TECHNICAL REPORT ON THE
PAUL TRAP MASS SPECTROMETER
DEVELOPED IN THE MASS SPECTROMETRY LABORATORY
S. Renuka Prasad, S. Sevugarajan, Vikram A. Sarurkar,
Jiss Paul, Dnyaneshwar Gorde
DEPARTMENT OF INSTRUMENTATION
INDIAN INSTITUTE OF SCIENCE
BANGALORE 560 012
March 2003
Mass Spectrometry Laboratory, IISc
31
COPYRIGHT  2003 BY MASS SPECTROMETRY LABORATORY
DEPARTMENT OF INSTRUMENTATION
INDIAN INSTITUTE OF SCIENCE
BANGALORE –560 012
INDIA.
Mass Spectrometry Laboratory, IISc
2
Preface
This report provides technical details of a Paul trap mass spectrometer
fabricated in our laboratory. The report has two motivations. First, it aims to
consolidate and put together at this point of time, the status of our efforts. This
will help us in making a “road map” for future developments. Secondly, this
report hopes to be a primer, an introduction, for future generation of students
who enter the laboratory and wish to make a quick entry into the experiments
going on in the lab.
In addition, I think this report will also be of interest to groups of aspiring ion
trappers in the country who have been enthused and encouraged by our progress
to “get into” the area of ion trap mass spectrometry. They have often asked if
this is an “expensive” area of research. To this latter group I would like to say
that our modest success has been possible with a “shoe-string” budget and an
abundance of enthusiasm of dedicated young students who contributed bits and
pieces of their time.
March 2003
Mass Spectrometry Laboratory, IISc
A. G. Menon
3
Mass Spectrometry Laboratory, IISc
4
Contents
Page no.
1.
Introduction………………………………………………………….7
2.
Ion trap mass spectrometry………………………………………….8
3.
Technical specifications…………………………………………….16
4.
Performance characteristics………………………………………..25
Appendix 1
Circuit diagrams of electronic circuits…………………………….33
Appendix 2
Technical specifications of DAQ, PCI-MIO-16-E-1………………41
References……………………………………………………………….51
Mass Spectrometry Laboratory, IISc
5
Mass Spectrometry Laboratory, IISc
6
1
Introduction
The motivation of this technical report is to describe the design and
provide fabrication details of the Paul trap mass spectrometer that has been built
in our laboratory. This technical report gives both the theory of ion trapping in
Paul trap mass spectrometers and the technical specifications of mechanical
assembly, vacuum chamber and other electronic subsystems associated with our
laboratory’s Paul trap mass spectrometer. Section 2 gives the theory of the ion
trap mass spectrometry including development of equations of ion motion and
the conditions required for the ions to have stable trajectories inside the trap.
Section 3 provides the technical specifications of our trap electrodes, electronic
subsystems including constant current source, gating power supply, extraction
power supply, high voltage dc power supply, RF signal generator as well as the
vacuum system and graphical user interface are presented. Section 4 presents a
few mass spectra to demonstrate the performance our mass spectrometer.
Appendix 1 gives the detailed orcad layouts of all the electronic circuits
associated with the Paul trap mass spectrometer and the technical data related to
the National Instruments data acquisition device PCI-MIO-16-E-1 is given in
Appendix 2. At the end of the report a few important and pertinent references
are provided.
Mass Spectrometry Laboratory, IISc
7
2 Ion trap mass spectrometry
2.1 Motion of ions
The Paul trap mass spectrometer consists of a 3-electrode geometry
mass analyzer with a hyperboloid of one sheet forming the central ring electrode
and hyperboloid of two sheets forming the two-endcap electrodes (March and
Hughes, 1989). In an ideal Paul trap, the potential distribution inside the trap is
quadratic and the field varies linearly. The potential distribution has both
spherical and rotational symmetry due to the geometry of the ion trap. Legendre
polynomials are used repersenting such potential distributions inside the ion
trap. If Pn is the Legendre polynomial of order n, then the potential distribution
inside the trap in terms of spherical co-ordinates are given by (Brown and
Gabrielse, 1986; Beaty, 1986)
∞
ρn
n=0
ron
φ ( ρ ,θ ,ϕ ) = φo ∑ An
Pn (cos θ )
(2.1)
(1.1)
where ro = radius of the ion trap and φo is a time dependent potential given by
φo = U o + Vo cos Ωt
(2.2)
where Vo = zero-to-peak voltage of the RF potential
Uo = magnitude of the dc potential.
An = dimensionless weight factors for different terms.
The various terms corresponding to n = 0,1,2,3…etc. represent the multipole
components of the potential.
In case of pure quadrupolar ion trap only the term corresponding to n =
2 is non-zero. Therefore the potential distribution in a pure quadrupole ion trap
in terms of cylindrical co-ordinates r and z can be written as

