US 3803410 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
United States Patent 1191 1111 3,803,410 Banner 1 Apr. 9, 1974 MEANS FOR PRODUCING ELECTRICAL  References Cited MARKER PULSES UNITED STATES PATENTS  Inventor: Aubrey Edward Banner, Sale, 2,624,783 1 1953 Medzel 324/42 England 3,244,876 4/1966 Kanda et a1... 250/41.9 D 2,624,783 1 1953 N d 1 [731 Asslgnee= Asswaied Elecil'lcal Indusmes 3,244,876 4/1966 kznc ig et al 250/214 Limited, London, England OTHER PUBLICATIONS  Filed: Mar. 26, 1970 Caldecourt et a1., Revlew of Sc1ent1c Instr., Vol. 25,  Appl. No.: 20,481 No. 10, 1954, pp. 953-955.
Related US. Application Data  Continuation of Ser. No. 607,412, Jan. 5, 1967, Prlrrlary Examlt'er lames Lawrence abandoned. Asszstant ExamznerB. C. Anderson  Foreign Application Priority Data  ABSTRACT Jan. Great A ma pectrometer in a magnetic analyzer may be used for scanning ions during operation. A U-S- 0 means is provided to measure the magnetic flux den 2 50/298 sity in the magnetic analyzer and utilize such measure-  Int. Cl. H01 39/34 ment in the determination f the mass/charge ratios f  Field of Search 324/42, 47; 250/419 ME, ions 22 Claims, 5 Drawing Figures PECOIZDER IZ/ INTEGRATOR 5 CIRCUIT t 26 seam/Ne CIRCUIT i MAB/ 52 MEANS FOR PRODUCING ELECTRICAL MARKER PULSES CROSS REFERENCE TO RELATED APPLICATIONS 1. This application is a continuation of Ser. No. 607,412, filed Jan. 5, 1967, now abandoned.
2. Application Ser. No. 571,049, filed Aug. 8, 1966,
by B. N. Green under the title Improvements in or Re- BACKGROUND OF INVENTION 1. Field of the Invention Mass Spectrometers are used to analyze substances both qualitatively and quantitatively. In the typical mass spectrometer, an ion source liberates ions from a sample and electrically propels the liberated ions through an exit slit. The ions typically are liberated either with an electron beam or a spark source.
After the ions pass through the exit slit, they are focused on a suitable collecting and/or detecting device. In a so-called single focusing mass spectrometer, an e'lectro-magnet is usually used to establish a magnetic field. Electrons passing through the exit slit of the source travel in an evacuated path through the magnetic field where they are deflected by the field. In a so-called double-focusing mass spectrometer, the ion path is first through an electrostatic analyzer and then into the magnetic analyzer.
The magnetic analyzer deflects ions according to their mass/charge ratios. In other words, the amount of deflection are of an ion as it travels through the magnetic field depends upon both the mass of the ion and the charge of the ion. For any given propelling force from the ion source and deflecting force in the analyzer or analyzers employed, only ions having a specific mass/charge ratio will pass into the detector. Ions of higher or lower mass/charge ratios will either be deflected more or less than the amount of deflection appropriate to focus the ion on the detector.
In analyzing a mass spectrum of a given sample, it is customary to utilize a technique known as scanning. With scanning, the mass spectrometer is adjusted over a range of values so that substantially all ions present in a given sample under study will at one time or another be focused on the detector. In this manner the detector can measure and record each type of ion present so that the operator can both identify the types of ions present and determine the quantity of each type.
There are two types of scanning which have been employed in the past. The first type is so-called voltage scanning where the accelerating voltage used in the ion source and the electrical potential used in the electrostatic analyzer are maintained in a fixed ratio and varied together. The second type of scanning is so-called magnetic scanning. Magnetic scanning is employed in both single and double focusing instruments and is the type of scanning to which this disclosure is directed.
As will be explained more completely below, the mass/charge ratio of those ions which are focused on the collector is directly proportional to the square of the magnetic flux density in the magnetic analyzer. The magnetic flux density is proportional to the current passing through the coils of the electromagnet apart from hysteresis effects. In the past it has generally been assumed that because the current in the magnetic coils and the magnetic flux density are proportional, the mass/charge ratio of a focused ion was proportional to the square of the current supplied to the magnetic coils. While this is theoretically correct, apart from hysteresis effects, in practice it has proved to introduce inaccuracies into the analysis of any given mass/charge spectrum.
2. Description of the Prior Art Prior to the two above-referenced co-pending applications, the fact that the magnetic flux density was not necessarily directly proportional to the current in the magnetic coils, apart from hysteresis effects, has not been generally recognized. As noted in those referenced applications, there are discontinuities which occur in the magnetic flux. For example eddy currents are generated which distort the magnetic flux density with the result that variations in the magnetic flux density are not directly proportional to the current applied.
SUMMARY OF THE INVENTION With this invention, a mechamism is placed in the magnetic flux field which is responsive to the magnetic flux density, or variations therein. This device preferably takes the form of a flux detection coilwhich is mounted saddle-like along the ion tube. A change of magnetic flux induces a flow of current in the flux detecting coil, the total electrical charge induced in the circuit being directly proportional to the change of flux acting on the ion beam. The output from this coil is integrated by means of a novel integrating circuit and is used to determine, at any specific time, exactly what change from the initial value of magnetic flux has occurred. The initial magnetic flux is represented by a suitably adjusted, constant potential, and this is added algebraically to the integrated current output from the coil to give the actual magnetic flux at any instant. The resulting potential is fed into a novel squaring circuit where it is squared and subsequently fed into a novel mass marker circuit. The output of the mass marker circuit is then conducted to a recorder. The recorder also receives the output of the mass spectrometer detector. By utilizing both recorded signals, that is the signal from the mass marker and the signal received from the mass spectrometer detector, it is possible to determine the identity of any given ion and the quantity of any given type of ion present with an accuracy well beyond that which has been obtained with prior art mechanisms and techniques.
