CA2234754A1 - Multiple reaction monitoring mass spectrometer and method - Google Patents

Abstract

A method of operating a mass spectrometer system in which ions are trapped in a first quadrupole mass spectrometer operated in an RF-only mode, at relatively high pressure, and all ions except parent ions are then removed therefrom using FNF or a similar technique. The ions are then fragmented in the first spectrometer, using radial or axial excitation or other desired method, and are then transferred to a resolving mass spectrometer for analysis.

Description

CA 022347~4 1998-04-14 Title: MULTIPLE REACTION MONITORING MASS
SPECTROMET:ER AND METHOD

FIELD OF THE INVENTION
This invention relates to a multiple reaction monitoring mass spectrometer and method.

BACKGROUND OF THE INVENTION
Multiple reaction monitoring (MRM) is a method which is usually performed by a single mass spectrometer system. In the MRM
method of operation, the rxlass spectrometer is programmed to look for 10 multiple compounds, by "peak hopping", i.e. the system looks at m/z ratios where peaks are expected to occur if the compounds of interest were present.
Current MRM methods of operation require relatively expensive mass spectrometers, are relatively slow in practice, and require a 15 substantial quantity of sample. Therefore there is a need for an improved MRM system.

BRIEF SUMMARY OF THE INVENTION
In one aspect this invention provides a method of operating a mass spectrometer comprising: trapping ions in a first multipole rod-type 20 mass spectrometer, applying an auxiliary field to said first mass spectrometer to remove all ions therefrom except for parent ions, fragmenting the parent ions in said first mass spectrometer to produc fragment ions, and transmitting said fragment ions from said first mass spectrometer into a second and mass resolving mass spectrometer.
Further objects and advantages of the invention will appear from the following description, taken together with the accompanying drawings.

CA 022347~4 1998-04-14 BRIEF DESCRIPTION OF THE DRAWINGS
In the drawings:
Fig. 1 is a diagrammatic view of a mass spectrometer with which an MRM method accvrding to the invention may be used;
Fig. 2 shows application of FNF to rods of the Fig. 1 mass spectrometer;
Fig. 3 shows a filtered noise field signal for use with the Fig. 1 system;
Fig. 4 shows application of a dipole field to rods of the Fig. 1 10 mass spectrometer;
Fig. 5 shows auxiliary rods for use with the Fig. 1 mass spectrometer;
Fig. 6 is a diagrammatic view showing a modified mass spectrometer system for use with the method of the invention; and Fig. 7 is a diagrammatic view showing a still further modified mass spectrometer system fvr use with the method of the invention.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
Reference is i irst made to Fig. 1, which shows a mass spectrometer system 10 having an atmospheric pressure ion source 12, 20 which produces ions which travel through an orifice plate 14, a vacuum chamber 16 pumped to (fox example) about 2 torr, and then through an orifice in a skimmer 18 into a vacuum chamber 20 pumped to between 1 and 8 millitorr. Vacuum chamber 20 contains a first quadrupole mass spectrometer Q0.
Vacuum chamber 20 is connected through orifice 22 in lens IQ1 to a further vacuum chamber 24 containing a second quadrupole mass spectrometer Q1. Vacuum chamber 24 is pumped to a low pressure of less than or equal to about 5 x L0-5 torr, and contains a detector 26 to receive and detect ions from Q1. The detected ions are recorded by a data system 30 27.
Both quadrupoles Q0 and Q1 can be driven from the same CA 022347~4 1998-04-14 quadrupole power supply 28, which can supply RF and DC to both Qo and Ql. Power supply 28 also supplies suitable DC bias voltages on orifice plate 14, skimmer 18, the rods of (2~ and Ql, and lens IQl. (The DC bias voltages applied to the rods of Qo and Ql are commonly referred to as the rod offset 5 voltages.) In the MRM analysis mode, ions are trapped in Qo simply by raising the IQ1 DC bias voltage about 5 volts above the Q0 rod offset DC
voltage (for positive ions). QOis operated at this time as an RF-only quadrupole, in which the RF is set to reject ions below the mass range of 10 interest and to contain the ions. At the same time, a filtered noise field (FNF) is applied to Qo. The technique of applying FNF to a quadrupole device is well known and is the subject of U.S. patents 3,334,225 by Langmuir and 5,187,365 by Kelly. In this method, and as shown in Fig. 2, a dipole field is applied from an FNF source 30 to two of the rods of Qo. The 15 field has the appearance shown at 32 in Fig. 3 and has a notch 34 therein.
Thus a dipole excitation signal is applied to excite all of the ions in Qo except for ions having a fundamental frequency of ion motion corresponding to the frequency of the notch 34. Ions having that frequency, referred to as parent ions, are accumulated in Qo. The 20 accumulation time depends upon the parent ion flux and the containment capability of Qo. For exarmple space charge effects in Q0 may limit the number of parent ions in Ql) to about 106 ions. The accumulation time is preferably between about 10 milliseconds and several seconds.
At the end of the accumulation period, further ion input 25 from source 12 is blocked by setting the DC bias potential on the orifice 14 to result in a repulsive field between the orifice and the skimmer 18. The parent ions in Qo are then fragmented by collision activated dissociation (CAD). The q value of the parent ions at the moment of dissociation should be relatively low so that there will be a reasonable range of 30 fragment ions. For example, if the parent ion has q = 0.15, the daughter ion mass range is approximately 0~ 195 = 6.