φ ( r , z ) = A2  z 2 −

r2
2



(2.3)
where A2 is the weight of quadrupole component and has a value of –2 for pure
quadrupolar potential distribution. The differential equation that governs the
motion of a single ion can be obtained from (Landau, 1976)
ρ
dρ (r , z )
e
= − ∇φ (r, z )
2
m
dt
Mass Spectrometry Laboratory, IISc
(2.4)
8
ρ
where ρ is the position vector and ρ2 = r2 + z2. Substituting equation (2.3) in
equation (2.4) we get
κe
d 2u
=
[U o + Vo cosΩt ]u
2
dt
mro2
(2.5)
where κ = -2 when u refers to the radial, r, direction and κ = 4 when u refers to
the axial, z, direction. Equation (2.5) can be rewritten in the canonical form of
the Mathieu equation as
d 2u
+ (a u − 2q u cos 2ξ )u = 0
dξ 2
(2.6)
with the substitutions ξ = Ωt/2 and
a z = − 2a r =
q z = 2q r =
− 8eU o
mro2 Ω 2
(2.7)
.
4eVo
.
mro2 Ω 2
(2.8)
The parameters au and qu are referred to as Mathieu parameters. The solution to
the equation (2.6) (McLachlan, 1947) is given by
u = Au U e (ξ ) + Bu U o (ξ )
(2.9)
where Au and Bu are arbitrary constants and
U e (ξ ) =
U o (ξ ) =
∞
∑C
2 n ,u
cos(2n + β u )ξ
(2.10)
∑C
2 n ,u
sin(2n + β u )ξ
(2.11)
n = −∞
∞
n = −∞
If we define ωu,n as the angular frequency of order n for motion in
direction u (= r,z) it can be shown that


1
2


ω u , n =  n + β u Ω
0≤n≤∞
Ωt

Θ ξ =

2 

(2.12)
when n = 0 the fundamental frequency ωu,0 in either r or z direction is given by
ωu is referred to as the secular frequency of the ion in the respective direction.
Mass Spectrometry Laboratory, IISc
9
ωu =
1
βuΩ
2
(2.13)
Substituting equation (2.10) and (2.11) into equation (2.9) and solving
for C2n,u co-efficients, we get recursion formulae for C2n,u co-efficients. On
elimination of C2n,u co-efficients we get the continuous fraction expression for βu
in terms of au and qu as (March, 1992; March and Hughes, 1989)
β u2 = a u +
+
q u2
(β u
+ 2) − a u −
q u2
2
(β u
+ 4) − a u −
2
(β u
q u2
+ 6 ) − a u − ...
2
q 2u
(β u
− 2) − a u −
q u2
2
(β u
− 4) − a u −
2
(β u
q u2
− 6 ) − a u − ...
2
(2.14)
Stable regions of ion trajectories are characterized by values of au and qu for
which βu lies between 0 and 1.
Fig. 2.1 Mathieu stability plot.
Mass Spectrometry Laboratory, IISc
10
Fig. 2.1 shows the condition under which ions will be stable inside an
trap. The lines along which the βz remains constant are known as the iso-βz lines,
similarly the lines along which βr remains constant are known as iso-βr lines.
Let us now take a closer look at Fig. 2.1 to see how it could be of help
from the point of view of the Paul trap mass spectrometer. What the stability
plot implies is that all au- qu values lying outside the area bounded by βz = 0 to 1
and βr = 0 to 1 will have unstable trajectories. A choice of an au-qu can be
translated to dc voltage (Uo) and RF amplitude (V0) values by substitution in
equation (2.7) and (2.8) above (for a given mass, frequency and ro ). Similarly,
any point chosen within the stable region will indicate stable trajectories at the
computed Uo and Vo values (again for a given mass, frequency and ro ). This
provides us with a very simple technique to operate the instrument as a mass
spectrometer.
Let us choose to operate the mass spectrometer with Uo = 0 volts. What
this means is that the "operating line" lies along the au = 0 axis. Along this line,
the βz = 1 curve crosses this axis at a qu value of 0.908 (referred to in literature
as qcut-off). If we re-write equation (2.8) so that we may use commonly used units
(i.e.: m in amu, Vo in volts, ro in cms, and Ω in MHz) it takes the form
qz =
0.0978Vo
mro2 Ω 2
(2.15)
For qz = 0.908, ro =0.7cm and Ω = 1MHz, equation (2.16) gives a relationship
between mass and Vo as
m = 0.2198Vo(0-p)
(2.16)
Equation (2.16) provides us with the relationship, which can be used by the
mass spectroscopists in interpreting the intensity-voltage histogram as a mass
spectrum.
In ideal Paul traps the ion trajectories can be calculated from equation
(2.6). The form of ion trajectory in an r-z plane resembles Lissajous patterns
composed of two frequency components ωr,0 and ωz,0 with a superimposed
micromotion of frequency Ω/2π due to the RF drive frequency (Dawson, 1976).
Consequently, in z direction, the motion of ion can be represented by
z=Z+Γ
(2.17)
where Γ is the displacement due to micromotion and Z is the displacement due
to macromotion (which is the secular motion of the ions). When Γ << Z (that is
z ≅ Z), it can be shown that the acceleration due to RF drive d2Γ/dξ2, averaged
over a period of the RF drive is equal to zero and the acceleration of the secular
motion d2z/dξ2 averaged over the same period is given by
Mass Spectrometry Laboratory, IISc
11
1 2
d 2z

