1
V.I. Troyan et al., Eur. J. Mass Spectrom. 21, 1–12 (2015) Received: 24 January 2015 n Revised: 8 April 2015 n Accepted: 8 April 2015 n Publication: 21 April 2015
EUROPEAN JOURNAL OF MASS SPECTROMETRY
Multisectional linear ion trap and novel loading method for optical spectroscopy of electron and nuclear transitions Victor I. Troyan,a Peter V. Borisyuk,a Andrey V. Krasavin,a Oleg S. Vasiliev,a Vitaly G. Palchikov,b Ivan A. Avdeev,a,c Denis M. Chernyshev,a,c Sergey S. Poteshina,c and Alexey A. Sysoeva,c a National Research Nuclear University MEPhI (Moscow Engineering Physics Institute), 115409, Kashirskoe shosse 31, Moscow, Russian Federation. E-mail:
[email protected] b Institute of Metrology for Time and Space at National Research Institute for Physical-Technical and Radiotechnical Measurements, Mendeleevo, Moscow Region, 141579, Russian Federation c
Linantec Ltd, 115409, Kashirskoe shosse 31, Moscow, Russian Federation There is a growing need for the development of atomic and nuclear frequency standards because of the important contribution of methodsfor precision time and frequency measurements to the development of fundamental science, technology and the economy. It is also conditioned by their potential use in optical clocks and quantum logic applications. It is especially important to develop a universal method that could allow one to use ions of most elements effectively (including ones that are not easily evaporated) proposed for the above-mentioned applications. A linear quadrupole ion trap for the optical spectroscopy of electron and nuclear transitions has been developed and evaluated experimentally. An ion source construction is based on an ultra-high vacuum evaporator in which a metal sample is subjected to an electron beam of energy up to 1 keV, resulting in the appearance of gaseous atoms and ions of various charge state. The linear ion trap consists of five successive quadrupole sections including an entrance quadrupole section, quadrupole mass filter, quadrupole ion guide, ion-trap section and exit quadrupole section. The same radiofrequency but a different direct current voltage feeds the quadrupole sections. The instrument allows the mass- and energy-selected trapping of ions from ion beams of various intensities and their localization in the area of laser irradiation. The preliminary results presented show that the proposed instrument and methods allow one to produce effectively up to triply charged thorium ions as well as to trap ions for future spectroscopic study. The instrument is proposed for future use in optical clocks and quantum logic application development.
Keywords: thorium, ion trap, optical spectroscopy, nuclear spectroscopy, mass spectrometry
Introduction Excitation of electron transitions in atoms is the basis of modern spectroscopy and metrology. Resent advances in atomic clocks instrumentation and applications development1–5 stimulated the search for an even more accurate time reference. Experiments using atomic clocks have already provided the most accurate tests of general relativity,1,6 as
ISSN: 1469-0667 doi: 10.1255/ejms.1329
well as set the most stringent limits on the range of possible temporal deviation of fundamental constants. 7 However, a number of fundamental physical limitations caused, in particular, by black-body radiation, lead to the existence of the limit of reproducibility and stability of the transition frequency of 10–18.
© IM Publications LLP 2015 All rights reserved
2
Multisectional Linear Ion Trap and Novel Loading for Electron and Nuclear Transition Spectroscopy
The use of nuclear transitions would allow the accuracy of measurements to be increased by a few orders of magnitude in view of their isolation by the electron shell and, therefore, much less sensitivity to external excitations.8 The energy of the majority of nuclear transitions is in the range of 103–106 eV, which is inaccessible with existing sources of coherent irradiation. The only known exclusion is the isomeric energy level of isotope 229Th with a theoretically estimated9 energy of 7.8 ± 0.5 eV. The energy is in the vacuum ultraviolet (VUV) range and can be accessible with existing lasers. The transition between the 229Th nuclear states could potentially be used as a frequency reference8,10 with a fractional uncertainty approaching 10–20. This could help improve the accuracy of global satellite navigation systems, detect the dependence of the transition frequency on the gravitational field and remotely detect deposits of rare-earth elements, oil and gas condensate fields, and provide a chance to create a gamma-ray laser.11 The nuclear frequency standard would also solve some problems of fundamental physics, in particular the measurement of a number of fundamental constants (the fine structure constant, the gravitational constant) with high accuracy and thus check the foundations of general relativity and cosmology. Triple-ionized thorium has a more convenient structure of levels for fluorescence detection and laser cooling compared with lower ionization states.3 Searching for the nuclear isomer transition of 229Th presents a significant challenge. Any charge state should exhibit the isomer transition,12 including neutral, single-, double- and triple-charged 229Th. However, triplecharged 229Th is especially attractive because of the availability of a forbidden transition and because its ionization energy is much higher than the isomeric transition energy.3 Therefore, the search for the nuclear isomer transition of 229Th can be simplified by the development of effective methods to produce, localize, cool and spectroscopically study the 229Th3+ ions. The importance