the ion pump.
The estimated current at said second voltage takes into account the ion current only, as it is calculated on the basis of a theoretical relationship, while the measured current at said second voltage is the result of both the contribution of the ion current and of the leakage current, if any, at said second voltage.
Therefore, a leakage current at said second voltage can be estimated as the difference between the measured current at said second voltage and the estimated current at said second voltage.
According to embodiment of the invention, the method further includes the step of calculating the pressure in the ion pump according to the well-known linear relationship between the ion current and the pressure.
At the first voltage, the leakage current is substantially negligible and the measured current coincides with the ion current, so that the pressure can be correctly estimated starting from said measured current at said first voltage.
At the second voltage:

if the measured current at said second voltage substantially coincides with the estimated current at said second voltage, then no leakage current is affecting the ion pump at said second applied voltage, and the pressure can be calculated starting from the measured current at said second voltage, according to the well-known linear relationship;
if the measured current at said second voltage differs from the estimated current at said second voltage, then a leakage current is affecting the ion pump at said second applied voltage, so that a correction factor has to be applied to the measured current before calculating the pressure, said correction factor being derivable from the difference between the estimated current at said second voltage and the measured current at said second voltage. Preferably, the method according to the invention provides for the following steps:
keeping the ion pump running at said second, higher voltage on a predetermined time interval;
measuring the measured current at said second voltage, i.e. while said second, higher voltage is applied;
calculating the pressure inside the pump starting from the measured current at said second voltage and applying the obtained correction factor, if needed.

Thanks to the method of the invention, allowing to obtaining a correction factor taking into account the possibility that a non-negligeable leakage current exists, the ion pump can be operated at higher voltages and it still provides for an accurate estimation of the pressure inside the pump.
It is well known among those skilled in the art that leakage current in ion pumps generally increases as the time during which the pump is operated elapses.
Therefore, once the aforesaid predetermined time interval has lapsed, the first voltage can be applied again and the above-described steps can be repeated, so that it is possible to verify whether the leakage current at the second voltage is still accurately estimated and the corresponding correction factor is still correctly determined and, if needed, it is possible to correct the estimation of the leakage current at said second voltage and adjust and update the correction factor to be used in calculating the pressure during the subsequent time interval.
The length of the time interval can be set by the user.
Alternatively, the length of the time interval can be determined upon the operating conditions of the pump.
As a further alternative, the length of the time intervals can be adjusted depending on the accuracy of the estimation of the leakage current: if the leakage current at said second voltage turns out to be accurately estimated, the length of the time intervals could be increased, while if the leakage current at said second voltage turns out to be inaccurately estimated, the length of the time intervals should be decreased.
According to embodiments of the invention, the first voltage is about 3 kV.
According to embodiments of the invention, the second voltage is selected between about 5kV and about 7 kV.
According to further embodiments of the invention, at least one further, higher voltage can be set and, for said further voltage or for each of said further voltages, the method further provides the steps of:

calculating an estimated current at said further voltage based on the theoretical relationship between the ion current at the first, lower voltage and the ion current at said further voltage;
applying said further voltage;
measuring the current at said further voltage;
comparing the estimated current at said further voltage with the measured current at said further voltage;
estimating a leakage current at said further voltage as the difference between said measured current at said further voltage and said estimated current at said further voltage.

If the measured current at said further voltage differs from the estimated current at said further voltage, a correction factor can be determined and applied in order to accurately estimate the pressure in the ion pump starting from the estimated leakage current at said further voltage.
According to embodiments of the invention, the first voltage is about 3 kV, the second voltage is about 5kV and a further voltage is about 7 kV.
According to a preferred embodiment of the invention, if the measured current at said second voltage is different from the estimated current at said second voltage, then the measured current at said second voltage is inputted into an algorithm comparing said measured current with a theoretical model based on the equation of Fowler-Nordheim. Thanks to the comparison with such theoretical model following the Fowler-Nordheim equation, an indicator of the reliability of the estimation of the leakage current at said second voltage is provided.
In addition, information concerning the source of said leakage current can be obtained. More specifically, it can be ascertained whether the leakage current at said second voltage is due to filed electron emission or to other internal or external factors.

Brief description of the drawings

Further features and advantages of the present invention will become more evident from the following detailed description of an embodiment of the invention, which is given by way of non-limiting example with reference to the annexed drawings, in which:

Figure 1a schematically shows an ion pump with a known diode configuration;
Figure 1b schematically shows an ion pump with a known triode configuration;
Figure 2 shows exemplary curves representing FEE current versus voltage at different electrode spacing according to the theoretical model according to the Fowler Nordheim equation;
Figure 3 is a flow chart schematically showing the main steps of the method of the invention;
Figure 4 is a block diagram schematically showing a pumping system suitable for implementing the method of Figure 3.