=
−
+
a
qu  z

u
2 
dξ 2

(2.18)
which, when written in terms of time, becomes
d 2z
1 2  Ω2

a
qu 
z
=
−
+

u
2  4
dt 2

Ωt


Θ ξ =
2 

(2.19)
Equation (2.19) corresponds to simple harmonic motion of the ions and if ω0 is
used to refer to the ion secular frequency, then
d 2z
= − ω 02 z
dt 2
(2.20)
Comparing equation (2.13) and (2.19) it turns out that
1 

β z =  a u + q u2 
2 

1/ 2
(2.21)
The expression (2.21) is valid for qz values less than 0.4. This approximation is
known as Dehmelt or adiabatic approximation (Wuerker et al., 1959; Dawson,
1976).
In practice, however, several geometrical and constructional
imperfections introduce non-linearities in the field distribution. The real trap
potential deviates from the ideal one due to truncation of the electrodes,
deviations from the hyperbolic shape, possible misalignments, space charge of
the stored ion cloud and existence of aperture for admitting electrons as well as
for the collection of destabilized ion. The fundamental properties of the ion
motion within the idea Paul trap is hence altered due to the introduction of
higher order terms in the quadrupolar field, thus exhibiting effects differing
considerably from those of the linear trap. Some of these effects are as follows:
the non-linearity of the RF field in the axial and radial direction, perturbation in
ion secular frequencies (the characteristic frequency with which the ions
oscillate inside the trap) and coupling between axial and radial motion. The
experimental conditions in practical traps appear as non-linear parameters in the
restoring term in the equation of ion motion. These higher order contributions
can be visualized as “perturbations” to the ion motion in a pure quadrupolar
field. The equation of ion motion in axial and radial directions incorporating the
experimental and practical constraints resembles the well-studied Duffings
equation.
Mass Spectrometry Laboratory, IISc
12
2.2 Block diagram of the Paul trap mass spectrometer
The block diagram of a Paul trap mass spectrometer is given in Fig.
2.2. It consists of a three-electrode geometry mass analyzer, an electron gun and
a detector. The electronic circuits include an electrode gun supply, an RF power
source, a dc high voltage detector power supply and a interface card to enable
PC control of the instrument.
In our mass spectrometer the analyte gas molecules are ionized in situ
by electron bombardment. Electrons are produced by thermionic emission when
a rhenium filament is heated by a constant current source. An extraction
electrode in combination with the gating electrode focuses ions into the central
cavity of the trap.
The gating electrode is used to gate the electron beam into the ion trap
cavity. When positive potential is applied to the gating electrode the electron
beam is allowed into the ion trap cavity, and when negative potential is applied
the electron beam is blocked. The electron beam gating control block consists of
positive and negative voltage power supplies. A high voltage pulsing circuit
connects the two power supplies alternately to the gating electrode, thereby
allowing the electron beam in or blocking it.
The high voltage RF supply supplies a 1MHz, 2kVp-p RF potential for
producing the confining field. The RF power supply is based on a crystal
oscillator for high frequency stability. A high-Q LC tuned circuit at the output
couples the output of the supply to the ion trap.
The electrometer amplifier having a gain in about of 106 amplifies
output from the electron multiplier detector. It also has a fast response time to
faithfully amplify the fast rising mass peaks.
2.3 Timing diagram
In order to obtain the mass spectrum of an analyte gas the different
electronic subsystems related to electron gun and amplitude of the RF potential
needs to be co-ordinated. The timing diagram for a simple scan in a typical
experiment is shown in Fig. 2.3. At the start of the experiment, an ionization
pulse is applied to the gating electrode. This allows the electron beam in the trap
and ionizes the analyte gas molecules inside. The RF potential at this instant is
held at a value such that the lowest mass of interest is trapped. After this
ionization phase the ions are allowed to cool in the cooling period. After the
cooling period the RF is ramped to generate the mass spectrum and electron
multiplier detector detects the ejected ions. The mass spectrum from each
experiment is then averaged to get a better S/N ratio.
Mass Spectrometry Laboratory, IISc
13
Fig. 2.2 Block diagram of a Paul trap mass spectrometer
Mass Spectrometry Laboratory, IISc
14
.
Ionization Cooling
Ramping
time
time
time
Figure 2.3: Timing diagram of Paul trap mass spectrometer
Mass Spectrometry Laboratory, IISc
15
3 Technical specifications and circuit description
3.1 Mass analyzer
The hyperbolic profile of the endcap and ring electrode is given by
(Knight,1983)
r2 z2
−
= −1
ro2 z o2
r2 z2
−
=1
ro2 z o2
(end cap)
(ring electrode)
(3.1)
(3.2)
In an ideal trap the radius of the central ring electrode ro and half the
distance between the two closest points of end-cap electrodes zo are related by
the expression ro2 = 2z02. The radius of the ring electrode r0 in our machine is
7mm and z0 (= r0/√2) works out to 4.95mm. The electrodes are fabricated from
austenitic stainless steel flanges whose diameter is 88.9mm (3.5”) and the
surfaces are machined at the center of the flange using a CNC lathe. The
truncation of the electrodes has been made at 3ro. For achieving the required
axial distance between the tips of the end cap electrodes as 9.9mm (=2zo)
appropriate Teflon spacers have been used between the electrodes. The endcap
electrode which faces the filament side has a 1mm diameter hole at it’s center