of these methods is also conditioned by their potential use in optical clocks and quantum logic applications. It is especially important to develop a universal method that could allow one to use effectively the ions of most elements proposed for the mentioned applications, including ones that are not easily evaporated. Laser cooling is the most common method used to study the quantum properties of individual ultra-cold (1 µK) atoms and ions13 confined in various traps. Significant progress with the use of such physical systems has been achieved in the traditional ultra-high resolution laser spectroscopy, as well as in new areas such as quantum logic,14,15 optical frequency and time standards.1 When the ions are spatially localized in a small volume of a trap it becomes possible to use effectively ultra-deep laser cooling techniques, including the method of sympathetic cooling.1–5 Two advanced research groups have started to address the problem with experimental optical spectroscopy of thorium ions: the Georgia Institute of Technology (USA) and Physikalisch-Technische Bundesanstalt (Germany). The German team created a laser system for the spectroscopy of electronic transitions and designed a thorium ion trap for
Coulomb crystals of Th+. First, Coulomb crystals containing up to 103 ions of Th+ were produced and detected. Singly charged ions 229Th+ and 232Th+ were produced from metallic thorium16,17 irradiated by a laser and were loaded into a linear quadrupole ion trap.18
Figure 1. Schematic picture of the linear ion trap for the optical spectrometry of electron and nuclear transitions (a) and overall view of constructed instrument (b): 1, vacuum chamber; 2, ion source; 3, linear ion trap; 4–7, lasers for first-stage cooling (not currently installed); 8 and 9, lasers for second-stage cooling and excitation of fluorescence (not currently installed); 10, a tunable laser for an isomeric transition search in the region of VUV (not currently installed); 11, a CCD camera (not currently installed); 12 and 13, gate valves; 14, turbo pump HiPace 300M; 15, titanium sublimation pump ST22; 16, ion pump VacIon Plus 300; 17, turbo pump; 18, rotary vane pump; 19, rotary vane pump RV5; 20 and 21, vacuum sensors; 22, helium.
V.I. Troyan et al., Eur. J. Mass Spectrom. 21, 1–12 (2015) 3
Producing Th 3+ and its loading into the ion trap for subsequent spectroscopic study poses a significant challenge because thorium is a refractory metal with a high melting point and low vapor pressure.19–21 Besides, generating multiply ionized species is much less efficient than generating singly ionized ones.19 A method to produce triply charged ions of 229Th 3+ and 232Th 3+ from thorium nitride enriched by the 229Th isotope was recently described by the US team.19,22 The third 355 nm harmonics of a neodymium yttrium–aluminum–garnet laser irradiation was focused on the target positioned perpendicular to the trap. The ion trap operated synchronously with the laser irradiation. The method described allowed the production of multiply charged trapped thorium ion clouds, which could be cooled to form crystals for nuclear laser spectroscopy. The main disadvantages of the direct laser loading of thorium ions are the significant variation of the number of ions loaded into the trap and the significant sample consumption. Different approaches were considered as candidates for producing ions for thorium nuclear spectroscopy. Inductively coupled plasma (ICP) mass spectrometry can effectively extract ions directly from liquid solution of thorium salts.23 Laser time-of-flight mass spectrometry can be considered as a potentially useful method for producing and separating thorium ions.24–26 A highly effective method to produce triply charged ions is the use of electron-beam metal evaporation and an ionization source.27 Compactness and sensitiveness are also important advantages of the quadrupole mass analyzer, which is why a number of miniaturized quadrupole-based instruments were constructed for environmental applications. 28–30 The modern low-cost rapid-prototyping technology was recently described for quadrupole mass filter31 and linear trap32 instrumentation. Owing to high reactivity of thorium ions, collision cooling is undesirable. Even when operating with helium as the collision gas, thorium ions efficiently react with impurities. In addition, thorium is a refractory metal with a high melting point and low vapor pressure, so high-energy methods are required to ionize it. For this reason, thorium ions are produced in a broad range of energy. Which is why a vacuum trapping method is of interest for development that would be effective for a variable range of energies. In the present work, for the first time we describe a method and instrumentation for thorium nuclear spectroscopy based on an electron-beam metal evaporation and ionization ion source with a quadrupole electrode system of linear configuration. The method and instrumentation could potentially improve the effectiveness of thorium sampling, minimize the requirements for sample quantity and decrease the danger of radioactive contamination, and could be used in optical clocks and quantum logic applications. The novelty of the described approach is the use of an electron-beam metal evaporation and ionization source with a linear quadrupole ion trap as the mass- and energy-selecting instrument that can improve the effectiveness of producing and trapping thorium ions with various charges.
Figure 2. Schematic of linear ion-trap electrodes installed with (a) an electron-beam evaporation and ionization source and (b) a surface ionization source (all dimensions are in centimeters).