Description of an Embodiment of the Invention

As mentioned above, the leakage current can arise from both external sources and internal sources. However, the main cause of leakage current is the field electron emission (FEE) from cathodes, by which free electrons are emitted from the cathode surface when strong electric field is applied.
The leakage current due to field electron emission, so-called "FEE current", is strongly dependent on the electrode spacing and, once the construction and geometry of an ion pump is set, on the applied voltage.
With reference to Figure 2, a set of curves is shown representing the FEE current versus voltage at three different electrode spacing.
In Figure 2, the FEE current is determined by the model based on the Fowler Nordheim equation as $I = A \times J$ where A is the emitting area in which the field acts and $J = 1.54 \cdot 10^{- 6} E^{2} Φ^{- 1} \exp (- 6.83 \cdot 10^{9} Φ^{1.5} E^{- 1} f (y))$ where E is the surface electric field, Φ is the work function of the emitting surface, y= 3.79.10^-5 E^0.5Φ ^-1 and f(y) is an elliptical function.
In the above equation, $E = β {Vd}^{- 1}$ where V is the applied voltage, d is the electrode spacing and β is a "shape factor" depending on the height and width of the protrusion on the cathode surface.
With reference to the curves shown in Figure 2, A = 2.9.10^-14 m² and β = 2000.
It is worth mentioning that a shape factor β = 2000 represents a quite dramatic situation, with a surface having a strong unevenness.
The curves of Figure 2 clearly show the strong dependence of the FEE current from the applied voltage, irrespectively of the selected electrode spacing.
In fact, starting from the actual set of values here discussed and considering an electrode spacing of 6 mm, the FEE current for an applied voltage 7 kV is 1.2.10^-3 A; a reduction of the applied voltage to 5 kV (at the same electrode spacing) gives a FEE current of 1.4·10^-5 A; a further reduction of the applied voltage to 3 kV gives 0.93·10^-9 A, i.e. a reduction of more than four order to magnitudes when compared to an applied voltage of 5 kV and of more than six order of magnitudes when compared to an applied voltage of 7 kV.
The same applies with different values of electrode spacing.
As a result, when a voltage of 3 kV is applied, the FEE current is absolutely negligible (even considering a non-favorable surface structure of the electrodes), while it becomes more and more relevant as the applied voltage increases.
Therefore, by applying a sufficiently low voltage - e.g., about 3 kV or below such value - the leakage current (or at least its main component) is negligible and the measured overall current substantially coincides with the ion current, i.e. with the pressure-dependent current. This means that the ion pump can be reliably used as a pressure indicator provided that a sufficiently low voltage is applied.
However, it is disadvantageous for the user to operate the ion pump at low voltage, especially if low pressure conditions are reached.
Accordingly, the method of the invention allows to reliably estimate the pressure in the ion pump even if the higher voltage is applied.
With reference to Figure 3, to this purpose the method according to a preferred embodiment of the invention essentially comprises the following steps:

at a given time t = t* (step 10), applying a first voltage, lower voltage between the electrodes of the ion pump (step 15); said first voltage is chosen sufficiently low so that the leakage current is substantially negligible; for instance, said first voltage can be about 3 kV;
measuring the current at said first voltage, i.e. while said first voltage is being applied (step 20); since the leakage current at the first voltage is negligible, the measured current at said first voltage entirely consists of ion current (i.e. pressure-dependent current);
calculating the pressure inside the ion pump starting from said measured current at said first voltage (step 25);
calculating an estimated current at a second, higher voltage based on the theoretical relationship between the current at said first voltage and the current at said second voltage (step 30); for instance, said second voltage can be about 5 kV or about 7kV;
applying said second voltage between the electrodes of the ion pump (step 35);
measuring the current at said second voltage, i.e. while said second voltage is being applied (step 40);
comparing said estimated current at said second voltage with said measured current at said second voltage (step 45) and estimating a leakage current at said second voltage as the difference between said measured current at said second voltage and said estimated current at said second voltage;
if the measured current at said second voltage is equal to the estimated current at said second voltage, calculating the pressure inside the ion pump starting from said measured current at said second voltage (step 50);
if the measured current at said second voltage is different from the estimated current at said second voltage, calculating a correction factor for taking into account the leakage current at said second voltage (step 55) and calculating the pressure inside the ion pump starting from said measured current at said second voltage and said correction factor (step 60).