Accordingly, the objects of the invention are to provide a novel and improved mass spectrometer equipped with a novel and improved detecting and recording mechanism including novel and improved integrating, squaring and mass marker circuits, and improved methods of operating a mass spectrometer.
Other objects and a fuller understanding of the invention may be had by referring to thefollowing description and claims taken in conjunction with the accompanying drawing:
DESCRIPTION OF THE'DRAWING FIG. 1 is a sectional side elevation of a mass spectrometer;
FIG. 2 is a circuit diagram of ancillary electrical equipment providing a progressively varying signal indicative from instant to instant of the mass number reached during a scan of the mass spectrum;
FIG. 3 is a circuit diagram of a preferred alternative form of ancillary equipment for providing such signal;
FIG. 4 is a circuit diagram of means in accordance with the invention for providing electrical marker pulses as said signal varies during a scan; and,
FIG. 5 is a fragmentary sectional view of parts of the magnetic analyzer as seen on an enlarged scale from the plane indicated by the line 55 of FIG. 1.
DESCRIPTION OF THE PREFERRED EMBODIMENTS The mass spectrometer illustrated in FIG. 1 is of a double-focusing type. That is, a mass spectrometer in which ions are both electrostatically and magnetically focused. The spectrometer has an ion source chamber 1 into which a specimen carrying probe 3 can be inserted. The specimen carried by the probe 3 or altematively a gaseous sample introduced to the chamber in a conventional manner is ionized with apparatus well shown in the art.
An electrode 5 to which an accelerating voltage of, say, 8 kilovolts is applied, repels ions liberated from the source. The ions are repelled as a beam which passes first through an electrostatic analyzer 7.
The electrostatic analyzer includes opposed conductive plate 7? between which a potential difference V is maintained. The ion beam then passes through a slit defined by a monitor collector 9 and into an ion tube 10. Ions in the tube 10 pass through a magnetic analyzer 11, and into a detector shown generally at 12.
In the magnetic analyzer 11 a pair of electromagnet coils 11A, 1 18, FIG. 5, establish a strong magnetic field B. The field B is directed in a direction transverse to the path of the ions. Since the ions are charged particles their paths will be curved in the magnetic field. The angular deflection of an ion in passing through the magnetic analyzer 11 will depend upon the accelerating voltage, since that determines the speed of the ion, upon the intensity of the magnetic field in the analyzer l1, and upon the mass/charge ratio of the ion. The deflected ions focused on the detector 12, or certain of them if different groups of ions are deflected to different degrees, pass through an adjustable slit in a member 13 which forms a part of the detector. In the detector, ions passing through the slit in the member 13 are picked up by a collector electrode 15 associated with an electron multipler 17.
The output from the electron multiplier 17 is used after amplification in an amplifier 19 to provide a record of the ions passing through the adjustable slit in member 13.
The mass spectrometer as described above is evacuated in a suitable manner as by pumps 18, 18a. The structure thus far described is well known in the art.
One method known of scanning a range of a mass spectrum is so-called magnetic scanning. With magnetic scanning the voltages used in the electrostatic analyzer 7 and on the accelerating electrode 5 are maintained constant. Scanning is accomplished by slowly decreasing the current used in the electromagnet coils 11A, 11B of the magnetic analyzer 11. This progressively changes the strength of the flux B and therefore the deflections of all the ions passing through the magnetic analyzer.
Since the output from the electron multiplier 17 indicates the number of ions passing through the slit member 13 at any given time, it is possible to identify those ions if the output of the multiplier is properly correlated to the conditions of the mass spectrometer at that given time. Thus, when a signal stimulated by the electron multiplier 17 is presented on a cathode ray tube as the vertical deflection with a horizontal scanning speed corresponding to the decay of the magnetic field in the magnetic analyzer 11, the trace shows peaks when ions are present having such a mass number that they are deflected to pass through the slitted member 13.
Mass analysis depends upon the identification of the peaks produced by the specimen during the scanning of the mass spectrum. It is advantageous to provide, on the record of the peaks, markers corresponding exactly to known mass/charge ratios. The apparatus described below provides this advantage. Since the charge is almost always unity, the mass/charge ratio is referred to herein merely as the mass.
In order to ascertain the number representing the mass which is instantaneously being recorded, it is necessary to know accurately the value of B, the magnetic flux density to which the ion beam is subjected, and V, the accelerating voltage which determines the velocity of the ions as they pass through that magnetic field. The mass M is given by where k is a constant for the particular mass spectrometer and can be determined by doing an analysis of a known specimen.
If V is held constant when the mass spectrometer is scanned, M is directly proportional to B. If the precise value of B is known M can be determined and therefore the ions detected can be identified.
As noted above, the value of B should theoretically vary in direct proportion to the current applied to the magnet coil 11. But in practice, there are other factors such as hysteresis effects and generated eddy currents which distort this relationship of the flux B to the applied current.
With the present invention the actual flux acting on ions in the tube 10 is measured. In the preferred arrangement, a flux variation detection coil PC, FIGS. 1 and 5, is positioned in juxtaposition to the tube 10 within the coil 11. The variation in the flux B induces a current in the flux variation detection coil PC which is fed to an integrator circuit 26 to a squaring circuit 27 and then to a mass marker 21.