for admitting electrons into the trap for ionizing the neutral analyte gas
molecules. The other endcap electrode, which faces the electron multiplier, has a
3mm hole at its center for collecting the destabilized ions.
3.2 Electron gun
The electron gun consists of a filament, an extraction electrode and a
gating electrode. Our filament has a triple filament assembly and is a spare of
the filament of the thermal ion source of a written-off AEI-MS702 mass
spectrometer. We use both rhenium and tungsten filaments. The extraction
electrode, gating electrode and the filament are all electrically isolated and
mounted on one-endcap electrode.
Mass Spectrometry Laboratory, IISc
16
3.3 Detector
The electron multiplier used in our machine is an ETP’s electron
multiplier (model number AEM 5000A). It provides an amplification of about
106 for a bias voltage of –2.5kV.
3.4 Mounting flange
The Paul trap mass spectrometer including the mass analyzer, electron
gun and electron multiplier are mounted on CF100 stainless steel flange.
Provision has been made on this flange for attaching two needle valves for
admitting sample gases. Vacuum feed throughs have been made locally using
Teflon and stainless steel welding rods. First, Teflon cylinders were tight fitted
into the flange and then 2mm stainless steel welding rods are pierced into the
Teflon. Two 5-pin feed throughs and three single-pin feed throughs have been
used. These feed throughs were tested to withstand a pressure of 10-6torr.
3.5 Vacuum
The mechanical assembly of the Paul trap is positioned in a vacuum
chamber. Vacuum sealing of the demountable flange is done using a neoprene
O-ring. The vacuum is maintained at 10-6 torr with a help of a diffusion pump/
LN2 trap backed by a rotary pump. Pirani and Penning gauges are used for
measuring the pressure.
3.6 Electronics
The Paul trap mass spectrometer requires the following power supplies
for its operation:
Constant current source: A current source that can deliver a current from 0A to
8A to the filament to generate the electrons. Typical value of current used in our
experiments is about 4A.
Filament bias supply: The filament bias supply is a voltage source capable of
outputting 0 to –150V/10mA. It is a variable supply, on which current source
floats. Typical value set in our experiments is about –14.5V.
Extraction electrode supply: This supply is capable of delivering a variable
output voltage from 0 to 130V/10mA. This range of voltage will give flexibility
to the experimenter to optimize the extraction voltage for his experiments. In our
experiments, the extraction electrode is biased at 30V.
Gating electrode supply: This supply is capable of delivering a switched ±150V
to the gating electrode. The switching of the gating electrode enables control on
Mass Spectrometry Laboratory, IISc
17
the entry of electrons into the trap. This switching is done through a pulsing
circuit.
RF power supply: This variable RF generator outputs 1.5kVp-p at a frequency of
1MHz for analyzing masses up to 150amu.
High voltage dc power supply: This supply is capable of delivering 0 to –
3kV/1mA dc voltage for biasing the electron multiplier. This switched mode
power supply is kept at –2.5kV in our experiments.
Electrometer amplifier: Electrometer amplifier is required for further
amplifying the ion signal obtained from the electron multiplier. The gain of this
amplifier is about 106.
The description of these circuits will be discussed in the following sections.
For ionization of the sample gas molecules, we require sufficient
number of electrons with sufficient energy. These electrons are produced by an
electron gun. An electron gun assembly used by us consists of a filament, an
extraction electrode and a gating electrode as shown in Fig. 3.1.
Figure 3.1: Electron gun assembly
Electrons are produced by thermionic emission from a filament with
high work function. The number of electrons produced is function of the
temperature of the filament and thus passing larger current through the filament
results in higher filament temperature as well as higher electron emission. Since
it is desirable to have a constant electron current the instrumentation need to
maintain a constant filament temperature. To a good approximation this is
achieved by passing a constant current through the filament. The appropriate
electronics for passing a constant current through the filament must take into
consideration the variation of filament resistance with temperature (both
tungsten and rhenium, the filament material used in our experiments, have
positive temperature coefficient).
The simplest approximation to a current source is using a voltage
source V and a series resistor R connected to a load RL as shown in Fig. 3.2.
Mass Spectrometry Laboratory, IISc
18
Here as long as RL<<R the current is nearly constant because any small
variation in the RL will not seriously affect the combined resistance. But this
resistor current source has several drawbacks. For good approximation to a
constant current through RL we must use large voltages, with high power
dissipation on resistor R. In addition, the current is not easily programmable.
Fig. 3.2: Resistor current source
An alternative scheme is to use a transistor based constant current
source. The basic concept of the current source is shown in Fig. 3.3. This circuit
works by applying VB to the base of transistor (with VB > 0.6, so that the
transistor will be in active region). Here the power dissipation will depend on
the VCE of the transistor and, consequently, a proper choice of VCC can reduce
the power dissipation. Further, programming of the current source can be
achieved by controlling VB.