Experimental A schematic diagram of the linear quadrupole ion trap designed and built in our laboratory is shown in Figure 1(a). Preliminary ideas of the method are described in the accepted patent application.33 A general view of the instrument is shown in Figure 1(b). The instrument includes an ion source, linear ion trap and vacuum system. Preliminary tests of the instrument were recently described.34
Ion source Two ion sources were used in the study. The first ion source construction is based on the commercial ultra-high vacuum evaporator EFM2 (Focus, Taunusstein, Germany) normally used in epitaxial growth processes. In the original construction the electron beam was emitted from circular 0.1 mm diameter tungsten filament. The energy of the beam was adjusted from 300 eV to 1000 eV. The maximum registered emitted electron current reached a value of 150 mA. The electron beam heated a graphite crucible containing the sample and vaporized it. The vaporized sample was ionized by a beam of fast electrons, resulting in the appearance of ions of various charge states. The ion beam was formed in a direction opposite to the electron beam [Figure 2(a)]. The external stainless-steel collimator of the evaporator was removed and replaced by an assembly of ion-lens electrodes. This allowed us to improve the ion-beam focusing. In addition, a sufficient energy of the electron impact resulted in the effective production of triply charged thorium ions. The source was installed at a distance of 20 cm from the linear ion trap. The second ion source was a home-made device based on a surface ionization principle. In some experiments it was installed on a grounded decelerating grid [Figure 2(b)] 0.2 cm distant from the left side.
Linear ion trap The segmented linear ion trap consists of five successive quadrupole sections, including a 15 mm radiofrequency (RF)-only entrance quadrupole section (Q1), a 150 mm quadrupole mass
4
Multisectional Linear Ion Trap and Novel Loading for Electron and Nuclear Transition Spectroscopy
filter (Q2), a 100 mm RF-only quadrupole ion guide (Q3), a 50 mm RF-only ion trap (Q4) and a 30 mm RF-only exit quadrupole section (Q5). Each quadrupole section consists of stainless-steel cylindrical rods with a diameter of 8 mm mounted at a diagonal distance of 7.1 mm between the surfaces of the rods (see Figure 2). Direct current (DC) bias voltages (U bias ) are applied independently to all the electrodes of all the quadrupole sections (Q1–Q5) to create some potential barrier for the injected ions. For the quadrupole mass filter (Q2), the DC offset voltage is also applied in regard to the potential of the Q2 axis and the actual voltages for both couples of equipotential rod electrodes can be described as Ubias + [U + V cos(wt)] and Ubias – [U + V cos(wt)], where U is the DC offset voltage, V is the alternative current amplitude and t is time. In fact, the Q2 power supplies produce mathematically calculated U bias + U and U bias – U voltages in such a way as to satisfy the equality 2U/V = a/q for every filtered mass. The value V was proportional to the filtered m/z and was equal to 338 V for m/z = 250. For the described quadrupole mass filter, the unit-resolving power the a/q value was equal to 0.18 for m/z = 200–250 and was slightly varied for lower masses to optimize the sensitivity. No DC offset voltage was applied for the RF-only quadrupole sections (Q1, Q3–Q5) and actual voltages for both couples of the equipotential rod electrodes are Ubias + V cos(wt) and Ubias – V cos(wt). Magnitudes of RF voltage for the sections
Q1, Q3, Q4 and Q5 were equal to the value V for the quadrupole mass filter (Q2). Two decelerating grids are located in the front of the entrance quadrupole section (Q1) to adjust the kinetic energy of the ions. The exit quadrupole section (Q5) is followed by two flat electrodes and an electrostatic energy analyzer (ESA). The first and second flat electrodes have 2 mm and 1 mm, respectively, diameter apertures coaxial with the ion trap axis. A preliminary numeric simulation has shown that the proposed configuration of the trap can effectively filter thorium ions and localize them in the area of spectroscopic study. The quadrupole sections with the required electronics were built by Shibboleth LTD (Ryazan, Russia). The choice of lengths and the diameter of the rods, as well as the diagonal distance between the surfaces of the rods, were based on the dimensions of the commercial quadrupole mass filters available from the manufacturer. The secondary electron multiplier (SEM) VEU-6 (Baspik, Russia) was installed behind the ESA. High-energy filtration by the ESA allows one to detect ions in a specified kinetic energy range. The constructed ion trap allows both the mass and energy filtration of ions. The mass filtration is performed by the quadrupole mass filter (Q2). The bias voltage of the entrance quadrupole section (Q1) decreases the kinetic energy of the ions that enter the quadrupole sections (Q1–Q5) and rejects ions with undesirable low energies. Ions with energy exceeding the kinetic energy window of 10 eV cannot be sorted by the quad-
Table 1. Instrumental parameters of the linear quadrupole ion trap system.