CA 022347~4 1998-04-14 The CAD described above may be performed in various ways, including the following.
(1) The ions in Qo may be excited by applying radial excitation. Here, about 1 vclt peak to peak of dipolar AC is applied, at the notch frequency, for about 10 ms to 40 ms, using dipole source 40 (Fig. 4).
Source 40 may be the same as FNF source 30 but operated in a different mode. The dipole excitation causes the parent ions in Qo to increase their amplitudes of oscillation and to fragment by collisions with the background gas in Qo. The method is analogous to that used in an ion 10 trap, where ions are excited to fragment by excitation at their secular frequency of oscillation in the trap. This method requires that the ions be gently excited so that their amplitude of oscillation does not exceed the space in between the rods, in order that the ions not strike the rods. Again, QOis operated at this time in an RF-only mode, the RF being supplied 15 from RF source 41, which is part of the quadrupole power supply 28.

(2) The ions in Qo may also be excited axially by applying an oscillating axial electric fi eld as described in copending U.S. patent application serial no. 08/514,372 filed August 8, 1995. The axial excitation is not a resonant process, so the frequency may be optimized 20 independently of the ion mass. Ions in Qo are accelerated forwards and backwards along the axis of Qo, experiencing several oscillations, and thus increasing the number of collisions and the probability of fragmentation.
In addition the ions may be accelerated to higher energies without risking the loss of ions to the rods, since the acceleration is along the axis. The 25 method chosen for creating the axial oscillating field may be any of the methods disclosed in said copending application, including for example providing four segmented auxiliary rods 42 located between the main rods of Qo as shown in Fig. 5. Appropriate DC voltages are applied to the segments 44 of the auxiliary rods 42 to create the desired axial field along 30 the length of the main rods of Qo. During this process, QOis operated in an RF-only mode.
In the axial oscillation method, in order to achieve sufficient CA 022347~4 1998-04-14 ion-neutral collision energy for fragmentation, the ions may typically need to gain about 50 eV of energy between collisions. Since the average ion-neutral collision distance is approximately 1 cm at 10 millitorr, therefore the axial oscillation field should be at least approximately 50 volts per 5 centimeter.

