WO2012035165A1

WO2012035165A1 - Dynamic trap ion mobility spectrometer

Info

Publication number: WO2012035165A1
Application number: PCT/EP2011/066229
Authority: WO
Inventors: Alexei Boulbitch; Andreas Diewald
Original assignee: Iee International Electronics & Engineering S.A.
Priority date: 2010-09-17
Filing date: 2011-09-19
Publication date: 2012-03-22
Also published as: EP2616804A1

Abstract

A dynamic ion trap comprises a chamber that includes a series of electrodes partitioning the chamber into a succession of sections, each section being fluidly connected to its nearest neighbour sections, and a control unit operatively connected to the electrodes to locally create an electric field across at least two neighbouring sections and to displace the electric field in a stepped manner along the succession of sections.

Description

DYNAMIC TRAP ION MOBILITY SPECTROMETER

Technical field

[0001 ] The invention is in the area of gas sensing devices. More precisely, this device is able to separate volatile organic compound molecules in the gas phase according to their mobilities.

Background Art

[0002] The necessity to sense volatile organic compounds in air is related to multiple tasks. In the area of security it may be related to the task of detection of explosives as well as detection of drugs in airports and other public places. It may be required in military applications for detection of toxic substances in the field. It may be related to the improvement of the quality of life by detecting harmful gases in streets, in public buildings, in private houses and in cars or in aircraft cabins . It may also become a part of technological processes in food production and may be used in diagnostic of certain diseases. All these tasks require the use of gas sensors that would be (i) reliable, (ii) fast and (iii) cheap. From the point of view of timing some applications admit measurement times of few tens of minutes, but most of those enumerated above require measurement times of a few seconds at maximum. From the point of view of a price of exchangeable parts, it should be on the level of few Euros.

[0003] Finally, it is desirable to have a programmable sensor. That is, in case of emergence of a new volatile compound to be detected one does not need to make a completely new sensor, but only needs to include new data into the computer library, still using the same device.

[0004] At present there are two main devices on the market solving such a problem and having the property of being programmable. These are the Ion Mobility Spectrometer (usually referred to as IMS) and Differential Mobility Spectrometer (DMS). The latter device and principle has no standard name, and about a dozen of names are met in literature. Here it is worth noting also FAIMS (Field Asymmetric Ion Mobility Spectrometer) as second most often met name. [0005] The IMS method is based on the measurements of the time of flight of ions subjected to a constant electric field moving in a tube through the air at room conditions. The mobility of the ion in air is considered as its finger print, and is directly related to the time of flight. Typically the IMS devices are large, since the longer is the time of flight, the more precisely the mobility may be determined. However, handheld devices are at present on the market, with the tube as long as several centimetres [1 ].

[0006] The second device is the DMS (sometimes called FAIMS). In contrast to IMS this one measures the so-called, ion mobility increment. This method is based on the following idea: the ion passes between two electrodes to which a periodic voltage is applied at high frequency. This signal is in addition asymmetric. Due to a non-linear dependence of the mobility upon electric field the ions drift to one of the electrodes. Application of an additional slowly varying electric field enables one to separate one ion species from all the others [2]. The non-linear dependence of the ion mobility upon the field is quite different from the ion mobility itself and therefore, DMS is complementary to the IMS: they measure different properties and one of them can resolve molecules that the other fails to separate.

[0007] One can also mention the so-called, aspiration spectrometer, representing the version of IMS with a more sophisticated configuration [3].

[0008] One could expect to drastically reduce the price of these sensors by going to micromachining and producing inexpensive sensors on chip. However, this proved not to be possible with either IMS of DMS, since certain device size is necessary for its functioning.

[0009] It is the length of the tube in the case of the IMS device that must exceed a certain value, since the time of flight is related to the drift distance.

[0010] In the case of the aspiration mobility spectrometer, although designed to be miniaturized, there is still a geometrical dimension that plays the role of the length of the drift tube.

[001 1 ] In the case of DMS it is the length of the electrodes, since the device resolution is related to this length, as well as the minimal inter-electrode distance, since the ions are moving back-and-forth between the electrodes should not hit them not to be neutralized. Technical problem

[0012] It is an object of the present invention to provide a microscopic gas sensing device based on the measurement of the ion mobility.

General Description of the Invention

[0013] This object is achieved by a dynamic ion trap device according to claim 1.

[0014] A dynamic ion trap device according to the invention comprises a filtering chamber that includes a series of electrode structures partitioning the filtering chamber into a succession of sections, each section being fluidly connected to its nearest neighbour sections, and a control unit operatively connected to the electrode structures and configured to perform a filtering cycle by which an electric field is locally created across at least two neighbouring sections and the electric filed is displaced in a stepped manner along the succession of sections.

[0015] Similarly to Ion mobility Spectrometry (IMS), the present invention exploits the mobility of a given ionized molecule or analyte in a carrier gas such as (but not limited to) air, and under an external electrical field. This mobility serves as a "fingerprint" of the molecule to be filtered/detected and is based on the relationship:

v = KxE

where v is the ion velocity, E the electric field intensity and K is the ion mobility in the carrier gas.

[0016] A merit of the present invention is the implementation of a moving trap for filtering of ions, e.g. of organic volatile compounds, moving through air, wherein a static electric field is applied for a certain time (hereinafter referred to as duty time), on two electrode structures constituting a fraction of the filter, followed by its stepwise propagation to the neighboring electrodes structures thus, moving along the filter. Adavantageously, electric field parameters,. e.g. the voltage and the duty time, are configured to enable only a single sort of ions to move together with the trap, while filtering out the other ion sorts by their segregation outside of the moving trap followed by neutralization of the segregated ions on electrodes and evacuation of the corresponding molecules.