Part Ion source
Quadrupole ion trap
Parameter
Setting
Anode voltage, V
300–1000
Filament current (automatically adjusted to allow required emission current), A
1–2
Electron emission current, mA
≤150
Ion current, µA
≤20
Ion lens voltage, V
–150
Rod diameters, mm
8
Inscribed diameter, mm
7.1
Number of quadrupole sections
5
Length Q1/Q2/Q3/Q4/Q5, cm
1.5/15/10/5/3
Bias voltage for Q1–Q5, V
≤400
Q1–Q5 frequency, MHz
1.22
Q1–Q5 magnitude of RF voltage, V
Vacuum
2.7–338
Q2 DC offset voltage, V
0.24–30.4
Mass range for Q2, m/z
2–250
SEM detector voltage, V
2500
Turbo pump HiPace 300M (Pfeiffer), L s–1
300
Rotary vane pump RV5 (Edwards), m3 h–1
5 –1
Titanium sublimation pump ST22 (Vacuum Generators Inc.), L s ∙cm Ion pump VacIon Plus 300 (Varian), L s–1
–2
4 300
Ion-trap pressure range in the described experiments, torr
2 × 10 –7 to 2 × 10 –8
Available ion-trap pressure range, torr
2 × 10 –7 to 2 × 10 –10
V.I. Troyan et al., Eur. J. Mass Spectrom. 21, 1–12 (2015) 5
rupole mass filter Q2 and pass freely through all quadrupole sections. Without the energy filtration, these ions could reach a detector and decrease the mass resolution. During a mass spectra recording, the energy filtering, performed by the ESA, removes these unwanted ions, thus increasing the resolution. It should also be noted that the use of the ESA decreases the transmission of ions by one order of magnitude during a mass spectra recording. The ESA does not affect the mass-selected ion-trapping process, does not decrease the transmission of ions and cannot remove higher energy ions. For this reason, special attention has to be paid to trap the ions in the kinetic energy window below 10 eV in order to be sure that the ions are mass selected. In the transport mode all the ions pass through the ion trap and ESA toward the SEM detector that allows energydependent mass spectrum recording. This mode allows calibration of the quadrupole mass filter (Q2), choosing the energy maximum of the selected isotope and the estimation of the ion current intensity for the required charge state of the chosen isotope. In the trapping mode, DC bias voltages on the quadrupole sections are changed in such a way to trap and localize the mass- and energy-filtered required charge state of the chosen isotope in the 5 cm ion trap (Q4).
Vacuum The trap was installed in a stainless-steel vacuum chamber with a diameter of 200 mm (Figure 1). The window in the top of the chamber was made for the observation of fluorescence light emitted by the trapped ions. Six windows around the circumference of the chamber were made for the transmission of laser beams, allowing laser cooling and further excitation of the trapped ions. Four windows were designed for Doppler cooling of the ions, one window for resolved sideband cooling and one for the excitation of fluorescence, and the last two windows for the ultra-high resolution spectroscopic study of electron, atom and nuclear transitions. For these purposes, we plan to use a recently described approach 35 based on 1087 nm lasers for the first-stage cooling, 984 nm and 690 nm lasers for the second stage cooling and the excitation of fluorescence, a tunable laser for an isomeric transition search in the VUV region and a charge coupled device (CCD) camera (Figure 1). In the current study the lasers were not installed. To maintain a pressure down to 10–10 torr the vacuum system consisted of a 300 L s–1 ion pump (VacIon Plus 300, Varian), a 4 L s –1∙cm –2 titanium sublimation pump (ST22, Vacuum Generators Inc.), a 210 L s–1 turbo pump (TMH 261, Pfeiffer, Germany), a rotary vane pump (RV5, Edwards, UK) and a 300 L s–1 turbo pump (HiPace 300M, Pfeiffer, Germany). The assembly of the first four pumps represents the original pumping line of the UHV Surface Science Systems Multiprobe MXPS RM VT AFM-25 (Omicron Ltd., Germany). The assembly is connected to the trap vacuum chamber by an ultra-high vacuum gate valve. The HiPace 300M turbo pump is connected to a trap vacuum chamber by another ultra-high vacuum gate valve. The experiments were mainly performed at 10–7–10–8 torr in
order to avoid spending time heating the instrument after loading the sample. A few experiments were done to prove that the instrument could operate at 10–10 torr. Instrumental parameters of the linear quadrupole ion trap are summarized in Table 1. If not specified otherwise, the conditions listed in the table are used for all the experiments described in this work.
Software Control over the linear quadrupole ion trap and data processing were performed with a special code developed by Shibboleth Ltd (Riazan, Russia). The code provides full control of the power supplies, control over the RF generators, assignment of a data-collection algorithm and data-collection control, as well as data representation and visualization (mass spectrum, reading parameters of the experiment, calibration of the mass scales).
Chemicals The linear ion trap was tested using solid gold and thorium, as well as thorium nitride, RbCl and CsI.
Safety considerations High voltages are applied to the ion source and detector. Care must be taken to avoid electric discharges and injury. In the current work, only 232Th samples were used because of its lower radioactivity compared with the 229Th isotope. When using 229Th samples special provisions must be made to avoid exceeding the maximum allowed laboratory level of radio activity of 500 Bq.