Since (a) the integrated output of the coil PC is directly proportional to the change of flux B and (b) the actual magnetic flux is proportional to the sum of the integrated coil output and the constant potential (0) the mass M is directly proportional to B and (d) the output of the squaring circuit will be directly proportional to the mass M. Accordingly, the ions can be identified with a precision and accuracy far exceeding the capabilities of prior mechanisms.
As shown in FIGS. 1 and 5, the coil PC rests saddle fashion on the tube 10. The coil PC has elongated central portions 22A, 22B disposed symmetrically about the axis of the tube 10. The coil has end portions 23A, 238 which pass around the tube 10 to connect the portions 22A, 22B.
If a larger induced current is desired the coil PC may be in the form of a coil or coils wound on one or both of the coils 11A, 11B of the magnetic analyzer 11 or around the actual pole pieces of the magnet. The coil PC shown in the drawings is preferred because it most accurately reflects the flux B actually acting on the ion beam in the tube 10. v
In FIG. 2 another structure for the value of B is shown. I-Iere Hall Effect probes 24A, 24B are placed side by side on a common substrate 25 to insure that they are at substantially the same temperature. As an alternative to the coil PC the substrate 25 with the two probes 24A, 24B is placed in the gap of electromagnet 11 at the position indicated by the dashed lines in FIG. 1. Thus, both probes experience the samevalue of magnetic flux density B as the ions passing through the analyzer.
Each Hall Effect probe 24A, 24B is in the form of a thin rectangular plate of semiconductor material carrying an electric current in the direction of its length.
Since each plate is subjected to a magnetic field which is normal to it, an electromotive force is developed which is at right angles both to the direction of the current and to the magnetic field.
Referring now to FIG. 2, a direct current of constant magnitude 1, is causedto flow from a positive pole 3.1? through the length of the Hall Effect probe 24A to the negative pole 31N. The output voltage u, of probe 24A where K, is a constant related to the characteristics of the probe 24A. One output terminal of probe 24A is connected directly to an input terminal of an amplifier 33, and the other output terminal is connected through a resistor R to the second input terminal of the amplifier 33. An output current 1 of the amplifier 33 is applied through the length of the Hall Effect probe 24B to the second input terminal of the amplifier 33. The amplifier 33 is of a high gain type having a relatively high input impedance. The amplifier 33 give an output current 1 proportionally related to the input voltage u, by the expression:
Since the probe 243 is subjected to the magnetic field B and to a lengthwise current of 1 its output voltage u is given by:
2 Kg 12 B where K is a constant, so that:
2 =(K1K2 B2 Thus, so long as I, is also constant, 14 is proportional to B As noted above, if V, the accelerating voltage, is maintained constant during a scan, then B and therefore u are porportional to the mass,'in which event a could be taken as the input to the mass marker to be described.
If Vmay vary, however, a voltage V derived proportionally from it is applied through a resistor R1 to an electromagnetic coil 41 associated with another Hall Effect device 43. The electromagnetic coil 41 produces a magnetic field having a magnitude H in a direction which is normal to the semiconductor plate of the Hall Effect device 43. A current I, is passed lengthwise through the plate of the Hall Effect device 43, flowing from a positive supply pole 45?, through a resistor R2, a current controlling device 47, and the plate to the negative supply pole 45N. The output voltage V out of the Hall Effect device 43 is applied in opposition to the voltage u to provide a net voltage v applied as an input to an amplifier 49. The output from the amplifier 49 is applied through a lead 51 as a control signal for the current controlling device 47.
The current I flowing in the coil 41 is given by IC VA/RI where R is large, and the voltage V developed across the resistor R by the flow of current I is given by:
Also, the output voltage V of the device 43 is given by:
where K is a constant.
The action of the amplifier 49 is to provide an output signal in lead 51 which will vary current I in a manner tending to eliminate any difference voltage applied to the amplifier 49, so that in effect I is constantly adjusted to make V equal to U It follows that, since whence I is proportional to 8 V (I' being constant) so that for a given value of R with V proportional to V:
V is proportional to 8 V The voltage V is therefore proportional to mass M and is applied to the marker circuit (FIG. 4).
In the preferred structure, current I is generated in the coil PC proportional to the change of magnetic flux density with time. That is, I ozdB/dt. The principle on which the arrangement of FIG. 3 functions is that the current I is integrated to give an output voltage V, indicative of the field strength reached at time t relatively to its strength at the beginning of the integra- Referring to FIG. 3, an integrator circuit is designated generally by the numeral 26. The current I from the pick-up coil PC is fed to an integrating amplifier A1 having an integrating capacitor C1 connected between its output and input. The output voltage from this amplifier is combined with a calibrating voltage, derived from a potential dividing chain including variable resistor VRl at the input of an operational amplifier A2, the output voltage V of which will be proportional to the magnetic flux density B of the magnetic analyzer 11. The integrating amplifier A1 is chosen as having low input offset voltage and current drifts. Constant input offset voltages and currents can be cancelled out by adjustment of a variable resistor VR2 included in a further potential dividing chain and connected to the amplifier input through a high resistance R2. Variations of input offset voltage and current can be reduced by placing the integrating amplifier A1 in a temperature controlled ambient.
Significant drifts in the output of the integrating amplifier A1 can occur because of leakage of charge within the integrating capacitor C1. To counteract this, a current 1 equal and opposite to the leakage current I is derived by applying the output voltage V, of the integrating amplifier A1 to the input of a unity gain inverting amplifier A3 an adjustable portion of the output of which is fed back, from a potential dividing chain including variable resistor VR3, to the input of the amplifier Al via a high resistance R3. A resistor R4 is connected in series with the inverting amplifier A3, and a resistor R is connected across the inverting amplifier A3.