Fig. 3.3: Basic transistor current source
In the present work we have configured a constant current source,
which sources current to a load returned to ground. An error amplifier is used to
maintain the base bias of the transistor as a constant for a given load current.
The basic concept of this circuit is shown in Fig. 3.4. The feedback is obtained
from a small current sensing resistor Rs, which is connected in series with the
load RL. The error amplifier will compare the voltage across the sensing resistor
Rs with Vref, the reference voltage. By varying the Vref we can program the
current source.
Mass Spectrometry Laboratory, IISc
19
Figure 3.4: Variable constant current source
We will next focus on achieving requisite electron energy. This is
achieved by floating the filament constant current source with respect to the
extraction electrode. The final energy attained by the ion is a combination of this
floating voltage, extraction electrode bias as well as the RF amplitude of the ring
electrode during the ionization period. We have provided a wide range of
voltage from 0V to –150V in order to give variable energy to electrons.
Referring to the Fig. 3.1 we should focus the electrons in to the trap to
ionize the sample molecules with the help of an extraction electrode is used for
this purpose. For focusing the electrons a positive dc voltage is required to be
applied to the extraction electrode. This will accelerate as well as focus the
electrons in to the trap. Normally the applied dc voltage is of the order of
+14.5V. The voltage source used is of voltage series pass voltage regulator.
Detailed constant current source circuit diagram is given in Appendix 1.
The second electrode in the electron gun assembly is called the gating
electrode. This electrode is used to gate the electrons for user-determined times
into the trap. This allows the user to control the ionization time and hence the
number of ions being produced inside the trap. This timing will lead to the need
for a pulsing circuit. This pulsing circuit will give positive or negative dc
voltage to the gating electrode to pass or block the electrons. From experiments
it was determined that a voltage of +/-150V would be required to accelerate or
block the electrons.
The basic block diagram of the pulsing circuit is shown in Fig. 3.5. The
pulsing circuit should respond to the input pulse. The switching time of the
circuit should be minimized to shut the electrons quickly. This circuit is
interfaced with the computer where we can determine the pulsing time in terms
of ionization time and cooling time using LABVIEW software and interfacing
card. The switching device switches the positive or negative supplies to the
gating electrode. The switching device used in the circuit is transistor. The
isolator will electrically isolate the circuit with computer.
Mass Spectrometry Laboratory, IISc
20
Figure 3.5: Basic pulsing circuit
The voltage sources mentioned in the above systems uses a series pass
voltage regulator which gives a very good output voltage regulation. The basic
block diagram of the regulator is shown in the Fig 3.6. A series regulator places
the regulating element (series pass control device) in series with the load, and it
is the voltage across the regulating element that is varied to adjust the load
voltage. A feedback voltage from output is fed to the comparator. This compares
the feedback voltage with a reference voltage and the difference signal controls
the series pass device. The series pass control device used in the circuit is a
transistor and the comparator is based on an operational amplifier. The feedback
circuit is a simple resistive voltage divider. The pulsing circuit diagram is given
in detail in Appendix 1.
Figure 3.6: Basic series pass regulator
2.3 RF power supply
For trapping the ions inside the trap RF supply is used to produce a
quadrupolar potential distribution inside the trap. RF quadrupolar field causes
mass selective trapping of the ions inside the Paul trap. The voltage of the RF
power supply can varied up to 1.5kVp-p at 1MHz in order to have a wide mass
range with sufficient trapping strength. The basic concept of the RF generator is
shown in Fig. 3.7.
Mass Spectrometry Laboratory, IISc
21
Figure 3.7: RF generator
Since the output voltage of the RF generator should be variable, we use
a modulator along with a crystal oscillator that will give 1MHz square wave.
This is applied to amplitude modulator, which modulates the amplitude of the
1MHz signal according to the modulating voltage. Finally the modulated signal
is applied to a power amplifier, which amplifies the signal as well as tunes the
signal with the help of a tank circuit (primary coil) to get a perfect 1MHz sine
wave. By proper coupling of this primary RF signal with a secondary tank
circuit we can generate the required RF voltage, which will be applied to the
trap.
The important point to be noted while implementing the RF supply is
that we should take into consideration of various capacitance like trap
capacitance, cable capacitance etc. Hence, it is better to have a tunable
inductance in the secondary to compensate the capacitance change so that the
tank circuit will tune to 1MHz. This will ensure maximum coupling between the
primary and secondary of the tank circuits. The full circuit diagram of RF
generator is given in Appendix 1.
High voltage dc power supply
An electron multiplier is used to detect the ions ejected out from the
trap. In our laboratory we use ETP’s electron multiplier model number AEM
5000A, which works on the principle of secondary emission and amplifies the
input ion signal by an amount of up to 107 depending on the dc voltage applied