Results and discussion
Energy distribution in ion beams produced from thoriated tungsten
An important challenge for both the optical spectroscopy of electron and nuclear transitions and for quantum logic applications is associated with the necessity of getting ions from different metals, including refractory ones, which are produced with a wide range of energy distributions. Furthermore, the ions have to be cooled in an ultra-high vacuum environment accessible for laser irradiation. The effectiveness of trapping the ions is expected to be dependent on the shape of the energy distribution and the position of its maximum. Estimation of the distribution is complicated by the fact that many parameters might affect the energy distribution, including the ionization energy and cross-section, electron, ion and atom flux density distribution etc. In order to define experimentally the expected energy distribution of the ions produced from thoriated tungsten rods at different conditions, some preliminary measurements were made before the development of the ion trap. The energy distribution of the ion beam obtained was estimated with the use of a simple planar wafer grid energy analyzer installed in front of the ion source. The analyzer consisted of three parallel
6
Multisectional Linear Ion Trap and Novel Loading for Electron and Nuclear Transition Spectroscopy
Figure 3. Energy distribution, obtained with the simplest energy analyzer, based on planar wafer grids, in an ion beam produced by electron-beam metal evaporation and ionization of thoriated tungsten rods for electron-beam energies equal to (a) 1000 eV, (b) 700 eV, (c) 500 eV, (d) 300 eV. Table 2. Parameters of the energy distribution obtained for an ion beam produced by electron-beam metal evaporation and ionization of thoriated tungsten rods.
a
Accelerating voltage, V
Electron emission current, mA
Maximum of energy distribution,a V
Width of energy distribution at half-maximum,a V
1000
59.9
156 (16)
157 (10)
700
32.8
154 (16)
90 (10)
500
18.4
140 (14)
100 (10)
300
6.6
122 (8)
90 (5)
(Relative standard deviation).
ion grids and an ion collector separated by 2 mm. The first and the third grids were grounded and the second one was fed by a potential that reflected low-energy ions. The ion current j measured by the ion collector was recorded as a function of potentials V applied to the second grid and the energy distribution was derived as –dj/dV. The energy distribution in the ion beam produced by the electron-beam evaporation and ionization of thoriated tungsten rods was studied for electron-beam energies equal to 1000 eV, 700 eV, 500 eV and 300 eV [see Figure 3(a)–(d), respectively]. The electron-beam energy is defined by the voltage applied between the cathode (a filament) and the anode (a thoriated tungsten rod). The filament current was kept equal 1.7 A. Table 2 shows the parameters of the energy distribution obtained for an ion beam produced by electron-beam metal evaporation and ionization of the thoriated tungsten rods. As was expected, a higher electron emission current was obtained at higher electron-beam energies because of the constant filament current. The decrease of accelerating voltage from 1000 eV down to 500 eV does not significantly affect the maximum of the energy distribution and results in a 10% decrease (from 155 eV down to 140 eV). The width of the
energy distribution at half-maximum was estimated as 100 eV for a 500–700 eV accelerating voltage.
Figure 4. A schematic representation of ion energy versus their position in the ion trap showing the energy-scan approach. The shaded area corresponds to all the ions leaving ion source. The hatched area in the ΔUk kinetic energy range corresponds to the ions normally mass filtered at kinetic energy below 10 eV. The bias voltage (Ubias) for all the quadrupole sections is kept the same. The bias voltage is varied from Ubias = 0 (a) through intermediate values (b) to Ubias = Uh – ΔUk.
V.I. Troyan et al., Eur. J. Mass Spectrom. 21, 1–12 (2015) 7
100 90 80 70 60 50 40 30 20 10 0
Ion current, %
Au+
Au++ Au+++
100
120 140 160 180 200 Kinetic energy of ions, eV
Figure 6. Mass-selective energy distribution for singly, doubly
Figure 6. Mass-selective energy distribution for singly-, doubly- and triply-charged and triply gold Au+ (m/z 197), Au++ (m/z 98.5) + charged ions of ions of gold Au+++ (m/z 197), Au++ (m/z 98.5), Au+++ (m/z 65.7). Plotted by moving and Au (m/z 65.7), plotted by a moving average with a step of average with step 5 eV. 5 eV.
Figure 5. Mass spectra of (a) 138Cs (10% peak width 0.7 Da) and 85 Rb and 87Rb (10% peak width 1 Da).