[0017] Accordingly, the control unit is advantageously configured to displace the local electric field along a filtering path at a pre-defined pace in accordance with the ion analyte to be detected/filtered (respectively in accordance with its mobility). In practice, the voltage is applied between a couple of electrode structures in the chainso as to create an electric field across two neighbouring sections (or possibly more neighbouring sections). The voltage is applied for a given duration the "duty time". Initially, the local field is created at the beginning of the filtering chamber by applying a voltage during a duty time on a pair of electrode structures covering at least the first two sections of the filtering path. At the expiration of the duty time, the electrical field is moved forward in the propagation direction along the filtering path, by applying the voltage on the next, nearest electrode structure in the chain, following each of the previously active electrode structures. Each time, the voltage is thus switched to a next electrode structure pair, and applied during a duty time, preferably substantially identical. And this respective energizing of electrode structure pairs is repeated along the chain of electrodes until the electrical field arrives at the end of the filtering path.

[0018] It will be appreciated that the duty time and the voltage across the electrode structures (to create the electrical field) are two electric field parameters that can be readily controlled in the present device in such a way that the ion analytes of interest be able to follow the propagation of the trap along the filtering chamber, and so that the ions of interest remain within the trap at the end of the filtering cycle and can be delivered to a detector chamber. In practice, the control unit is advantageously configured to operate with a duty time τ and voltage U that are appropriate for filtering ions having the desired mobility, K. Indeed, the relationship between the duty time, τ, voltage, U, and the mobility K can be approximated as follows: K D²/xU where D is the inter-electrode distance.

[0019] Let us now illustrate the jumping of electrode field in the filtering chamber. Suppose, for the sake of exemplification, that the chamber contains N electrode structures arranged in series along the filtering path, the electric field is typically applied to extend across two neighboring sections, which means that the voltage is applied between electrode structure i and the next nearest electrode structure i+2 (since one electrode structure is present between two sections). At the expiry of the duty time, the electrical field is moved forward in the desired propagation direction of the ion cloud by applying the voltage between electrodes directly following the previously active electrodes, i.e. i+1 and i+3. Preferably, a similar electrical field (same length and strength) is created every time between the respective pair of electrode structures during a measurement cycle.

[0020] Preferably, neighbouring sections are separated from one another by a single electrode structure . In such case, supposing that the electrode structures are regularly distributed along the progression direction, the duty time (τ) during which the electric field is applied between a pair of electrode structures may be calculated as follows:

x nL²/K_eU

[0021 ] where K_e is the mobility of the ions that are able to pass the filter exhibiting given parameters, U, L and τ. Further, U is the voltage across the two active electrode structures, L is the distance between two nearest neighbour electrode structures and n the number of sections over which the electric field extends. When the electric field extends over two sections, n=2. Hence, Fixing U, L and τ, thus, determines the value of mobility of molecules able to pass through the filter. The corresponding mobility is, therefore, referred to as "targeted mobility" (i.e. the mobility that it is intended to measure).

[0022] Hence, ions having the targeted mobility K_e will follow the electric field. In contrast, ions having a higher mobility outpace and collide with the electrode structures. They are neutralized and fall out of from consideration. They are further carried away by the air flow. The ions with mobility smaller than K_e file over and finally find themselves behind the repruslive electrodes. They are then separated and later collide with a trailing, active electrode structure.

[0023] Preferably, the control unit is configured is to repeat the filtering cycle with different electric field parameters to detect ions with different mobilities. Indeed, voltage and/or duty time can be readily varied to adapt to be able to filter a variety of ions with different mobilites. In one practical embodiment, the filtering cycle may be repeated by scanning voltages over a pre-defined range corresponding to a range of mobilities. [0024] In one embodiment, each electrode structure comprises at least two elongate electrode members, parallel to one another. The electrode members may have variety of shapes, in particular the electrode may be shaped as rods of circular of quadrilateral cross-section. Alternatively, the electrode members may take the form of conductive pads or strips.

[0025] Within one electrode structure, the various electrode members may be electrically independent and hence individually connected to the control unit. Alternatively a bridging element may electrically connect together the electrode members of a same electrode structure, so that a single connection from the electrode structure to the control unit is sufficient. When using such bridging elements, they are preferably located outside the chamber so as to prevent direct collision with ions. This being said, even when the electrode structures comprise electrode members that are independently connected the control unit, the latter is preferably configured so that all electrode members belonging to a given electrode structure are operated jointly, i.e. they are switched on or off simultaneously and a same voltage is applied.

[0026] According to a further design, each electrode structure comprises a ring- shaped electrode member. Such electrodes, although more expensive, permit reducing fringe field issues.

[0027] In the various possible embodiments, the electrode members are preferably made from a metal such as copper, copper alloy or more noble conductive metals.

[0028] Electrode structures preferably extend transversally to the progression direction (filtering path) of the ion cloud in the filtering chamber, so as to normally correspond to a given position along the filtering path.

[0029] In the case of elongate or rod shaped electrode members, fringe fields due to the truncation of this electrode members may be reduced by applying low conductive layers on the chamber walls at the longitudinal extremities of the electrode members.

[0030] It may further be noted that the current in the filtering chamber may be increased by using electrode structures with a plurality of parallel elongate electrode members, which results in an enhanced field homogeneity towards the center of the filtering chamber. [0031 ] In practice, the dynamic ion trap device according to the invention hence comprises a series of electrode structures distributed along a filtering path for the ions to be detected, and where two consecutive electrode structures (nearest neighbourgs) are separated by a section of filtering path. The voltage to create the electrical field is preferably applied between a given electrode structure and a next nearest neighbour electrode structure, so that the electric field spans over two neighboring sections and there is one electrode structure in between the "active" electrodes (the electrodes structures to which the voltage is applied). However, one may design the control unit so that more than one electrode structure is located in- between the pair of active electrode structures, whereby the electrical field will extend over more than two sections.