The setting up procedure for the integrating amplifier Al involves connecting the input of the integrating amplifier A1 to ground by means of switch S3; with V approximately zero, adjusting VR2 for zero drift; applying a positive input to the integrating amplifier Al to set up a voltage across the integrating capacitor C1; and adjusting VR3 until there is no drift in this last mentioned voltage. In this condition, a current 1,, which will be equal and opposite to the capacitor leakage at all voltages across it is being applied to the input of the integrating amplifier A1. Only changes in leakage current are then significant and these can be minimized by maintaining a constant ambient temperature. Further improvement can be achieved by connecting a temperature sensitive resistance, such as a silicon resistor, in series with the resistors R4 and R5, depending on the signs of the temperature coefficients of the temperature sensitive resistance and the insulation resistance of the integrating capacitor C1. The effect of this extra component, which can be adjusted by means of a low temperature coefficient resistance in parallel with it, is made equal and opposite to that due to variation with temperature of the insulation resistance of the capacitor C1.
The output voltage V aB from the integrator circuit 26, and a voltage V representative of the accelerating voltage are applied (by way of one section (a) of a mu]- ti-position switch S2 in the case of V as three inputs V V V V and Vy V, of a multiplier divider circuit which in general produces an output voltage proportional to V 'V /Vy and in the present case acts as a squaring circuit producing the output voltage Vm aV /V (or 01V if the accelerating voltage V is main tained constant during a scan). The principal on which the squaring circuit 27 functions is that a time interval T proportional to V Vy V /V is defined by the part of the circuit comprising operational amplifiers A4, A5 and A6, while an integration is being performed on V V by an integrating amplifier A7. The integration is stopped at the end of the time T. The output voltage V of the integrating amplifier A7 at any time t during the integration, and the output V taken from a following unity gain amplifier A8, are therefore given by:
so that on termination of the integration at time t V aVZT= V T but T (XVX/ Vy VB/VA V KVx'V /Vy K V V,, where K is a constant.
In practice it is convenient to operate with a constant acceleration voltage and to adjust the squaring circuit parameters so that V 0.1 V The output voltage V which is taken to the marker circuit of FIG. 4 through a second section (b) of the multi-position switch S2, will then be 10 volts when V is 10 volts.
Considering the squaring circuit 27 in greater detail, the operational amplifier A4 and the capacitor C2 constitute an integrator circuit to which is applied a source of potential V proportional to the accelerating voltage V or constant if V is constant. A transistor TRl connected across the integrating capacitor C2 is switched on and switched off under control of a bistable circuit BS. When TRl is switched on, the integrator A4 is reset by discharge of capacitor C2, the output of the integrator A4 in this reset condition being adjusted to be 0 volts, to within lmV, by means of atrim potentiometer connected to the integrator A4, not shown, connected to the integrator A4. When transistor TRl is switched off by the bistable circuit BS, a linear integration immediately commences, the output of amplifier A4 going negative with respect to 0 volts.
This output potential is algebraically combined with V,, at the input of the operational amplifier A5. Amplifier A5 is operated as a simple unity gain inverting .amplifier with negative feedback through a resistor R10. It is also possible to operate the operational amplifier A5 with a gain greater than unity, if desired. Thus, as the output of the operational amplifier A4 goes progressively more negative from zero during the integration, the output of the amplifier A5, which starts at a potential equal and opposite to V progressively rises towards 0 volts.
The amplifier A6 with its associated resistors R6-R9 and limiting diodes D1, D2, constitutes a level detector trigger circuit 30, the trigger level being set to 0 volts by means of a variable resistor VR4. This amplifier can be a dc. amplifier with such high gain that except at virtually zero input voltage it gives the full output voltage permitted by the limiting diodes. At the beginning of an integration by amplifier A4, when the output of the operational amplifier A5 is negative, the output potential of the level detector 30 is positive with respect to ground. When the output of the operational amplifier A5 reaches the trigger level of zero volts (i.e., when V Ta V and Ta V V the output of the level detector' circuit 30 switches to a negative potential, and a corresponding positive potential change is applied to the bistable circuit BS by a pulse shaping circuit PS. This edge triggers the bistable circuit BS into its SET 1) state, and the transistor TRl is thereby switched into its saturated state to reset the amplifier A4.
A transistor TR2 is also switched off when the bistable circuit changes to its SET (1) state. This stops current flow into the integrating amplifier A7 and thus stops the integration of V Transistors TRl and TR2 remain in these conditions until the bistable circuit BS is later RESET by a pulse from a monostable circuit MS2. When this occurs a new integration is commenced.
As an integration of V, proceeds in the integrating amplifier A7, the output of the integrating amplifier A7 goes increasingly negative until the input current is cut off at TR2, that is, when the bistable circuit BS is in its SET (1) state as a result of the level detector 30 being triggered. Thus, the time for which the integrator is allowed to run is determined by the operation of the bistable circuit and has been seen to be proportional to V,,/ V
When the bistable circuit goes into its SET state at the end of the integration, it applies a leading edge signal to a monostable circuit MSl, which in its turn switches on a transistor TR4. The final output potential (V V of the integrating amplifier A7 is thus applied to a storage condenser C3. The allowed time for this read-out operation is determined by the time constant of the monostable circuit MSI. When MSl reverts to its stable (0) state, the transistor TR4 is switched off, and a monostable circuit M82 is triggered to its SET 1) state. This switches on transistor TR3 for a time determined by the time constant of the monostable circuit M82, the second integrator thus being reset to give 0 volts at the output of A7. When the monostable circuit MS2 reverts to its stable state, it applies a potential edge to reset the bistable circuit BS, and the whole cycle is repeated continuously.