Energy distribution obtained for ions of gold In order to obtain the best effectiveness of the trap, the loaded ions should have energies close to the maximum of the energy distribution. That is why the energy distribution of potentially useful ions has to be studied. Figure 4 gives a schematic representation of the energy-scan approach. Mass spectra are recorded sequentially at various kinetic ion energies with an energy range width (ΔUk) equal to 10 eV. This value (ΔUk = 10 eV) is governed by the fact that a reliable 1 Da filtration of the existing quadrupole mass filter is obtained with ions of kinetic energy below 10 eV. This fact was demonstrated experimentally by installing the surface ionization ion source and adjusting the kinetic energy of ions that enter the quadrupole mass filter (Q2). This was possible because the ion trap design allowed one to use it with different ion sources, including ones that effectively produce ions potentially usable for quantum logic applications. Figure 5 shows cesium and rubidium surface ionization mass spectra obtained reproducibly for ions with a kinetic energy below 10 eV. The width at 10% of the mass peak intensity for 138Cs, 85Rb and 87Rb is below 1 Da. The ion-beam intensity for gold was studied in the energy range between 0 and U h = 250 eV, where U h is the upper boundary of the scanned energy range. Ion deceleration was achieved by a quadrupole bias voltage, Ubias, applied to all the quadrupole sections (Q1, Q2, Q3, Q4 and Q5). Ubias varied from
0 to Uh – ΔUk = 240 eV [Figure 4(a) and (b)] with an increment of 1 eV allowing one to record the ion-beam intensity in an energy range between 0 and Uh. The measurements for Ag+, Ag++ and Ag+++ were performed within 48 minutes. The data record for a selected charge state of the ions takes around 15 minutes. Ions from the hatched zone from Ubias up to Ubias + ΔUk are assuredly mass filtered, because their kinetic energy is below 10 eV while they are in the quadrupole mass filter (Q2). The decelerating grids reject the ions that leave the ion source with a kinetic energy below Ubias. Ions that leave the ion source with a kinetic energy higher than Ubias + ΔUk have a probability of passing through the quadrupole mass filter (Q2) without mass filtration that is dependent on their energy. To exclude the influence of these ions on the results of mass analysis, the ESA prevents the transport of high-energy ions to the detector. The energy distribution in the ion beam produced by the electron-beam evaporation and ionization of pure gold was studied with an electron-beam accelerating energy equal to 600 V. This electron-beam energy was chosen as a result of optimization of the signal intensity and sample consumption. The signal from the gold sample normally appeared at an electron-beam energy of 300 V. At higher energies the time of sample consumption significantly decreases. Figure 6 shows an example of a typical energy distribution for ions of gold. No significant differences were found in the values of the energy distribution maxima for the studied charge states of gold ions. The maximum of the energy distribution for singly charged ions corresponds to 154 eV. The width of the energy distribution at half-maximum for singly charged ions is equal to 29 eV. The maximum of the energy distribution for doubly charged ions corresponds to 159 eV. The width of the energy distribution at half-maximum for doubly charged ions equals 29 eV. The maximum of the energy distribution for triply charged ions corresponds to 155 eV. The width of the energy distribution at half-maximum for triply charged ions equals 27 eV. Similar values were obtained for ions produced from thoriated tungsten. The main differences were that an electronbeam accelerating energy value of 800 V was found to be the best for producing thorium ions from different matrixes. In
8
Multisectional Linear Ion Trap and Novel Loading for Electron and Nuclear Transition Spectroscopy
Table 3. Relative intensities of ions observed for gold and thorium normalized to those of singly charged atomic ions.
A+++
A++
Gold
0.04 (m/z 65.7)
0.24 (m/z 98.5)
1 (m/z 197)
Not observed
Thorium
0.18 (m/z 77.6)
0.32 (m/z 116)
1 (m/z 232)
2.52 (m/z 248)
addition, different ratios between the intensities of the charge states were observed for ions of gold and thorium (see Table 3). Figure 7 demonstrates the typical mass spectra for gold and thorium obtained with electron-beam evaporation and ionization of pure gold and thorium nitride.34
Ion trapping The trapping algorithm should allow operation under ultrahigh vacuum conditions, localization of ions in the area of laser irradiation, containing the maximum amount of ions from beams with a wide energy distribution and keeping ions in the laser irradiation area for a duration of up to a day. Two trapping approaches were studied (see Figure 8). In both cases, during the ion-trapping process the quadrupole mass filter (Q2) was
Figure 7. Mass spectra34 of (a) Th and (b) Au.
A+
AO
operated in a mass-selective mode in order to transmit the selected target isotope only. The trapping must be performed in a kinetic energy range below 10 eV to allow reliable 1 Da filtration. The first trapping approach [Figure 8(a)] is characterized by a relatively large area of trapping that allows one to load ions effectively from low-intensity beams. First [Figure 8(a), (1)], ions are decelerated by a bias potential down to a kinetic energy below ΔUk = 10 eV. The bias potential, Ubias, is chosen to be close to the maximum of the energy distribution of the required isotope. To trap ions in the required energy range [Figure 8(a), (2)] the bias voltage of exit quadrupole section (Q5) was increased up to 10 V (ΔUk) while keeping the bias potential of the other quadrupole sections constant. Ions with energies higher than
V.I. Troyan et al., Eur. J. Mass Spectrom. 21, 1–12 (2015) 9
Figure 8. A schematic representation of the algorithm with single-stage (a) and multistage (b) ion trapping. The shaded area correFigure 8. A schematic representation of the algorithm with single-stage (a) and multistage (b) ion trapping. sponds to all the ions leaving the ion source. The hatched area in the ΔUk kinetic energy range corresponds to the ions normally mass The shaded area corresponds leaving ion are source. Theto hatched areaenergy in ΔU energy range k kinetic filtered at kinetic energy below 10 eV.to Noall ionsions of the targetedthe isotope supposed have a kinetic higher than upper boundary of energy rangeto (Uhthe ). ions to be mass-filtered normally in 10 eV. No ions of targeted isotope is supposed to have corresponds kinetic energy higher than upper boundary of energy range (Uh).