[0032] During a detection/measurement cycle, the duty time is preferably kept constant, to keep a constant field distribution. Alternatively, the duty time may be variable during a measurement cycle, in particular at the beginning of the measurement cycle to initially carry a greater number of ions that may have unlucky initial conditions.

[0033] Auxiliary electrodes (or compressing electrodes) may be used in order to help keeping the ion cloud more compressed towards the filtering chamber centreline, and hence prevent the ion cloud from colliding too fast with attractive electrodes. These electrodes produce a repulsive electric field that tends to compress the ion cloud.

[0034] Besides, ion clouds may be confined in the filtering chamber by magnetic fields created along electrode pathways by either permanent or electro-magnets. For such embodiment, one may e.g. consider the following relationship:

m K U

E >

e H L

[0035] which when fulfilled, allows that the ion trajectories are confined to the filetring chamber by the magnetic field B. Here m is the ion mass, K is its mobility, U is the voltage across the electrodes structures, e is the elementary charge, H is the smallest among the width and the height of the pathway between the electrode members, L is the distance between the electrode structures. The above equation can be fulfilled for microscopic range of the device sizes. [0036] Preferably, the dynamic ion trap device comprises an enrichment chamber upstream of the filtering chamber.

[0037] For increased processing speed, the dynamic ion trap device may comprise a plurality of filtering chambers arranged in parallel for independent processing, whereby each channel is configured for detection of a distinct single analyte.

[0038] The invention may be used in an ion mobility spectrometer comprising a dynamic ion trap, an inlet gate for introducing ions into the dynamic ion trap, an outlet gate for evacuating ions from the dynamic ion trap and a detection chamber for detecting ions evacuated from the ion trap.

[0039] In one embodiment, the ion mobility spectrometer comprises a plurality of dynamic ion traps arranged in parallel.

[0040] The dynamic ion trap may also be used as (or as part of) a micropump.

[0041 ] According to a preferred embodiment, a spectrometer comprising a dynamic ion trap in series with a differential mobility spectrometer.

[0042] These and other aspects of the present invention are recited in appended claims 2 to 15.

[0043] According to another aspect, a method of operating a dynamic ion trap is recited in claim 19. Preferred embodiments thereof are recited in dependent claims 20 to 25.

[0044] According to a further aspect of the invention, there is proposed a micropump comprising a dynamic ion trap device as described above. Such micropump hence allows trapping one or several ion types in a carrier gas (e..g air), based on their respective mobilities, to deliver them at the filter outlet.

[0045] In order to increase the throughput of the micropump, several electric traps can be operated in sequence along the series of electrode structures. Therefore, one may trigger two or several filtering cycles as mentioned above (although in the context of the micropump the cycle is not for "filtering" purpose as such), at different timings. In such case however the various local electric fields (trap) are separated from one another by at least one portion of sections with a passive electrode structure (i.e. not currently involved for creating the electric field). Brief Description of the Drawings

[0046] Preferred embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings in which:

Fig. 1 is a block-scheme of one embodiment the ion-trap filter system, which comprises a first gate (A), a first filter (B), a second gate (C), and a detector (D). A more advanced version of the device may include a second filter (E).

Fig. 2 are schematic diagrams showing the principle of functioning of the present dynamic ion trap. The electrodes suck the cloud of ions in (image 1 ). The activated attractive electrode pair (shown in black) and repulsive pair (shown in gray) successively switch (images B-F) pulling the ionic cloud.

Fig. 3 are graphs of (A) the eigenmobility as the function of the voltage applied across the second nearest neighbors electrodes at the value of the ratio of L²/x = 0.01 cm²/s and (B) the frequency range for the inter-electrode distance from 10 to 100 μπΊ necessary to get such a ratio. Fig. 4 is a graph illustrating the dependence of the resolution upon the number of the electrode pairs.

Fig. 5 are schematic diagrams illustrating the reciprocating motion driving by switching of electrodes voltage. The ions move towards the right end of the trap filter (A and B). After the last phase of the motion to the right, the electrodes are switched such that the one that has been attractive (black) during the previous step (shown in image B) becomes repulsive (gray) during the next step (image C), while the one that has been repulsive during the previous step becomes attractive. The electrodes switch further driving the ions backwards.

Fig. 6 is a schematic view of the enrichment chamber. The inlet (A) allows the air with ions of both signs (schematically shown by the circles) to enter the chamber. The outlet (B) with the gate (C) representing a pair of electrodes kept at some potential allow the air to leave the chamber, neutralize the ions of one sign and arrest those of the other sign. The gate electrodes are set to zero when the concentrated ions are allowed to leave the chamber. Description of Preferred Embodiments

[0047] Dynamic trap IMS [0048] The present example concerns an ion filter of a new type. The filter is designed to work in normal air at atmospheric pressure like conventional IMS or DMS devices. Its operation principle is based on the separation of ions on the basis of their mobilities in an external electric field, this feature is in common between this filter and IMS devices.

[0049] The operation of the present dynamic ion trap will be first described hereinbelow with respect to the embodiment of Figs. 1 -6 that employ electrode structures featuring a pair of rod-shaped electrode members. In the following an electrode structure is thus often simply referred to as "electrode pair".

[0050] The filter is based on the fractionated electrode pairs forming a chain. The voltage is applied across one pair of electrodes and its next nearest neighbour pair forming a local electric field. The term "next nearest" is used to designate the second neighbour electrode pair, not the one directly after.