The output potential which is stored by capacitor C3 is applied to the input of the differential operational amplifier A8, which is connected in such a manner to constitute a unity gain, non-inverting follower circuit with very high input impedance. Thus, capacitor C3 does not discharge significantly during the time of an integration, and low impeda'nces may be driven from the low output impedance of the operational amplifier A8.
Temperature compensation in the squaring circuit can be provided by including a temperature sensitive resistor Rla, Rlb having a very low temperature coefficient and being a much larger resistance than Rla.
The operation of the squaring circuit can be analyzed as follows:
Let V, output potential of the amplifier A4 with respect to 0 volts at time t, point 0 being a virtual earth point I input current through Rla Rlb R1 Since the output of amplifier A4 is at 0 volts at the commencement of the integration But I potential across R /R, "V /R Hence V, V 't/R C (1) But V, V when t= T VB V4'T/R1'C] T: VB/VA i' r Let V output potential of A7 with'respect to zero volts at time I, point X being a virtual earth point R total input resistance to A7 Then V V -t/R"C (as for V in (1) above) Therefore, at time T:
V0 VB'TIRICZ Substitute for T from (2) in (3) Hence V3/R"C2 VB/VA R 'C R C /R'C VB2/VA V k'V /V where k is a constant and equals R 'C,/R"C
The squaring circuit just described has been found to be capable of operating within range of error as low as 0.01 percent at a full scale output of 10 volts, with maximum error at reduced outputs between full scale and 10 percent of full scale being approximately 1 3mV and with an operating time for a squaring action of 1 millisecond. The circuit is therefore much more accurate than other known squaring or multiplying circuits the accuracies of which are in general limited to about 0.1 percent of full scale output unless large voltages,
e.g., of the order of volts, are employed. In a transistorized system such large voltages would often be excessive.
The voltage V a V V derived in the circuit of FIG. 2 or FIG. 3 gives from instant to instant an indication of the mass of the ions which are being intercepted by the collector electrode 15 (FIG. 1). Although one could apply the output voltage V and the output from the collector electrode 15 to some device such as a computer to obtain a mass analysis, in practice it is very expensive to provide analogue-digital peripheral equipment which can deal with the two very rapidly varying voltages in an on-line manner to feed a digital computer. The apparatus of FIG. 4 is therefore used to provide marker pulses which accurately indicate certain masses as ions of those masses are being collected by the collector electrode 15.
Referring now to FIG. 4, which is designed for a decreasing mass scan, the function of this circuit is to provide an output pulse on an output lead 61 as the voltage V (representing the mass instantaneously being mea sured by the collector-electrode (15) passes each fifth mass number. The circuit is centered on a difference amplifier 63, which constitutes a signal comparator and in the preferred embodiment shown is a high gain operational amplifier with an input noise and drift better than SOuV for 1 in 10 accuracy at mass 50. The gain of the difference amplifier 63 is arranged 'to be 100 by the provision of a feedback loop having for example a resistance 64 of 1.5 megohms. With the resistance 64 at this value, a pair of amplifier input resistors R5 and R6 are of 15 kilohms each. The voltage V is applied as a negative voltage to the difference amplifier 63 throughthe resistor R5, and the second input V, to the difference amplifier 63 is applied through a unity gain, high input impedance amplifier A9 from a digital potentiometer 69.
The digital potentiometer 69 consists of a series chain of nineteen resistors, designated by numerals RAl-Rll, RAZ-RIZ, and RJ respectively, connected in that order with the outer end of resistor R] connected to ground and the outer end of resistor RAl connected to an accurately maintained source of about 10 volts. Of these resistors, the resistors RAl, RAZ and RJ have a resistance of 0.5R ohms, resistors RBI, RB2, RDl and RD2 have a resistance of IR ohms, resistors RC1 and RC2 have a resistance of 2R ohms, resistors RBI and RE2 have a resistance of SR ohms, resistors RFl, RF2, RHI and RHZ have a resistance of 10R ohms, resistors RG1 and RG2 have a resistance of 20R ohms, and resistors RI] and R12 have a resistance of 50R ohms, where R is a constant chosen such that the potentiometer will not be significantly loaded by the input impedance of the amplifier A9: R may for'example be 10 ohms.
With the exception of the resistor R], each resistor is shunted by a switch having the same reference as the resistor but with the prefix R replaced by S e.g., resistor RAl is shunted by Switch SAl. Each of these switches is a mercury wetted reed switch having an associated 5 operating coil which again has the same reference but with C as its prefix letter. The operating coils CAI-Cll for switches SAl-Sll are energized in the RESET conditions of bistable circuits BA-BI respectively. Coils CA2-CI2 for switches SA2-SI2 are respectively energized by the inverted outputs of the bistable circuits BA-BI, that is, in their SET (1) conditions. Thus with the bistable circuits all in their RESET (0) conditions, switches SAl-Sll are closed and switches SA2-Sl2 are open, this being the condition illustrated.
The bistable circuits BA-BI are so interconnected that pulses applied to an input lead 79, being also the pulses fed out on the lead 61, cause the switches SAL-- SA2 to be opened and closed according to a predetermined sequence to produce in a potentiometer output lead 81, which is connected from the junction of resistors RH and RA2, a step-wise varying voltage. In the particular arrangement shown, the bistable circuits BA-BI are interconnected as a cyclic counter comprising a division-by-two section (BA followed by two decadic binary sections (BB-BE) and (BF-BI). Each bistable circuit alternates between its l and 0 conditions in response to successive input pulses delivered to it (except in the case of BAD on successive reversions of the immediately preceding bistable circuit to its 0 condition. The following tables set out the resulting sequential operation of the switches. In these tables the 0 state corresponds to the first switch of the pair, e.g., the SCI switch, being closed, and the l state corresponds to the second switch of the pair, e.g., the C2 switch, being closed. It will be seen from the tables that the first two sections behave as a binary counter except that at a count of 8 a feedback connection is arranged to step the (0 0 0 l 0) state to a (0 l l l 0) state, and that at a count of 18 the (0 0 0 l 1) state is stepped to (0 l 1 l l In the second decade section similar stepping takes place at counts of 40 and 90.