the energy of mass filtration overcome this barrier and leave the trap. Ions with energies below ΔUk turn round to face the direction of the entrance quadrupole section. Then [Figure 8(a), (3)] the potential of the entrance quadrupole section (Q1) was increased up to 10 V (ΔUk) in order to isolate the ions with energies below ΔUk in the region between the entrance (Q1) and the exit (Q2) quadrupole sections. Sometime after turning off the ion source [Figure 8(a), (4)] all high-energy ions leave the trap and only the target isotopes remain in the trapping area. After that, the ions are expected to be cooled by laser irradiation. Then, the bias voltage of the ion-trap section (Q4) is decreased [Figure 8(a), (5)] in order to collect all the trapped ions in the area of laser beam excitation (Q4). Every stage of the first trapping approach takes 5–10 s as conditioned by software limitation. In another trapping approach studied (multistep) the first [Figure 8(b), (1)] and the second [Figure 8(b), (2)] stages are the same as in the previous case. Then the bias voltages of the third and the fourth quadrupole (Q3, Q4) sections [Figure 8(b, (3)] are decreased step-by-step by 0.1 V up to 20 times (the number of trapped ions is proportional to the number of steps). Each step of the third stage takes 10 s, which is why stage 3 was performed within 200 s. The process isolates the ions in the region between the quadrupole mass filter (Q2) and the exit quadrupole section (Q5). Then the bias voltages [Figure 8(b), (4)] of both the entrance quadrupole section (Q1) and the quadrupole mass filter (Q2) are increased up to the value Uh in order to isolate ions with energies below ΔUk in the
region between the quadrupole mass filter (Q2) and the exit quadrupole section (Q5). After that the ions are expected to be cooled by the laser irradiation and the bias voltage of ion trap quadrupole section (Q4) is decreased [Figure 8(b), (5)] in order to collect all the trapped ions in the area of laser beam excitation (Q4) in the same manner as in the first trapping approach [Figure 8(a), (5)]. Each stage except the third one of the second trapping approach takes 5–10 s. Third stage was performed within 200 s. Times of other stages were conditioned by software limitation. The current configuration of the instrument does not allow one the direct control of the number of trapped ions. The planned equipping of the instrument with lasers will make it possible to control the number of trapped ions inside the trap by measuring the intensity of their fluorescence. We estimated the number of ions trapped by SEM counting the ions extracted from the trap after stage (4) of both algorithms [Figure 8(a) and (b)]. To extract the ions the bias voltage of Q5 section was decreased to 0 V. We expected that ion losses in the energy analyzer could cause more than an order of magnitude reduction in the number of ions counted by the SEM. Both trapping approaches were studied with the surface 33 ionization source [Figure 2(b)]. This ion source produced Rb+ + and Cs ion beams with narrow energy distributions characterized by a 9.5 eV maximum and 0.5 eV full-width at halfmaximum. The time of extraction for a particular ion was measured as a time interval between the moment of switching the Q5 bias voltage and the ion detection event. The multistep
10
Number of ions per second, s‐1
3000 2500
Multisectional Linear Ion Trap and Novel Loading for Electron and Nuclear Transition Spectroscopy
evaporation and ionization source with a narrow energy range and install it very close to the ion trap.