[0051 ] The ions subjected to this field are forced to move towards one of the two mentioned electrode pairs. The voltage drop stays on the mentioned electrodes during a certain time, τ, which is followed by the jumpwise switching of the voltage to its nearest neighbour (rather than next nearest neighbour) pair, for both electrode pairs to which the previous voltage was applied, along the chain. Thus, a moving electric trap is realized. Its motion separates out the ions with certain mobility that are able to move "in phase" with the trap motion. The ions that have either a too small or a too high mobility are filtered out by the trap and are then neutralized during their collision with the electrodes. The filter may have microscopic geometrical parameters.

[0052] The resolution of the filter can be increased by (i) increasing of the number of electrode structures in the chain/filtering chamber, and (ii) by a special choice of the signal providing a multiple runs of the ion cloud along the filter back-and-forth.

[0053] The resolution can be further increased by placing the dynamic trap sensor in series with another sensor type, such as for example, the DMS.

[0054] For a complete gas sensor, the filtering chamber is normally combined with a detection chamber and gates. The latter should control the delivery of the ions into the filter and prevent the ions from moving from the detection chamber back to the filtering chamber. [0055] The sensitivity of the device may be increased (i) by enrichment of the air by analyte ions achieved in the enrichment chamber and (ii) by a parallel scheme of the filter geometry, in which several equal filters contribute to the same detection chamber.

[0056] The sensor can have an overall size ranging from few hundreds micrometers to few millimeters and may be operated with voltages in the range from 0.1 to 10 Volts.

[0057] The sensor device may comprise a first gate (A), a first filter (B), a second gate (C), and a detector (D). The more advanced version of the device may include the second filter (E). The block-schemes of the simple and advanced devices are shown in Fig. 1 .

[0058] The main part of the filter consists of a chain of pair electrodes as it is shown in Fig. 2. The cloud of ions is sucked into the filter with the velocity v_d by air flow (Fig. 2). The potential (attractive with respect to the analyte) is applied to one pair of the electrodes (indicated by black in Fig. 2) that suck the part of the cloud with the corresponding sign into the filter. The ions of the opposite sign are kept outside the filter volume by the same potential. The electrode pair behind the cloud gets a repulsive potential during the next step (it is indicated by gray). Thus, the electric trap is formed that stays on the electrodes during some time, τ. It is then switched down, while the two pairs ahead of are switched on.

[0059] Under the action of the electric field between the activated electrodes the ion cloud starts to move towards the attractive electrodes. After the stepwise propagation of the activated electrodes the ions that are fast enough to travel over the inter-electrode distance during the time τ again find themselves between the two activated electrode pairs. In contrast slow ions that cannot travel over the inter- electrode distance during the time τ appear outside backwards of the trap and neutralize. Too fast ions that pass considerably more than the inter electrode distance during this time reach the attractive electrodes and neutralize. Thus, the ions are selected with their mobilities in a certain interval.

[0060] This procedure lasts until the trap achieves the position with the attractive electrodes at the extreme of the filter. The procedure is terminated by switching the voltage off. The survived ions are carried then out of the filter into the detection chamber by the air flow. Alternatively, one may accelerate the cloud injection into the detection chamber by switching the attractive potential (black) of the last electrode pair off while the electrode pair preceding the last one becomes repulsive (as it is shown in Fig. 2F).

[0061 ] If the voltage U is applied across the activated electrodes (say, to the first pair and the third pair e.g. it is applied across the next-nearest neighbour pairs), and both the size of the entrance camera and the inter-electrode distance is ~L, the intensity of the driving electric field, E_d, acting on the ions is

E_d~U/2L (1 ) [0062] since the attractive electrodes are remote away from the repulsive one by two inter-electrode distances, 2L (Fig. 2). The velocity gained by an ion in this field is v~v_d+K₀ E_d~v_d+K₀U/2L (2)

[0063] where v_d is the drift velocity of air flow and K₀ is the ion mobility. We keep here the notation taken in the community working in the field of IMS and DMS.

[0064] As an example, let us estimate the relative weight of the two contributions in Eq. (2). Having in mind a device that should be low-consuming, assume the voltage U=1 V.

[0065] The ions have different mobilities, K₀. However, mobilities of most important organic volatile compounds at normal conditions are within the interval 0.5 to 2.5 cm²/V s, most often the mobilities being in the range of K₀~1 cm²/V s. For estimates the range from K_0min~0.5 cm²/ V s to K_0max~2.5 cm²/ V s will be chosen.

[0066] There is a certain freedom in choosing the limits of the drift velocity, v_d. Taking into account that the pump is the most problematic sub-device of the μ-IMS both from the point of view of its size, cost and energy consuming, and that the smaller is the power of the pump, the smaller will be the consumed power, we assume here that the drift velocity is in the range 1 mm/s to 1 cm/s.

[0067] Finally, for the inter-electrode distance, L, let us take two estimates J_/-10 μΓΠ and J₂~100 μηι.

[0068] With various combinations of these figures one finds K₀U/v_dL varying between -10 to~10000. One concludes therefore, that the velocity gained by the ion in the electric field by far exceeds the drift velocity v_d, and one can neglect the term v_d with respect to K₀U/L where appropriate.

[0069] Denote the time of switching of the electrodes (like that between the frames shown in Fig. 2) τ. The complete cycle consisting of jumping of the activated electrodes along n electrode pairs has the duration T_tot=n τ. The ions that move over the distance L during the time τ move "in phase" with the field and successfully pass the filter. Their mobility, K_e, therefore, is determined from the relation:

K_e~2L²/Ux (3)

[0070] The subscribt "e" at the mobility, K_e, stands for eigenmobility and indicates that this value corresponds to the mobility that is characteristic to the filter with fixed inter-electrode distance and switching time, determined by a certain voltage value.

[0071 ] Making use of the above figures and Eq. (3) one can estimate the interval of the switching time, τ, such that the mobilities between 0.5 cm²/ V s and 2.5 cm²/ V s will satisfy (3). The estimates yield τ between 10^"4 and 10^"6 s. On this basis and having in mind that the smaller is the frequency the less expensive is the device, one may advantageously choose τ~10^"4 s.