Output Pulse SAl/Z SBl/2 SCI/2 SDI/2 SEl/2 Volts 00 0 O 0 0 O 10.00 01 l 0 0 0 0 9.95 02 0 l 0 0 0 9.90 03 l l O 0 0 9.85 04 o o 1 0 0 9.80 05 l 0 l 0 0 9.75 06 0 l l 0 0 9.70 07 l l l O O 9.65 08 0 0 0, l 0 xxxx stepsto 0 l l l 0 9.60 09 l 1 l l 0 9.55 10 0 o 0 0 1 9.50 ll 1 0 0 0 l 9.45 12 0 l 0 0 l 9.40 13 l l 0 0 l 9.35 14 0 0 l 0 l 9.30 15 l 0 l 0 l 9.25 16 l I l 0 l 9.20 17 l l l O l 9.15 6O 18 0 0 0 l l xxxx slepsto 0 l l l l 9.10 19 l l l l l 9.05 20 0 0 O 0 0 9.00
These sections now having returned to their original state, they will commence a similar cycle, and will continue to recycle every 20 input pulses. At the 20th pulse, i.e., on returning to the (0 0 0 0 0) state, a carry pulse is passed to the second decade section.
Carry Pulse pulse No. SFl/Z 561/2 SI-Il/Z SIl/2 Output 0 000 0 0 0 0 10.00 1 020 l 0 0 0 9.00 2 040 0 l 0 O 8.00 3 060 1 l 0 0 7.00 4 080 0 O l 0 xxxx steps to 080 l l l 0 6.00 5 0 0 0 l 5.00 6 l 0 0 l 4.00 7 0 l 0 l 3.00 8 l l 0 l 2.00 9 0 0 l l xxxx steps to 180 l l l l 1.00 10 200 O 0 0 0 xxxx This section has now returned to its original state.
It will be seen that throughout the complete cycle the total effective resistance of the potentiometer 69 remains constant, facilitating the task of keeping the supply voltage to it steady. It can also be seen that the voltage of the output lead 81 relative to ground is decreased from an initial 10 volts in steps of 0.05 volt down to a minimum value of 0.05 volt.
The potentiometer 69 is provided with a reset switch RS which, when closed, resets all the bistable circuits to their 0 states.
The output from the difference amplifier 63, representing the decreasing difference between V and V from instant to instant during a scan, is applied to a level detector 85 which preferably is a high gain, low drift d.c. amplifier, being similar to the level detector A6 in the squaring circuit 27 in FIG. 3. It gives an output pulse only when the output of the difference amplifier 63 goes through the preset triggering level of the level detector 85, which in this case is 0 volts. The difference amplifier 63 and the level detector 85 both have a sign inverting action. Consequently with V and V applied respectively as negative and positive amplitudes, and with V always representing a future value of V because of the repeated stepping of V to progressively lower values, the output from the difference amplifier 63 will be positive and decreasing, and the level detector 85 will be producing a negative output clamped to a level of, say, 3 volts. When the output from the difference amplifier 63 goes through zero, the output from the level detector 85 abruptly changes to a positive value and the pulse edge represented by this positive-going change is sharpened up and inverted by a pulse-shaping circuit PS1. The pulse-shaping circuit PSi accordingly changes one input of an AND gate 75 from a positive potential to a negative gate-opening potential. During normal operation, a switch 73 is applying a negative potential also to the second input of the AND gate 75, which in response to the coincident application of these two negative input potentials feeds an output pulse by way of a second pulse-shaping circuit PS 2 to a monostable circuit MS3. This changes the monostable circuit MS3 to its unstable state and in doing so causes a pulse to be sent out over the lead 61. The pulse also passes over the lead 79 to step the digital potentiometer 69 to its next setting as previously described, thereby reducing the reference potential Vp by 0.05 volts. Any pulses that may be produced from the level detector 85 while the potentiometer 69 is settling down in its new setting are prevented from producing a pulse on the output lead 61 because of the presence of the monostable circuit MS3, which cannot again respond until it has returned to its stable condition. Thereafter, the action is repeated as V continues its decrease towards the new value of V As a result of this repeated action, a pulse is obtained on the output lead 61 at intervals of 0.05 volts in the decreasing value of V and these pulses are used in the mass spectrometer output recorder (FIG. 1) to produce markings on the record at corresponding intervals each representing mass numbers.