The ion pump is kept running at said second, higher voltage for a predetermined time interval Δt.
During said time interval, the current is measured (step 65) and the pressure inside the ion pump is calculated starting either from said measured current at said second voltage or from said measured current at said second voltage and said correction factor (step 70).
Thanks to the method of the invention, during the time interval Δt the ion pump can be operated at higher voltages and it still provides for an accurate estimation of the pressure inside the pump.
Once the time interval Δt has lapsed, the first voltage is applied again and the steps 15 - 60 are repeated, so as to verify whether the leakage current at said second voltage is still accurately estimated and, in case, adjust and update the correction factor to be applied. The length of the time interval Δt can be set by the user.
Alternatively, the length of the time interval Δt can be determined upon the operating conditions of the pump.
By way of example, the length of the time interval can be longer at the beginning of the pump life, when the cathode surface is smoother, and it can be shortened as the wear of the pump increases, i.e. as the cathode surface becomes rougher and the presence of unevenness on the cathode surface is likely to generate a higher leakage current due to field electron emission.
As a further alternative, the length of the time interval can be iteratively determined depending on the accuracy of the estimation of the leakage current at said second voltage. If the correction factor calculated at the end of the time interval is substantially the same as the previously calculated correction factor, the length of the time interval could be increased; conversely, if the correction factor calculated at the end of the time interval is sensibly different from the previously calculated correction factor, the length of the time intervals should be decreased.
It is evident that at each iteration of the method of the invention, one or more further higher voltages can be applied, and for each further voltage a corresponding leakage current can be estimated and a corresponding correction factor can be calculated.
According to a preferred embodiment of the invention, if the measured current at said second voltage differs from the estimated current at said second voltage, then the measured current at said second voltage is compared with a theoretical model based on the equation of Fowler-Nordheim (not shown in the flow chart of Figure 3).
Thanks to the comparison with such theoretical model following the Fowler-Nordheim equation, it is possible to ascertain whether the calculated correction factor is consistent with FEE theory.
In this way, an indicator of the reliability of the calculated correction factor and, consequently of the estimation of the pressure in the ion pump can be obtained and provided to the user.
In addition, information concerning the source of the leakage current at the second voltage can be obtained. More specifically, it can be ascertained whether said leakage current at said second voltage is due to field electron emission or to other internal or external factors. The method according to a preferred embodiment of the invention can be implemented in a pumping system 1 as schematically shown in Figure 4.
The pumping system 1 includes a ion pump 3 having the general structure shown in Figure 1a or in Figure 1b.
The pumping system further comprises a potential difference source 5 for applying a voltage between the electrodes of the ion pump 3.
The pumping system 1 further includes one or more sensors 7 for measuring the current absorbed by the ion pump 3.
The pumping system 1 further includes a controller 9 for controlling the operation of the ion pump 3.
The controller 9 is suitable for controlling the potential difference source 5 for applying a selected voltage to the electrodes of the ion pump 3.
In addition, the controller 9 is equipped with a memory unit 11 in which the theoretical relationship between the ion current and the applied voltage for the ion pump 3 (i.e. the geometrical and magnetic parameters of the ion pump being fixed) are stored.
Preferably, a theoretical model of the FEE current in the ion pump 3 according to the Fowler Nordheim equation is also stored in the memory unit 11.
The controller 9 is further equipped with a processing unit 13 which is adapted to calculate the estimated currents at different voltages starting from the theoretical relationship stored in the memory unit 11, receive the values of the measured currents coming from the sensor(s) 7, compare the estimated current with the measured current at the applied voltage, calculate a correction factor based on the difference between the estimated current and the measured current (if any), and calculate the pressure starting from the measured current and the correction factor (if applicable).
Preferably, the processing unit 13 also includes an algorithm for comparing the measured current with the theoretical model of the FEE current stored in the memory unit 11 and providing an indicator of the reliability of the calculated correction factor based on such comparison.
It will be evident to the person skilled in the art that the embodiment described above in detail should in no way be understood in a limiting sense, and that that numerous modifications and variants are possible without thereby departing from the scope of protection as defined by the appended claims.