across it. A high negative voltage of order of 3kV is required for the operation of
the electron multiplier. This voltage should be regulated and free from ripple as
this ripple might get coupled to the output. Using linear techniques it is rather
difficult to get a well regulated –3kV power supply. Hence a switched mode
power supply techniques are employed to get a regulated –3kV power supply.
The detailed circuit diagram of -3kv dc power supply is given in page 38 and 39
of Appendix 1. As shown in the circuit MOSFET IRF840 is used as a series pass
element, as a switch, in this case. When the MOSFET is ON energy is stored in
an inductor connected to the MOSFET. When the MOSFET switches OFF, the
energy stored in the inductor is transferred to another inductor through the ferrite
Mass Spectrometry Laboratory, IISc
22
core. The output current is rectified and filtered to give a smooth –3kV DC. The
turn’s ratio of the two inductors ensures the steeping up of the voltage. A PIC
micro controller is used to control the whole operation of the circuit. It also
provides the functions like shutting down the circuit, in event of output overload
or overheating of the MOSFET. A serial link is also provided for digitally
controlling the output voltage. The MOSFET switching frequency is 50kHz,
which ensures easy filtering of the output voltage even by small value
capacitors.
Electrometer Amplifier
An electrometer amplifier is employed for amplifying the signal from
the electron multiplier. An electrometer amplifier consists of a trans-impedance
amplifier, a current to voltage converter stage, followed by one or two voltage
amplification stages.
The transimpedance amplifier stage is built around LF351 OpAmp. The
feedback resistor sets the current conversion factor to 1V/µA. The bandwidth of
the amplifier is limited to 200kHz and the response time of the amplifier is about
5.6µs. Two voltage amplifier stages follow the transimpedance amplifier stage.
Each of the voltage amplifier stage has a gain of about 34, thereby giving an
overall gain in excess of 1000. Each of the OpAmp is provided with separate
supply filtering and decoupling network on the supply pins to minimize the
noise. The PCB for the electrometer amplifier is designed with due
consideration to noise reduction and is provided with ground planes on both the
sides. The decoupling capacitors are kept as close as possible to the supply pins
and the signal traces are made as short as they can be. Finally the board is
mounted in a metal box to shield the amplifier circuit from radiated noise. The
detailed circuit layout is shown in Appendix 1.
Graphical User Interface (GUI)
A user friendly Graphical interface between the computer and the mass
spectrometer has been developed by using National instrument’s LABVIEW
software and PCI-MIO-16-E-1 data acquisition device (DAQ).
National instrument’s PCI-MIO-16-E-1 data acquisition device (DAQ)
generates the required analog and digital signals according to the user’s
specification (such as sampling rate, amplitude, timing sequence, etc.,) given in
the LABVIEW program. The DAQ that we have in our lab has 16 analog input
channels, two 12-bit analog to digital converters with the sampling rate of 1.25
mega samples/sec. The detailed specification of the hardware is given in
Appendix 2.
Mass Spectrometry Laboratory, IISc
23
Generation of gating pulse and control signal
The gating pulse is generated with the help of general-purpose timing
signal connections available in the DAQ. The pulse is obtained by programming
the GPCTR1 counter/timer through the software. The counter output’s a pulse
depending on the total experimental time and the pulse duration depends on the
ionization time. The software calculates the total experimental time internally,
which is the addition of ionization time, cooling time and ramping time that are
set by the user in the program.
The control signal for controlling the output amplitude of the RF signal
source is generated by digital to analog converter present in the DAQ. The DAQ
hardware consists of two digital to analog converters, DAC0 and DAC1. We
have used DAC0 for our purpose. The software generates the digitized data and
downloads the data to the DAC0 through the computer and DAC0 generates the
corresponding analog output. Through the software user can set the ionization
time, cooling time, ramping time, initial RF voltage, voltage at ionization period,
voltage at cooling period and maximum RF voltage.
Acquisition and calibration of mass spectrum
The electrometer amplifier’s output is given to channel0 of the ADC
present in the DAQ. The analog signal obtained from the electrometer amplifier
is converted to digital data’s through the ADC and is displayed in the
oscilloscope block available in the front panel of the program. To obtain a better
S/N ratio the signal is continuously averaged and is displayed in averaged mass
oscilloscope block available in the front panel of the program.
For the purpose of displaying the calibrated the obtained averaged mass
spectrum we use XY plotter block available in the software. A small amplitude
RF signal obtained from the feedback of RF generator is digitized by using
channel1 of the ADC in DAQ and this voltage data serves as the input for X
terminal of the plotter block. The scale of the X-axis is calibrated in terms of
amu by using the Equation (2.16) and by substituting the digitized voltage data
obtained from Channel1 for V. The signal for Y input is averaged mass
spectrum.
The user has the flexibility of zooming the axis of oscilloscope and XY
plotter blocks. The program also has the option of restarting the averaging
whenever the user changes any of the experiment parameters.