The voltage V in the manner described above, provides an accurate indication of the massnumber of the ions reaching the collector electrode 15 from instant to instant during a scan. However, the starting point in the scanning of the mass spectrum will depend upon the values of the current in the electromagnet coil 1 1C and the voltage applied to the ion source, and is not exactly predictable. In order to set the apparatus up at the start of a scan, the RESET switch is operated and puts the digital potentiometer 69 into the condition shown in FIG. 4, i.e., giving its maximum output of volts. The other input to the difference amplifier 63 can only be a voltage less than this the maximum value of V also being 10) and so the output of the difference amplifier 63 will be at a negative potential with respect to ground. Thus, the output of the level detector 85 will be positive, the input to the AND gate 75 from the pulse-shaping circuit PS1 will be negative, and the AND gate 75 will be opened. If the switch 73 is changed over a series of clock pulses will be applied from a free-running multivibrator 70 to the opened gate 75 and will cause stepping of the digital potentiometer 69 until equality has been reached in the two inputs to the difference amplifier 63. When this point is reached, the level detector 85 will invert its output and, by removing the negative input to the AND gate 75 will render the AND gate 75 ineffective so that no more clock pulses will reach the digital potentiometer 69 and it will remain in the setting which it has then reached. 7
This initial setting will usually take place in a very short space of time, and the operator can then restore the switch 73 to isolate the clock multivibrator 70, which serves no further function. The scanning action can then be initiated, with V correspondingly decreasing from its initialvalue. v
To calibrate when using the FIG. 2 arrangement for providing the mass indicating voltage V the magnetic field of the mass spectrometer is adjusted so that the prominent peak at mass number =502 in a reference sample of heptacosafluorotributylamine is tuned in at thecollector electrode of the mass spectrometer, and the'variable resistor VR2 (in FIG. 2) is set so that a display connected to the mass marker unit (FIG. 4) indicates 502. It is now certain that during a future analysis when the mass number scanned is 502 a marker pulse will be produced, and furthermore that in each direction from this marker pulse other marker pulses will be produced at accurately known mass number intervals of 5. v
In using the arrangement of FIG. 3, the integrator circuit 26 is initially set up as follows. With the switch S3 connected to ground, an input of 0 volts is applied to the integrating, amplifier A1 and the variable resistor VR2 is then adjustedto eliminate any drift of the amplifier output from zero. A suitable potential is then set up across the capacitor C1 by changing the switch S3 to its intermediate position and the variable resistor VR3 is adjusted until there is again zero drift. The squaring circuit 27 is set up by first putting the switch S2 into its position 2. This sets the output of the digital potentiometer 69 to its maximum 10 volts and applies this output over the lead 69' to the input of the squaring circuit 27. With V constant, the output of the squaring circuit 27 should then be 10 volts 0.1 X 10 The input and output voltages are compared through equal resistance e.g., of kilohms each and the overall gain is adjusted by means of the variable resistor VR6 until these voltages are equal as indicated by a null on a meter M2. The switch S2 is then changed to its position 3 to set the digital potentiometer output to 2 volts 20 percent full scale) and apply it to the squaring circuit 27. This should give an output of 0.4 volts so that in comparing the output with the input through a 20K resistor and 100K resistor respectively a compensating adjustment can be made on a zero control member (not shown) in amplifier A8 until the meter M2 again reads zero. A check at an intermediate input voltage of 7 volts can be made with switch S2 in position 4.
To calibrate the apparatus after it has been set up as described, an analysis is performed on a reference substance and a specified peak is tuned in, e.g., the 502 peak of heptacosafluorotributylamine. The output from the integrating amplifier A1 is zeroed by short circuiting the capacitor C1 by switchv S3 and the calibrating potentiometer VRl is adjusted until the marker output corresponds to mass number 502. Switches S3 and S2 have previously been restored to their No. l positions, and the switch S3 has to be re-opened before a scan is commenced.
Switch S3 is re-opened and the magnetic field is readjusted to tune in another prominent peak in the reference sample spectrum, e.g., the 69 peak of heptacosafluorotributylamine. The display system attached to the mass marker unit should now read 69. Any discrepancyv can be corrected by suitably adjusting variable resistor VRS in the integrator circuit 26 and repeating scans from mass number 502 to 69 until correct readings are obtained.
If, due to small drifts in the integrator circuit 26, it is occasionally necessary to reset the integrator circuit 26, this can be done by retuning the 502 peak on the collector electrode 15, then shorting out. the integrator capacitor C l for a few seconds, and adjusting the variable resistor VRl .to give a reading of 502 on'the display system.
While the operation has been described in respect to decreasing field scan, similar arrangements could be used for increasing field scanning. The sequence switching of the digital potentiometer would have to be reversed and appropriate modifications made in other parts of the circuits to cater for reversed polarity of certain of the voltages. It is further contemplated that apparatus capable of being switched to deal alternatively with increasing or decreasing scans could be provided.
In order to facilitate reading of the marks produced on the recorder, provision can be made for each n mark to be made more prominent than the others. The counting action performed by the bistable elements of the digital potentiometer 69 can be utilized for this purpose. For instance, in order to produce a more prominent mark at each twentieth mass number, a lead 61 may, as shown, be connected to the 0 side of an additional bistable element driven from the 0 side of element BA to produce an output pulse on that lead on each fourth pulse which appears on leads 61 and 79. In the case of a galvanometer recorder, for instance, this output pulse on lead 61 could switch on a transistor in parallel with another transistor switch on by each pulse on lead 61 and feeding current to the galvanometer coil.
Although the invention has been described in its preferred form with a certain degree of particularity, it is understood that the present disclosure of the preferred form has been made only by way of example and that numerous changes in the details of construction and the combination and arrangement of parts may be resorted to without departing from the spirit and the scope of the invention as hereinafter claimed.
1. In combination with a mass spectrometer system including an ion source utilizing an ion accelerating voltage for producing a beam of ions and passing said beam into an ion tube, a magnetic analyzer for generating a flux field to differentially deflect ions in the ion beam in said ion tube according to the mass numbers of said ions, and ion detector means for detecting ions so deflected, the improvement comprising:
a. a stationary magnetic flux detection coil at least partially surrounding said ion tube and located in the magnetic analyzer flux field for producing an output responsive and proportional to variations from an initial value in the magnetic flux density in the magnetic analyzer;
b. an integrating circuit connected to receive the output from the magnetic flux detection coil, said integrating circuit comprising an integrating amplifier having an integrating capacitor, means for counteracting drift in said amplifier by injection of a variable compensating input, and means for counteracting the effects of leakage in said capacitor by injecting into it a replenishing current substantially equal and opposite to any leakage current, said integrating circuit producing an integrated output indicative of variations in said magnetic flux density.