[0072] With this switching duration for the inter-electrode distance Li one covers the actual interval of variation of mobilities (e.g. from 0.5 to 2.5 cm²/ V s) by varying the voltage of the device from 0.01 to 0.1 V. One can see more details in the image Fig. 3 that shows the curve K₀=K₀(U) for 0.01 V≤U≤0. 1 V for the L²h ratio equal to 0.01 cm²/s.

[0073] Resolution

[0074] If the ion has a mobility exceeding the eigenmobility, K_e, it overtakes the attractive electrode pair resulting in falling of the ion onto the attractive electrode and its neutralization. This takes place, if during the time (η-1)τ the ion travels over the distance n L. Therefore, among the ions with the mobility K₀≥K_e only those survive for which (n-I)vx≤n L yielding:

[0075] If the ion mobility is a bit smaller than K_e, it will gradually fall behind the moving trap. If in some moment of time the distance the ion passes appears to be smaller than L at the end of the duty time, the repulsive barrier jumps ahead of the ion. As the result the ion is forced to move in the direction opposite to that of the dynamic trap.

[0076] The condition that the ion with ¾ < ¾ still passes the filter is that its displacement during n time intervals τ, represented as ντ , is larger than n-1 inter- electrode distances: ντ n<(n-l)L. This yields the inequality

[0077] One concludes that after n jumps only those ions survive whose mobilities fulfill the following inequality: i— < T7^L < i +

n A_e 7i— i

[0078] This yields the resolution interval, r, of the device:

K_€ Mil - 1) (6)

[0079] Mobilities within the interval are not resolved from one another.

One can decrease the resolution interval, r, Eq. (6) by increasing the number of the electrode pairs, n. One can see that the resolution r<0.1 can be achieved at «>20.

[0080] Typically, the difference between the mobilities of the organic volatile compounds manifests itself in the first figure after the decimal comma. However, some molecules have the mobilities the difference between which only manifest itself in the second figure after the decimal comma. For example, mobility of TNT is 1 .59 cm²/V/s, while that of diphenylamine is equal to 1.54 cm²/V/s; the mobility of dimethylmethylphosphate is 1 .95 while that of benzene is 1 .94 cm²/V/s.

[0081 ] In order to achieve a resolution on the level of the second figure after comma, one should preferably have the number of pairs n~100, while for a still better resolution (e.g. to achieve the resolution that is better than that of modern IMS devices) one needs to go to «>500. This dramatically increases the difficulty of manufacturing and its cost. [0082] The same level of resolution can however, be achieved in a different way by manipulating the signal applied to the electrodes and without using of too long electrode chains. The idea is that the dynamic ion trap may make multiple runs back- and-forth thus, effectively increasing the number of electrode pairs in the filter. This is illustrated in Fig. 5.

[0083] Within the approach that is put forward here, the dynamic trap moves back and forth. The switching of the direction of the trap motion takes place after the termination of the last phase of its motion in a certain direction, the one during which the attractive electrode pair occupies the extreme position. This is illustrated in Fig. 5 in which the image (B) shows the last phase of the motion of ions to the right. In the present scheme the polarity of the electrodes of the last step of the first run is inverted (compare the Fig. 5 (B) and (C)).

[0084] If the ion cloud is forced to make m runs back and force, it is equivalent to run through a linear device with n_eff=n> m electrode pairs making the resolution equal to

2 n m— I

n m {n m - 1) ^

[0085] For example, with «=20 electrode pairs and m=50 runs one finds the resolution interval ^r = ^{2 x 10-3} .

[0086] Timing

[0087] The time of a single run is

(8)

[0088] The total time of flight, T„_,m, through the filter with n electrode pairs and consisting of m runs is expressed as follows:

(9) [0089] Since in such a situation the device is characterized by the resolution r Eq. (7), the whole intervals of mobilities of interest (say, from K_mm=0.5 cm²/V s to K_max=2.5 cm²/V s) should be divided into nxm intervals, corresponding to nxm scans swiping the whole interval. This means that one measurement of the complete interval of interest consists of nxm cycles with the duration of T„_{: m} each cycle. This makes the total measurement time, T_tot, to be expressed as follows:

[0090] Taking τ~10^"4 s and∞ «=100 one finds Γ_ίοί~1 s. If one has the target of the resolution r=10^"3, one finds Γ_ίοί~100 s.

[0091 ] The total measurement time obtained above may be too large to be practical in some cases. It can be however reduced by installing k parallel channels connected to k detector chambers, each chamber belonging to its own channel. In such case each channel may be configured so as to be responsible of its own interval of mobility, and the whole interval scanned by each channel is therefore k times smaller. One finds in this case

[0092] making in the case with the resolution r~10^~3 and k=10 channels in parallel the total time T_tot~10 s.

[0093] Sensitivity

[0094] Upon leaving the filter and the second gates the ions enter the detection chamber, where two electrodes are situated with a voltage applied across them. The ions are neutralized on these electrodes, giving rise to the electric current in the detector chain. The latter signal can be amplified and registered. Here we estimate its value.