Mass Spectrometry Laboratory, IISc
24
4. Performance characteristic
4.1 Mechanical assembly
The three trap electrodes, electron gun assembly, extraction electrode,
gating electrode and the electron multiplier are supported by three stainless steel
pillars. These pillars are mounted on a 6” diameter stainless steel flange. The
required electrical isolation is done by using Teflon washers/spacers. The
electrical connections from different power supplies are taken into vacuum
chamber by Teflon feed through’s that are fitted to the stainless steel flange. The
flange has a provision for attaching two needle valve for inletting the analyte
gasses. The output of the electron multiplier is taken out to the electrometer
amplifier through a stainless steel wire through one of the feed through in the
flange. To have a better S/N an electromagnetic shielding is provided to the
stainless steel wire by using a stainless steel wire mesh around the wire. The trap
electrodes and the entire mechanical assembly are shown in Fig. 4.1and 4.2
respectively.
4.2 Electronics
All electronic subsystems were tested for having achieved the desired
performance. These test included checking for achieving the required
specification and also for their load and line regulation. The final performance
of these subsystems will be seen in the quality of the mass spectra obtained. We
show below, as an example, the switching characteristics of the pulsing circuit.
Fig.4.3 shows the rise and fall time for a ±150V pulse. It may bee seen
that the rise time is about 16µs and the fall time is of the order of 270µs.
Although the fall time is large, it does not affect the performance of the mass
spectrometer.
The photograph of the chassis housing the electronic subsystem is
shown in Fig. 4.4.
Mass Spectrometry Laboratory, IISc
25
Fig. 4.1 Endcap and ring electrodes used in our laboratory Paul trap mass spectrometer
Mass Spectrometry Laboratory, IISc
26
Fig. 4.2 Full view of our laboratory Paul trap
Mass Spectrometry Laboratory, IISc
27
Fig. 4.3 Switching characteristic of the pulsing circuit.
Fig. 4.4 Chassis housing all the electronic subsystem
Mass Spectrometry Laboratory, IISc
28
4.3 GUI
A snap shot of the GUI developed using LABVIEW is given in Fig.4.5.
The user can vary any experimental parameter by setting appropriate values for
that particular parameter in the front panel of the GUI. The following section
gives the explanation of different functions of the software according to the
numerals marked in the front panel.
1. Using six controls, user can set ramping time, cooling time,
ionization time, maximum RF voltage, RF voltage at cooling and
RF voltage at ionization.
2. This chart shows the typical analog control signal generated by the
software.
3. This oscilloscope has the function of displaying the raw ion signal
data obtained from the electrometer amplifier.
4. The time averaged uncalibrated mass spectrum will be displayed in
this scope.
5. The final calibrated mass spectrum can be obtained from this XY
recorder.
6. Adjustments for varying the scales of the oscilloscopes.
7. Display indicating how many cycles of averaging has been done.
User can also restart the averaging by pressing “RESET
AVERAGING” button.
4.4 Mass spectra
The Mass spectrometer has to be operated in the following sequence for
obtaining the mass spectrum.
1. Switch ON the rotary pump and open the backing and the butterfly
valves.
2. When the pressure reading in the Pirani Gauge meter reaches
around 5x10-3torr, the diffusion pump should be switched ON. The
user has to wait till the pressure indicated by the Penning gauge
meter reaches 1x10-6torr.
3. All the electronic subsystems except the current source and RF
generator are switched ON.
4. Current source and the RF generator are next turned ON after
starting the control program in which all the experimental
parameters are set.
5. The RF signal and the electrometer amplifier output, which is the
mass spectrum, are fed into a 100MHz analog oscilloscope. The two
signals are superposed and the RF voltage corresponding to the
peaks in the mass spectrum are calculated for the purpose of mass
calibration.
6. The Electrometer amplifier output is also fed to a 100MHz Digital
scope for the purpose of averaging to have better S/N ratio. Typical
Mass spectrum of two samples – Benzene and Xylene are shown in
Fig. 4.5 and 4.6. It may be seen that the resolution obtained is
reasonable through out the range.
Mass Spectrometry Laboratory, IISc
29
2
5
1
4
3
6
7
Fig. 4.5 Front panel of the control program
Mass Spectrometry Laboratory, IISc
31
30
Mass Spectrometry Laboratory, IISc
31
Mass Spectrometry Laboratory, IISc
32
Appendix 1
Mass Spectrometry Laboratory, IISc
31
Mass Spectrometry Laboratory, IISc
31
34
Mass Spectrometry Laboratory, IISc
35
Mass Spectrometry Laboratory, IISc
36
Mass Spectrometry Laboratory, IISc
37
Mass Spectrometry Laboratory, IISc
38
Mass Spectrometry Laboratory, IISc
39
Mass Spectrometry Laboratory, IISc
40
Appendix 2
Mass Spectrometry Laboratory, IISc
31
Specifications
This appendix lists the specifications of PCI-MIO-16E-1 device.
These specifications are typical at 25 °C unless otherwise noted.
Analog Input
Input Characteristics
Number of channels
PCI-MIO-16E-1 .............................. 16 single-ended or 8 differential
(software-selectable per channel)
Type of ADC.......................................... Successive approximation
Resolution .............................................. 12 bits, 1 in 4,096
Max sampling rate (single-channel)1
PCI-MIO-16E-1 .............................. 1.25 MS/s
Relative accuracy ................................... ±0.5 LSB typ dithered,
±1.5 LSB max undithered
Analog Output