2. The system of claim 1, wherein the means for providing said replenishing current is an inverting amplifier connected to receive the output voltage of the integrating amplifier as its input and a variable potential divider connected to the output of the inverting amplifier for feeding a proportional current into the integrating capacitor opposite to the direction of leakage current therein.
3. The system of claim 1, wherein said detection coil rests saddle-fashion on said ion tube.
4. The system of claim 1, wherein the improvement further comprises a multiplier and divider circuit connected to receive the ouput from said integrating circuit, said multiplier and divider circuit producing an output; and a marker circuit connected to receive the output from said multiplier and divider circuit, said marker circuit being adapted to produce output pulses at intervals corresponding to mass number differences of a predetermined number of units of mass as the output from the multiplier and divider circuit passes through predetermined values.
5. The system of claim 4, wherein said predetermined number is five.
6. The system of claim 4, wherein said marker circuit comprises:
a. a signal comparator connected to receive the first input and as a second input a reference signal stepped to a future value of the first input each time the first input reaches a predetermined relationship with the reference signal;
b. means for producing a pulse consequent upon detection of said relationship by the signal comparator; and
0. means for stepping the value of the reference signal by a predetermined amount to a future value of the first input consequent upon said detection.
7. The system of claim 6, wherein the signal comparator comprises a high gain operational amplifier provided with a negative feedback loop degrading the overall amplification of the amplifier to a fraction of the gain of the amplifier without feedback.
8. The system of claim 6, wherein the means for stepping the value of the reference signal comprises a digital potentiometer including a series chain of resistance elements each individually shunted by a separably operable switch, a reference signal output connection connected to an intermediate point in said chain, and means responsive to successive detections of said relationship for operating the switches in a predetermined sequence such that the total resistance between the ends of the chain remains constant but the distribution of this total resistance on opposite sides of the intermediate point is varied in a stepwise manner according to the required stepping of the reference signal.
9. The system of claim 8, in which the means for operating the switches includes a chain of bistable elements each controlling a pair of the switches, the switches of each pair controlling the switching of equal resistance elements which are respectively on opposite sides of the intermediate point and are alternatively operable as the bistable element for controlling the pair is in one or the other of its two stable states, and the series chain of resistance elements is being connected to receive stepping pulses concomitantly with the production of the marker pulses. 10. The system of claim 8, wherein the resistance elements in the series chain are so chosen in their relative resistance values having regard to the switching sequence that successive step changes of the reference signal'are of equal magnitude.
11. The system of claim 8, including further switchingmeans for setting the switches to at least one predetermined combination of operated and unoperated conditions thereof, thereby to produce ,the reference signal with a correspondingly predetermined value.
12. The system of claim 4, wherein said multiplier and divider circuit operates on quantities respectively represented by X, Y and Z voltages to provide a V output signal, said X and Z voltages being mutually equal and proportional to said output of said integrating circuit, said Y voltage being proportional to said ion accelerating voltage utilized in said ion source, said circuit comprising:
a. first circuit means for receiving said X and Y voltages and providing a T signal having a time duration substantially proportional to said X voltage, and substantially inversely proportional to said Y voltage;
b. second circuit means for receiving said Z voltage and said T signal and providing a Z signal having a rate of change substantially proportional to said Z voltage and a time duration the same as said T signal; and
0. output storage means for receiving and storing said Z signal at the termination of said T signal to provide said V output signal, which is substantially proportional to XZ/Y.
13. The system of claim 12, further including means in said first circuit means for receiving an X voltage and providing an X signal having a rate of change substantially proportional to said Y voltage, and means for providing said T signal having a time duration substantially proportional to X/Y whereby said V output signal is substantially proportional to XZ /Y.
14. The system of claim 12, wherein said first circuit means comprises a first integrator, a summing operational amplifier, and a level detector for providing said T signal.
15. The system of claim 12, wherein said second circuit means comprises a second integrator.
16. The system of claim 12, wherein said first circuit means comprises a first integrator, a summing operational amplifier, and a level detector, and said second circuit means comprises a second integrator.
17. The system of claim 16, further including control means responsive to an output of said level detector to control said first and second integrators and said output storage means.
18. The system of claim 13, wherein said first circuit means comprises a first integrator, a summing operational amplifier, and a level detector for providing's'aid T signal.
19. The system of claim 13, wherein said second circuit means comprises a second integrator.
20. The system of claim 13, wherein said first circuit means comprises a first integrator, a summing operational amplifier, and a level detector, and said second circuit means comprises a second integrator.
21. The system of claim 20, further including control means responsive to an output of said level detector to control said first and second integrators and said output storage means.
22. The system of claim 4 wherein said multiplier and divider circuit comprises:
a. a first integrating amplifier connected to receive a modifying voltage as a first input;
b. means for combining an output from the integrating amplifier with a second input in a manner as to produce a resultant signal tending toward zero as the integration of the modifying signal proceeds;
c. a second integrating amplifier connected to receive a signal proportional to the second input;
d. means responsive to said resultant signal becoming zero for applying to an output storage element a signal proportional to the integration level reached by the second integration amplifier; and
e. means for restarting the first and second integrating amplifiers in a manner that the integrating actions of the integrating amplifiers are recurrently repeated.