(3) An "out" and "in" method may be used, in which parent ions are removed axially from Q0 into Q1 with low energy (e.g. less than 5 eV energy) so that they do not fragment, and are then axially returned from Q1 back to Q0 with about 50 eV energy for fragmentation purposes.
10 This technique may be employed by using appropriate bias or rod offset voltages on Q0 and Q1 and on lens IQ1, as well as by providing an exit lens shown in dotted lines as IQ2 at the exit end of Q1. The bias voltage on lens IQ2 will be set to reflect ions passing through Q1 back toward Q0. Axial fields may also be employed to achieve high efficiency. Here, Q1 as well as QOis operated in an RF-only mode, to contain all of the ions of interest.
When any of these three methods is used, a fragmentation spectrum of ions will now be contained in Q0. Since QOis operated as an RF-only quadrupole, with a cut-off frequency at or below the m/z ratio of the lowest mass ion of interest, it can contain all the unfragmented parent ions and all the fragments of concern.
Next, the DC bias on lens IQ1is lowered to, for example, 5 volts below the rod offset of Q1, which itself is several volts below the rod offset of Q0. The fragment ions and any remaining parent ions then pour from Q0 into Q1 and are counted. If the rate at which ions enter Q1is too high, then the rate at which the IQ1 voltage is lowered may be controlled to lower this voltage more gradually, in order to limit the ion count rate.
Normally during the counting process, Q1 will have resolving RF and DC applied to it from source 28, to act as a resolving quadrupole and to pass only one fragment mass. It is preferred to begin 30 parent ion accumulation for each fragment mass desired, provided that the fragmentation time (which is an "overhead") is a small fraction of the accumulation time, so that the duty cycle is not severely affected.