[0095] Denote the air number density

10¹⁹ cm^'3, the number density and concentration of the analyte ions as N_a and c:

(12) [0096] Denote further H and Wi e height and width of the channel. Assume further that m channels are arranged in parallel injecting the ions into the same detection chamber (which is different with respect to the configuration described in the previous Section). Then the charge, Q, injected into the detection chamber is expressed as Q~ e N_a L W H, where e is the elementary charge, e=L6x ]0^~19 A s. The time of discharge depends upon the time of drift of ions to the electrodes. The latter is dependent upon the voltage across the detection electrodes and the size of the detection chamber. At high voltages the drift time becomes negligible, but the discharge time anyway cannot be smaller than the time of delivery of the ions into the chamber ~τ. For this reason the electric current, J, in the detection chain is not larger than

[0097] Estimate of the current assuming H~ff~J~100 μηπ, τ~10^"4 to 10^"5 s

The typical concentration of the analyte is 1 ppm. However, one should take into account that only a small fraction of molecules of interest can be polarized in the polarization chamber. The realistic concentration of the ions passing the filter is therefore, c~10^"10. This yields J~l to 10 pA. [00981 Parallel scheme to increase sensitivity

[0099] However, the sensitivity may be still increased several fold by using the parallel scheme. It is analogous, though not identical to the scheme that has already been discussed above for the purposes of reducing the measurement time. If the relatively long time of the device operation may be tolerated, the parallel scheme may be applied differently. In this case all the channels are scanning the same mobility interval and inject their ions into the same detection chamber. The current of the channels is therefore, summed up, and the current Eq. (14) should be multiplied by the number of channels, k.

[00100] In the case of £=10 channels in parallel the estimates of the previous example yield 10 to 100 pA .

[00101 ] Enrichment

[00102] In order to increase the sensitivity the analyte may be enriched in the enrichment chamber. The enrichment chamber may be designed as follows. The chamber has at least one in- and one outlet, the latter being controlled by gate. Figure 6 displays the schematic view of the enrichment chamber. The inlet of the chamber allows the ions of the both signs to enter the enrichment chamber carried by the air flow. The outlet is equipped by the gate represented by a pair of electrodes kept at a certain potential. These electrodes will attract ions of one sign (negatively charged, if the potential is positive, and vice versa), but will prevent the ions of the opposite charge to leave the enrichment chamber and to enter the filter. Thus, the ions of a certain charge will be concentrated in the enrichment chamber.

[00103] To estimate the voltage necessary to stop the ions, observe that the motion of the ions is viscous, their drift velocities obeying the relation v = bf, where / is the force acting on the ion and driving its motion, and b=K( e is its mobility. In the case under consideration the force is due to "friction" of ion with respect to air. This should be opposed by the electrical force generated by the gate f_ei~U_g e/L_enr, where U_g is the gate potential, and L_mr is the size of the enrichment chamber (Fig. 6). The air is unable to carry out the ions, if f≥f yielding

[00104] The size of the chamber must be close to the inter-electrode distance in order to provide the possibility to deliver most of the ions stored in the chamber into the filter upon opening of the gate. Assuming v=v_rf~1 mm/s, J_e„_r~10 μηι and K₀~1 cm²/V s one finds ¾~10^"4 V. Thus, application of the gate voltage of 10^"1 to 10^"2V as used for the filter electrodes described above would be enough.

[00105] To estimate the enrichment achieved by application of this enrichment chamber note that the concentration of ions collected during the enrichment time, t_enr, in the chamber with the cross-section ~D² and with the length ~L_mr is independent of D: c~c₀ v_d t_mr/L_mr. The enrichment, u, is expressed as

[00106] where co=N_a/N_mr is the concentration of the analyte in the gas right after the ionization chamber and before the enrichment chamber. One finds:

[00107] Assuming the enrichment time, t_e„_r~1 s, and J_e„_r~100 μηι one finds u~10, while for L_enr~\ 0 μηι one finds w~100.

[00108] Going to lower values of L_mr one can hope to increase the concentration factor to ~10³. For very small L_mr however, the field generated by the gate may lock the inlet of the enrichment chamber for the analyte ions.

[00109] Enrichment of the analyte has a direct effect on the device sensitivity. The electric current, J, of the device with the enrichment chamber is related to that, J₀, without such a chamber as follows:

J=R Jo

[001 10] Thus, combining the enrichment with parallelization of the channels one may achieve the increasing the current 100 up to 1000 fold reaching the values of current up to 1 to 10 nA.

[001 1 1 ] Second filter

[001 12] The device may be designed either smaller, or faster still keeping a high resolution by applying a second filter based on the DMS principle. The main idea in such unification is that on one hand the both devices work at atmospheric pressure and therefore, they can be conjugated directly. That is, the ions leaving the first filter described above directly enter the second filter without any precautions, rather than the detection chamber. They enter the detection chamber only after passing the second filter. On the other hand the principle of operation of the dynamic trap filter is based on the measurement of the mobility, while that of the DMS is based on the measurement of the mobility increment. Since these two parameters are independent of one another, two substances that could not be separated by one filter will be separated by another with a very high probability and vice versa.

[001 13] If in this situation the DMS is placed as the second filter, it is not necessary to achieve the high resolution of the dynamic trap filter. It is for the following reason. The probability that two chemically distinct analytes present in the same air sample have close mobilities is rather small. The same statement is valid for probability that two analytes have close mobility increments. The probability that two distinct analytes have simultaneously close mobilities and mobility increments equal to the product of two above probabilities is negligible. For this reason, if two molecules of chemically distinct analytes are already pre-filtered by the dynamic trap and are nevertheless still not separated, they have close mobilities. In this case, however, with the high probability they possess the mobility increments that differ considerably from one another.

[001 14] Dynamic trap as a micropump

[001 15] Here another application of the dynamic trap is briefly described. Namely, it can be used in a regime of a pump.

[001 16] Consider the dynamic trap described above. Assume that the ions with the concentration c are injected into the trap, all the ions being of the same type, and the trap parameters are fixed such that the mobility of these ions represent the eigenmobility Eq. (3).

[001 17] In contrast to the device described above, to make the pump configuration one makes n/2 moving traps (this is in contrast to a single moving trap necessary for the filter. Within the trap the ions are subjected to the electric force, f~e U/L and move on average with a constant velocity. This implies that the force resisting their acceleration is acting on ions from the side of air due to friction.