Claims

Method for operating an ion pump immersed in a magnetic field and comprising at least one anode and at least one cathode and being configured for applying a voltage between said at least one anode and said at least one cathode, said method comprising the steps of:
a) applying said first voltage between said at least one anode and said at least one cathode (step 15)

b) measuring the current at said first voltage (step 20);

c) calculating an estimated current at a second voltage higher than said first voltage (step 30), based on a theoretical relationship between current at said first voltage and current at said second voltage in said ion pump;

d) applying said second voltage between said at least one anode and said at least one cathode (step 35);

e) measuring the current at said second voltage (step 40);

f) comparing the estimated current at said second voltage with the measured current at said second voltage (step 45) and estimating a leakage current at said second voltage.
Method according to claim 1, wherein said first voltage is chosen sufficiently low so that the leakage current at said first voltage is substantially negligible.
Method according to claim 1 or 2, further comprising the following step:
- while said first voltage is applied, calculating the pressure inside said ion pump starting from said measured current at said first voltage (step 25).
Method according to any of claims 1 - 3, further comprising the following steps:
- while said second voltage is applied:
- if said measured current at said second voltage is equal to said estimated current at said second voltage, calculating the pressure inside the ion pump starting from said measured current at said second voltage (step 50);

- if said measured current at said second voltage is different from said estimated current at said second voltage, calculating a correction factor taking into account said estimated leakage current at said second voltage (step 55) and calculating the pressure inside said ion pump starting from said measured current at said second voltage and said correction factor (step 60).
Method according to any of the preceding claims, wherein said ion pump is kept running at said second, higher voltage for a predetermined time interval (Δt), and wherein during said time interval, the current at said second voltage is measured (step 65) and the pressure inside the ion pump is calculated starting either from said measured current at said second voltage or from said measured current at said second voltage and said correction factor (step 70).
Method according to claim 5, wherein, once said predetermined time interval (Δt) has lapsed, steps a) - f) are repeated.
Method according to claim 6, wherein
- if the difference between said measured current at said second voltage and said estimated current at said second voltage is the same as the previously calculated difference between said measured current at said second voltage and said estimated current at said second voltage, the correction factor for calculating the pressure inside said ion pump starting from said measured current at said second voltage is maintained;

- if the difference between said measured current at said second voltage and said estimated current at said second voltage is different from the previously calculated difference between said measured current at said second voltage and said estimated current at said second voltage, a new correction factor for calculating the pressure inside said ion pump starting from said measured current at said second voltage is calculated.
Method according to claim 5 or 6 or 7, wherein the length of said time interval (Δt) is settable by the user and/or is adjustable upon the operating conditions of said ion pump and/or is adjustable upon the results of the comparison between said estimated current at said second voltage and said measured current at said second voltage.
Method according to any of the preceding claims, wherein, if said measured current at said second voltage differs from said estimated current at said second voltage, said measured current at said second voltage is compared with a theoretical model based on the equation of Fowler-Nordheim.
Method according to claim 4, wherein, if said measured current at said second voltage differs from said estimated current at said second voltage, said measured current at said second voltage is compared with a theoretical model based on the equation of Fowler-Nordheim and an indicator related to the consistency of said correction factor with said theoretical model is provided.
Method according to any of the preceding claims, wherein said first voltage is equal to or lower than about 3 kV.
Method according to any of the preceding claims, wherein said second voltage is equal to about 5 kV or equal to about 7 kV or higher than 7 kV.
Pumping system (1) comprising an ion pump (3) immersed in a magnetic field and comprising at least one anode and at least one cathode, a potential difference source (5) for applying a voltage said at least one anode and said at least one cathode of said ion pump, one or more sensors (7) for measuring the current absorbed by said ion pump and a controller (9) for controlling the operation said ion pump (3),
wherein said controller (9) controls said potential difference source (5) for applying a selected voltage,

wherein said controller is equipped with a memory unit (11) in which theoretical relationships between currents at different applied voltages for said ion pump (3) are stored, and wherein said controller (9) is further equipped with a processing unit (13) which is adapted to calculate the estimated currents at different voltages starting from the theoretical relationship stored in said memory unit (11), receive the values of the measured currents coming from said one or more sensor(s) (7), and compare the estimated current with the measured current at the applied voltage.
Pumping system (1) according to claim 13, wherein said processing unit (13) is further adapted to calculate a correction factor starting from the difference between said estimated current and said measured current at the applied voltage, and calculate the pressure in said ion pump starting from the values of the measured currents coming from said one or more sensor(s) (7) and from said correction factor.
Pumping system (1) according to claim 14, wherein a theoretical model of the current due to field electron emission in said ion pump (3) according to the Fowler Nordheim equation is also stored in said memory unit (11), and wherein said processing unit (13) is provided with an algorithm for comparing said measured current at the applied voltage with said theoretical model and providing an indicator of the consistency of said calculated correction factor with said theoretical model.