CA 022347~4 1998-04-14 If it is desired to look at two fragment ions from the fragments contained in QO, without accumulating a fresh set of fragments for each fragment to be detected, then first Q1 will be set (by controlling the RF and DC therein from power supply 28) to pass or resolve one set of 5 fragment ions, and then thereafter to pass or resolve the second set of fragment ions. However t]-e difficulty in this approach is that once the potential barriers preventing transmission of ions from QO to Q1 have been lowered, ions tend to leave exponentially, with 80% of the ions exiting QO in several milliseconds. Therefore, the ion count of the 10 fragment ions measured first would greatly exceed that of the fragment ions measured second. The problem is aggravated by reason of the fact that most existing quadrupole power supplies 28 have a settling time of about 1 ms. Therefore, by the time the instrument is ready to count the second set of fragment ions, they will mostly have gone.
A modified version of the instrument shown in Fig. 1 is shown in Fig. 6, in which corresponding reference numerals indicate parts corresponding to those of Fig. 1. In Fig. 6, quadrupole QO is divided into two separate quadrupoles QOa and QOb, each in its own vacuum chamber 20a, 20b and separated by an apertured lens IQO. Appropriate DC bias and rod offset voltages are provided by conventional means, not shown. The vacuum chamber 20a containing QOa is pumped to about 8 millitorr, while the vacuum chamber containing QOb is pumped to between .5 and 1 millitorr. In the Fig. 6 arrangement, the MRM analysis begins by accumulating parent ions in QOb, using FNF in QOb to remove unwanted ions. Both QOa and QOb operate in the RF-only mode during this process, so ions travel through QOa to QOb, where they are trapped by appropriate DC potentials on lens IQ1 and by appropriate rod offset voltages on QOa and QOb. Resolution is believed to be better at the somewhat lower pressures in QOb than at the higher pressure of QOa.
Once the desired parent ions have been accumulated in QOb, the entry of further ions into QOa is blocked by adjusting the DC bias voltage on interface plate 14' and/or on skimmer plate 18'. The parent CA 022347~4 1998-04-14 ions in QOb are then transferred back through lens IQO to QOa for fragmentation in QOa in the manner described previously. The transfer back is effected by resetting the rod offset voltages in QOa and QOb, and by applying an axial field if desired. Fragmentation is more efficient at the 5 higher pressures of QOa, since there is more gas with which the ions may collide.
Preferably each of QOa and QOb is about 4 inches in length, i.e.
each is about half the length of QO (which is about 8 inches long) in Fig. 1.
Preferably the ratio of the RF level applied to QOb, relative to that applied 10 to Q1, is high, e.g. 0.9, while the ratio of the RF level at QOa relative to that at QOb is e.g. about 0.5, to provide a substantial fragment mass range.
Again, only one quadrupole power supply 28 is necessary for all three quadrupoles.
If the ions are transferred from QOb to QOa by setting the rod 15 offset of QOa to about 50 volts below that of QOb (an ion dependent parameter), this will provide a high probability of fragmentation, and the process is quick (less than 5 milliseconds), whereas radial CAD will require about 20 to 40 ms. The shorter fragmentation time will improve the MRM
duty cycle. If radial CAD is used, it should be carefully controlled, to 20 minimize the risk of running the parent ions into the rods.
The ions travelling from QOb back to QOa and fragmented there are trapped in QOa by setting the rod offset voltages and the lens voltages appropriately. Then, the voltages are set to transfer the fragment ions from QOa back through QOb and through Q1 for detection by detector 25 26. As before, QOa and QOb are operated as RF-only quadrupoles, while Q1 is operated in a mass resolving mode at low pressure (e.g. 10-5 torr or less) to pass the fragment ion of interest.
Reference is next made to Fig. 7, which shows a further modified version similar to that of Fig. 1 but used to perform MS/MS. In 30 the Fig. 7 version, double primed reference numerals indicate parts corresponding to those of Fig. 1. After a desired parent ion is isolated in QO (e.g. using FNF) and fragmented there (e.g. by using axial or radial CA 022347~4 1998-04-14 excitation), a desired fragment ion is transmitted through Q1" (which is operated in a mass resolving mode, using RF and DC) into Q2, which is operated in an RF-only mode and is located in a chamber 50 at relatively high pressure (e.g. 8 millitorr). The fragment ions are fragmented again in Q2 (again using axial or radial excitation or other appropriate means of fragmentation) to form second fragment ions. A selected second fragment ion is then passed through Q3 (which is operated in a mass resolving mode at low pressure, e.g. 10-5 torr or less) and is detected by detector 26.
It will be appreciated that wherever FNF has been described 10 (i.e. a filtered noise field having a notch so as to reject all ions except those of a specific mass to charge ratio), the well known technique of dipolar frequency scanning may be used instead to eject unwanted ions (i.e. a dipolar field is applied at a particular frequency and is scanned to eject ions having mass to charge ratios below and above that of the desired ion).
15 Alternatively, a supplementary quadrupolar field may be applied and scanned to achieve this result, instead of a dipolar field.
In an alternative method of isolating ions of desired mass to charge ratio, the ions of mass to charge ratio interest are positioned along the q axis of the well known a/q stability diagram at q = 0.45 20 (approximately), and then a dipolar field is applied to create a dipolar "hole" on the q axis at q = 0.22 (approximately). The RF field drive amplitude is then increased to move the ions of interest to a q = 0.9, and then down to q = 0.22, thereby isolating the ions of interest in a mass range with a factor of 8. Ions having mass to charge ratios above and below that 25 of the desired ion are eliminated by this method. While scanning upwardly to eliminate low mass ions is a relatively fast process, scanning down to move high mass ions through the hole at q = 0.22 can be relatively slow, but the speed can be increased by applying an auxiliary or supplementary quadrupolar field instead of a dipolar field to create a hole 30 at q = 0.22.
Yet another method of isolating ions is to position the desired ions at q = 0.7, and then applying DC as in a standard resolving quadrupole CA 022347~4 1998-04-14 _ 9 _ having DC and RF applied, so that ions of the desired mass to charge ratio are at the tip of the eventual first stability region. However, care must be taken with this approach so that significant signal losses are not caused by gas present in the device.
Finally, reference is again made to Fig. 1, in which an optional solenoid valve 60 may be installed in vacuum chamber 24 to admit gas (e.g. helium or argon from gas source 62) into chamber 24 when desired.
In use of the modified Fig. 1 apparatus, gas from source 62 is 10 admitted into Q1 to pressurize it to a pressure of about 1 millitorr. Then Q0 is operated as an RF-only ion guide to admit ions from source 12 into Q1, where the ions are trapped. The gas pressure from source 62 serves to collisionally cool the ions.
The trapped ions in Q1 are then subjected to an isolation step 15 as previously described, e.g. using FNF or one of the other techniques described, to isolate ions having a desired mass to charge ratio. The isolated ions are then directed back into Q0 (which is also pressurized at this time to between about 1 and 8 millitorr, from the atmospheric pressure ion source 12, as previously described), in order to fragment the 20 ions to produce daughter ions, which are trapped in Q0.
After the fragmentation process has occurred, Q1 is evacuated to about 10-5 torr by its pump. Then the trapped daughter ions in Q0 are directed into Q1 (by appropriate adjustment of the rod offset potentials as described), and Q1 is operated as a resolving quadrupole (e.g. by applying 25 RF and DC thereto), to transmit only the daughter ions of interest, which are detected by detector 26.
Since gas can be introduced into chamber 24 very quickly (in less than 20 milliseconds), and can also be removed in about 20 milliseconds (or less with a sufficiently large pump), the process can be 30 carried out reasonably quickly. While the higher pressure in Q1 may in some ways affect resolution adversely, it is helpful in that it removes kinetic energy from the ions injected into Q1. This arrangement is CA 022347~4 1998-04-14 somewhat similar to that of Fig. 6, except that it is not necessary to divide QO into two parts.
While preferred embodiments of the invention have been described, it will be realized that various changes may be made within the 5 scope of the invention. For example, other multipole rod-type mass spectrometers may be used, e.g. octopoles or hexapoles. Rod-type mass spectrometer means a mass spectrometer which has a number of rods, which are usually but not necessarily parallel, and which are usually but not necessarily linear.