[001 18] Though in this problem the Reynolds number may become somewhat larger than unity, for the estimates here the solution of the Navier-Stokes equation describing the flow in a tube is adopted:

[001 19] where Q is the mass of the air passing the device during the unit time, P is the air pressure, Δ P is the difference between the air pressure at the in- and outlets, η is the air viscosity, p is the air mass density, f is the force acting on the ion within the trap and introduced above and N is the number density of ions, H is the width of the channel, L is the distance between the electrode pairs and n is the number of pairs.

P~fNL n=e Uc Nai_rn (18)

[00120] This is the maximal pressure difference that can be achieved by application of the device. Note that though Eq. (17) is the result of the solution of the Navier- Stokes equation, the conclusion (18) is valid in general (also beyond the limits of the application of the mentioned equation).

[00121 ] Estimate the pressure difference that can be achieved with such a pump. Assume the chain consisting of 20 electrode pairs, to which the voltage l/~1 V is applied, the air number density is N_a/,.=2.7x10¹⁹ cm^"3, e=1 .6*10^"19 C and assume the concentration of ions to be c~10 to 100 ppm. One finds Δ ~10³ to 10⁴ Pa.

[00122] The above being said, which concern a very detailed embodiment of the present device, some remarks remain to be made concerning further possible embodiments. [00123] Fig.7 illustrates an embodiment of the present device 100 where the filtering chamber 102 is of simple parallelepiped shape, although other shapes may be considered. The chamber 100 defines a filtering path along which six electrode structures 104-| ... 104₆ are distributed. Each electrode structure 104, comprises a pair of parallel rod-shaped electrodes members 106. Reference signs S1. . . S5 indicate 5 chamber sections defined between the respective nearest neighbor electrode structures 104,, 104_i+i . A control unit 108 is operatively connected to the electrode structures 104, so as to control the electric field in the chamber 102. The electrode members of a given electrode structure may be connected to the control unit by an individual, distinct circuit or wire. Alternatively, electrode members of a given electrode structure may be interconnected by a bridging member 1 10 as illustrated for electrode structure 104₅. Such bridging members 1 10 may be arranged inside or outside the chamber 102, in which latter case direct collision with ions is prevented. Even if the electrode members 106 are individually connected to the control unit 108, all electrode members belonging to a given electrode structure 104, are operated jointly, i.e. they are switched on or off simultaneously and a same voltage is applied.

[00124] As explained above, the control unit is operated to create a local electric field and displace the local electric field along the filtering path at a pre-defined pace in accordance with the ion type to be detected. Supposing that an ion cloud enters from the left side 1 12 of the chamber 102, the control unit controls the application of an appropriate voltage on a pair of electrode structures that are next nearest neighbours, and moves this field progressively. Accordingly, except for the inlet and outlet control steps already explained above, the control unit will apply, in sequence, a voltage between the following electrode structures (104-1 , 104₃); 104₂, 104₄); (104₃, 140₅); (104₄, 104₆). For every pair of electrode structures, the voltage is applied for during the length referred to as duty time.

[00125] To reduce fringe field effects, conductive patches may be applied on the internal wall surfaces of the chamber 102. For same purposes, low conductive layers may be applied on the chamber walls at both ends of the electrode members.

[00126] Turning now to Fig.8, there is shown an embodiment of the present device 200, where same or similar elements are indicated by same reference signs as in Fig.7, increased by 100. As can be seen, each electrode structure comprises 5 parallel electrode members 206. The electrodes members 206 of each electrode structure substantially coincide in position along the filtering path. Such configuration permits increasing the current in the filtering chamber by a parallelisation of similar pathways with same timing, and results in a more homogeneous field towards the internal region.

[00127] Fig.9 shows a further embodiment of the present device 300, where same or similar elements are indicated by same reference signs as in Fig.7, increased by 200. This embodiment employs a greater number of electrode structures, which are spaced by respective sections of reduced length. With such configuration, the control unit may apply the voltage to create an electric field over more than two sections, say e.g. 4. Starting from the situation where the voltage is applied between electrode structures 304i and 304₅ (grey electrodes in Fig.9), during the next duty time the voltage will then be applied between electrode structures 304₂ and 304₆, then electrode structures 304₃ and 304₇, and so on keeping the same spacing to have the same field distribution. The idea here is that by having more and narrowly spaced electrode structures, the applied voltage jumps more often, but the local electric field moves more smoothly.

[00128] In such configuration, the intermediate electrodes (those in-between the electrode structures to which the voltage is applied to create the local electric field) may be left floating or set to a predetermined potential.

[00129] Alternatively, the control unit may apply to the intermediate electrodes a voltage gradient depending on the position along the filtering path. This can result in an overall linearly changing potential, which results in a homogeneous electric field. In practice, the use of a greater number of electrode structures should however be balanced with the fact that it can decrease transmission.

[00130] It remains to be noted that so-called compressing electrodes may be used in order to help to keep the ion cloud more compressed towards the box centreline, and hence prevent the ion cloud from colliding too fast with attractive electrodes. These compressing electrodes produce a repulsive electric field that tends to compress the ion cloud.

[00131 ] Such compressing electrodes are disposed preferentially outside of the chamber 102. The compressing electrodes are preferably operated by the control unit in correlation with the operation of electrode structures in the filtering chamber, such that these compressing electrodes are only energized when the ion cloud achieves the position against the corresponding electrodes along the filtering path, and switched off again after this cloud passes them.

Reference Documents

[1 ] Baumbach, J. I. & Eiceman, G. A. Ion mobility spectrometry: Arriving on site and moving beyond a low profile. Applied Spectroscopy 53, 338A-355A (1999).

[2] Borsdorf, H. & Eiceman, G. A. Ion mobility spectrometry: Principles and applications. Applied Spectroscopy Reviews 41 , 323-375. (2006).