Output Characteristics
Number of DAC................................2
Resolution...............................................12 bits, 1 in 4,096
Type of DAC ..........................................Double-buffered, multiplying
FIFO buffer size
PCI-MIO-16E-1....................................2,048 samples
Data transfers ..........................................DMA, interrupts,
programmed I/O
DMA modes ...........................................Scatter gather (single transfer,
demand transfer)
Dynamic Characteristics
Mass Spectrometry Laboratory, IISc
42
Settling time for full-scale step...............3 µs to ±0.5 LSB accuracy
Slew rate .................................................20 V/µs
Noise.......................................................200 µVrms, DC to 1 MHz
Glitch energy (at midscale transition)
Magnitude
Reglitching disabled.................±20 mV
Reglitching enabled..................±4 mV
Duration...........................................1.5 µs
Stability
Offset temperature coefficient ................±50 µV/°C
Gain temperature coefficient
Internal reference.............................±25 ppm/°C
External reference............................±25 ppm/°C
Digital I/O
Number of channels................................8 input/output
Compatibility ..........................................TTL/CMOS
Timing I/O
Number of channels ............................... 2 up/down counter/timers,
1 frequency scaler
Resolution
Counter/timers ................................ 24 bits
Frequency scaler ............................. 4 bits
Compatibility ......................................... TTL/CMOS
Base clocks available
Counter/timers ................................ 20 MHz, 100 kHz
Frequency scaler ............................. 10 MHz, 100 kHz
Mass Spectrometry Laboratory, IISc
43
Base clock accuracy ............................... ±0.01%
Max source frequency............................ 20 MHz
Min source pulse duration ..................... 10 ns, edge-detect mode
Min gate pulse duration ......................... 10 ns, edge-detect mode
Data transfers .........................................DMA, interrupts,
programmed I/O
DMA modes........................................... Scatter gather
Triggers
Analog Trigger
Source
PCI-MIO-16E-1...............................ACH<0..15>, external trigger
(PFI0/TRIG1)
Level .......................................................± full-scale, internal;
±10 V, external
Slope .......................................................Positive or negative
(software selectable)
Resolution...............................................8 bits, 1 in 256
Hysteresis................................................Programmable
Bandwidth (–3 dB)
PCI-MIO-16E-...................................2 MHz internal, 7 MHz external
External input (PFI0/TRIG1)
Impedance........................................10 k.
Coupling ..........................................DC
Protection.........................................–0.5 to VCC + 0.5 V when
configured as a digital signal, ±35 V when configured as an analog trigger signal
or disabled, ±35 V powered off
Digital Trigger
Compatibility ..........................................TTL
Mass Spectrometry Laboratory, IISc
44
Response .................................................Rising or falling edge
Pulse width .............................................10 ns min
RTSI
Trigger lines ........................................... 7
Calibration
Recommended warm-up time ................ 15 min
Calibration interval ................................ 1 year
External calibration reference ................ >6 and <10 V
Onboard calibration reference
Level ............................................... 5.000 V (±3.5 mV) (over full
operating temperature, actual
value stored in EEPROM)
Temperature coefficient .................. ±5 ppm/°C max
Long-term stability ......................... ±15 ppm/
Bus Interface
Type .......................................................Master, slave
Power Requirement
+5 VDC (±5%)
PCI-MIO-16E-1................................... 1.1 A
Power available at I/O connector ........... 4.65 to 5.25 VDC at 1 A
Physical
Dimensions
(not including connectors) .................... 17.5 by 10.6 cm (6.9 by 4.2 in)
I/O connector
PCI-MIO-16E-1.............................. 68-pin male SCSI-II type
Hardware Overview
Mass Spectrometry Laboratory, IISc
45
Figure 3-1 shows a block diagram for the PCI-MIO-16E-1
Figure 3-1. PCI-MIO-16E-1 Block Diagram
I/O Connector
Mass Spectrometry Laboratory, IISc
46
Mass Spectrometry Laboratory, IISc
47
Mass Spectrometry Laboratory, IISc
48
Mass Spectrometry Laboratory, IISc
49
Mass Spectrometry Laboratory, IISc
50
Beaty, E.C. (1986) Calculated electrostatic properties of ion traps. Phys. Rev.A,
Vol. 33, 3645-3656.
Brown, L.S. and Gabrielse, G. (1986) Geonium theory: Physics of single
electron or ion in a Penning trap. Rev. Mod. Phys., Vol. 58, 233-311.
Dawson, P.H. (1976) Quadrupole Mass Spectrometry and it’s Application,
Elsevier, Amsterdam.
Knight, R.D. (1983) The general form of the quadrupole ion trap potential. Int.
J. Mass Spectrom. Ion Physics, Vol. 51, 127-131.
Landau, L.D. and Lifshitz, E.M. (1976) Mechanics, Third Edition, Pergamon,
UK.
McLachlan, N.W. (1947) Theory and applications of Mathieu functions,
Calarendon press, UK.
March, R.E. (1992) Ion trap Mass Spectrometry. Int. J. Mass Spectrom. Ion
Processes, Vol. 118/119, 72-135.
March, R.E. and Hughes, R.J. (1989) Quadrupole Storage Mass Spectrometry,
Wiley-Interscience Publications, New York.
Wuerker, R.F., Shelton, H., and Langmuir, R.V. (1959) Electrodynamic
containment of charged particles. J.Appl.Phys, Vol. 30 342-349.
Mass Spectrometry Laboratory, IISc
31