Claims (9)

1. A method of operating a mass spectrometer comprising:
trapping ions in a first multipole rod-type mass spectrometer, applying an auxiliary field to said first mass spectrometer to remove all ions therefrom except for parent ions, fragmenting the parent ions in said first mass spectrometer to produce fragment ions, and transmitting said fragment ions from said first mass spectrometer into a second and mass resolving mass spectrometer.

2. A method according to claim 1 and including the step of detecting a selected fragment ion passed through said second mass spectrometer.

3. A method according to claim 2 when said auxiliary field is a filtered noise field.

4. A method according to claim 2 when said auxiliary field is a dipolar field.

5. A method according to claim 2 when said auxiliary field is a quadrupolar field.

6. A method according to claim 2 wherein after ions are removed from said first mass spectrometer except for said parent ions, said parent ions are transferred into said second mass spectrometer and are trapped there, and are then injected back into said first mass spectrometer with sufficient energy to fragment said parent ions in said first mass spectrometer.

7. A method according to claim 6 wherein said first mass spectrometer is operated in RF-only mode while receiving ions from said second mass spectrometer.

8. A method according to claim 7 including the step of admitting gas into said second mass spectrometer to facilitate trapping said parent ions in said second mass spectrometer, and then after said ions are injected from said second mass spectrometer back into said first mass spectrometer for fragmentation, a substantial portion of said gas is removed from said second mass spectrometer prior to transmitting said fragment ions into said second mass spectrometer.

9. A method according to claim 1 wherein said first mass spectrometer comprises first and second rod sets, the pressure in said first rod set being maintained at a higher level than that in said second rod set, said ions being transferred through said first rod set to said second rod set and being trapped in said second rod set and said auxiliary field being applied to said second rod set to remove all ions therefrom except for said parent ions, said parent ions in said second rod set then being injected therefrom back to said first rod set with sufficient energy to fragment said parent ions in said first rod set to produce said fragment ions.