[3] Zimmermann, S. & Barth, S. in TRANSDUCERS Ό7 & Eurosensors XXI. 2007 14th International Conference on Solid-State Sensors, Actuators and Microsystems 1501 -4, (IEEE, Piscataway, NJ, USA, Lyon, France, 2007). Zimmermann, S., Barth, S., Baether, W. K. M. & Ringer, J. Miniaturized Low-Cost Ion Mobility Spectrometer for Fast Detection of Chemical Warfare Agents. Analytical Chemistry 80, 6671 -6676 (2008). Zimmermann, S., Abel, N., Baether, W. & Barth, S. An ion-focusing aspiration condenser as an ion mobility spectrometer. Sensors and Actuators B125, 428-434 (2007).

Claims

1 . Dynamic ion trap device comprising a filtering chamber having therein a series of electrode structures partitioning said chamber into a succession of sections, each section being fluidly connected to its nearest neighbour sections, and a control unit operatively connected to said electrode structures and configured to perform a filtering cycle comprising locally creating an electric field across at least two neighbouring sections and displacing said electric field in a stepped manner along said succession of sections.

2. Dynamic ion trap device as claimed in claim 1 , wherein said control unit is configured to, during a filtering cycle, displace said local electric field at predefined pace depending on ions to be filtered.

3. Dynamic ion trap device as claimed in claim 1 or 2, wherein said control unit is configured, during a filtering cycle, to displace said electrical field in accordance with the mobility of ions to be filtered.

4. Dynamic ion trap device as claimed in any one of the preceding claims, wherein said filtering cycle comprises: creating an electric field at the beginning of a filtering path of said chamber by applying a voltage during a duty time on a pair of electrode structures covering at least the first two sections of said filtering path, and displacing said electric field in a stepped manner along said filtering path by applying the voltage, during a duty time, on the next, nearest electrode structure in the chain following each of the previously active electrode structures.

5. Dynamic ion trap device as claimed claim 4, wherein said duty time is constant along the filtering path or variable.

6. Dynamic ion trap device as in claim 4 or 5, wherein during a filtering cycle, the duty time (τ) and voltage (U) are set according to the mobility (K) of ions to be trapped.

7. Dynamic ion trap device as in claim 4, 5 or 6, wherein said control unit is configured to repeat said filtering cycle with different electric field parameters to detect ions with different mobilities.

8. Dynamic ion trap device as in claim 7, wherein said filtering cycle is repeated by scanning over a given voltage range.

9. Dynamic ion trap device as claimed in any one of the preceding claims, wherein said control unit is configured to, during a filtering cycle, displace said local electric field back-and-forth along at least part of said filtering path.

10. Dynamic ion trap device as claimed in any one of the preceding claims, wherein neighbouring sections are separated from one another by one electrode structure.

1 1 . Dynamic ion trap device as claimed in any one of the preceding claims, wherein each electrode structure is connected to said control unit to be operable at a selected potential or let at floating potential.

12. Dynamic ion trap device as claimed in any one of the preceding claims, wherein an electrode structure comprises one, two or more electrode members.

13. Dynamic ion trap device as claimed in any one of the preceding claims, comprising at least one ionizing chamber upstream of said detection chamber.

14. Dynamic ion trap device as claimed in any one of claims 1 to 13, comprising at least one enrichment chamber upstream of said chamber.

15. Ion mobility spectrometer comprising a dynamic ion trap as claimed in any one of claims 1 to 14, at least one inlet gate for introducing ions into said dynamic ion trap, at least one outlet gate for evacuating ions from said dynamic ion trap and at least one detection chamber for detecting ions evacuated from said ion trap.

16. Ion mobility spectrometer as claimed in claim 15, wherein a plurality of dynamic ion traps devices are arranged in parallel.

17. Spectrometer comprising a dynamic ion trap device as claimed in any one of claims 1 to 14 in series with a differential mobility spectrometer.

18. A method of operating a dynamic ion trap device comprising a filtering chamber having therein a series of electrode structures partitioning said chamber into a succession of sections, each section being fluidly connected to its nearest neighbour sections, said method comprising a filtering cycle, in which a local electric field is displaced in a stepped manner along said succession of sections, said local electrical field extending across at least two neighbouring sections.

19. The method according to claim 18, wherein during said filtering cycle, said local electric field is displaced at pre-defined pace depending on the ion analyte to be filtered.

20. The method according to claim 18 or 19, wherein during said filtering cycle, said electric field is displaced in accordance with the mobility of the ion analyte to be filtered.

21 . The method according to any one of claims 18 to 20, wherein said filtering cycle comprises: displacing said electric field by successively applying a voltage between a pair of electrode structures from one end of a fitering path along said series of electrode structures to another end thereof, the voltage being applied each time during a duty time, and the voltage being applied, at the expiry of the duty time, to the next, nearest electrode structure in the chain following each of the previously active electrode structures.

22. The method according to any one of claims 18 to 21 , wherein during the filtering cycle, the voltage which is successively applied to the electrode structures to create the electric field and its duration are set according to the mobility ( ) of ions to be trapped.

23. The method according to any one of claims 18 to 22, wherein said filtering cycle is repeated with different electric field parameters to detect ions with different mobilities.

24. The method according to any one of claims 23, wherein said filtering cycle is repeated by scanning over a predetermined voltage range.

25. Micropump comprising a dynamic ion trap as claimed in any one of claims 1 to 12.

26. Micropump according to claim 25, wherein it is configured for pumping a mixture of air with at least one type of organic volatile molecule.

27. Micropump according to claim 25 or 26, wherein two or more filtering cycles are performed concurrently, with a shifted starting timing, so that several local electric fields travel along said series of electrode structures, with a given spacing between one another.