2000 1500
Conclusion
1000
A multisectional linear ion trap was constructed for the study of the optical spectroscopy of electron and nuclear transitions as well as for potential use in quantum logic applications. The 0 trap was shown to allow both the mass and energy filtration 0 5 10 15 20 25 30 35 Time, s of ions produced both by an electron-beam evaporation and Figure 9. Number of from ions extracted frommultistep the traptrapping using the ionization Figure 9. Number of ions extracted the trap using approach versus time of source and by a surface ionization source. Energy multistep approach time was of extraction extraction for cesium ions.trapping To extract the ions,versus Q5 biasthe voltage decreased down to 0 V. Time of is especially important for refractory metals ions filtration for cesium To extractasthe ions, the Q5between bias voltage was of switching the Q5 extraction for particular ionions. was measured a time interval the moment produced by sources with wide energy distributions. An initial bias voltage anddecreased the ion detection event. down to 0 V. The time of extraction for a particular study has shown that the proposed instrument and methods ion was measured as a time interval between the moment of allow us effectively to obtain singly and triply charged thorium switching the Q5 bias voltage and the ion detection event. ions as well as to trap them for future study. The multistep trapping algorithm was found to be approximately an order trapping algorithm was found to be approximately an order of magnitude more effective compared with an alternative of magnitude more effective compared to the single-stage approach with the other ion beam parameters were the same. approach, while keeping the other ion beam parameters the The stability of the ion-trap operation was demonstrated by same. In particular the second multistep trapping approach the fact that after storage of the ions in the trap for 10 h no allowed approximately 4700 Cs+ ions to be extracted from decrease of ion number was found. the trap (Figure 9), while the first approach allowed one to extract approximately 450 ions of Cs+. The DC bias voltage could decrease ion transmission through the Q2 section, which could be a reason why trapping is less efficient when using a single-stage algorithm. The effect of decreasing the This work was financially supported by the Rosatom Nuclear transmission in the Q2 section has not been studied quan- Energy State Corporation (project N4b.43.90.13.1136), by the titatively. Another reason could be associated with the fact Russian Foundation for Basic Research (projects No.14-08that multistep trapping algorithm preferably takes ions of low 00487_a) and by the Ministry of Science and Education of kinetic energy. Russia (project No. 3.1803.2014/K). The authors are grateful to The stability of the ion-trap operation was studied by Dr. Eugeny Y. Chernyak for fruitful discussions and kind assiscounting the number of ions extracted after a defined storage tance during different stages of the project. time. After 3 h storage time we did not observe any decrease in 34 ion number. After storage of the Rb+ ions in the trap isolated for 10 h, again no decrease in ion number was found. The multistep trapping algorithm was also studied with the electron-beam evaporation and ionization source for trapping Au+ ions with 1. B.J. Bloom, T.L. Nicholson, J.R. Williams, S.L. Campbell, the bias voltage increased up to a value close to the maximum M. Bishof, X. Zhang, W. Zhang, S.L. Bromley and J. Ye, of the energy distribution. The number of ions counted by “An optical lattice clock with accuracy and stability at the SEM was about 40. There are two explanations why a two the 10 −18 level”, Nature 506, 71 (2014). doi: http://dx.doi. orders of magnitude greater number of ions is trapped with org/10.1038/nature12941 the surface ionization source compared to the electron-beam 2. M. Takamoto and H. Katori, “Spectroscopy of the 1S 0 – 3P 0 evaporation and ionization source. First, the surface ionization clock transition of 87Sr in an optical lattice”, Phys. Rev. source was installed very close to the ion trap at a distance Lett. 91, 223001 (2003). doi: http://dx.doi.org/10.1103/ of 2 mm from the left side of the grounded decelerating grid, PhysRevLett.91.223001 while the commercial electron-beam evaporation and ioniza- 3. T. Ido, T.H. Loftus, M.M. Boyd, A.D. Ludlow, K.W. tion source was installed 20 cm from the grounded decelerHolman and J. Ye, “Precision spectroscopy and densityating grid (Figure 2). Second, the energy distribution of the Au+ dependent frequency shifts in ultracold Sr”, Phys. Rev. ions is more than an order of magnitude wider than the energy Lett. 94, 153001 (2005). doi: http://dx.doi.org/10.1103/ distributions of Rb+ and Cs+ ions produced by the surface ioniPhysRevLett.94.153001 zation source. Estimates show that the contributions of these 4. Z.W. Barber, C.W. Hoyt, C.W. Oates, L. Hollberg, A.V. two effects decrease the number of trapped ions by more Taichenachev and V.I. Yudin, “Direct excitation of the forthan three orders of magnitude. In order to improve the ion bidden clock transition in neutral 174Yb atoms confined to transmission it is planned next to develop an electron-beam 500
Acknowledgments
References
V.I. Troyan et al., Eur. J. Mass Spectrom. 21, 1–12 (2015) 11
an optical lattice”, Phys. Rev. Lett., 96, 083002 (2006). doi: http://dx.doi.org/10.1103/PhysRevLett.96.083002 5. T. Rosenband, P.O. Schmidt, D.B. Hume, W.M. Itano, T.M. Fortier, J.E. Stalnaker, K. Kim, S.A. Diddams, J.C.J. Koelemeij, J.C. Bergquist and D.J. Wineland, “Observation of the 1S 0 →3P 0 clock transition in 27Al+”, Phys. Rev. Lett. 98, 220801 (2007). doi: http://dx.doi. org/10.1103/PhysRevLett.98.220801 6. N. Ashby, T.P. Heavner, S.R. Jefferts, T.E. Parker, A.G. Radnaev and Y.O. Dudin, “Testing local position invariance with four cesium-fountain primary frequency standards and four NIST hydrogen masers”, Phys. Rev. Lett. 98, 070802 (2007). doi: http://dx.doi.org/10.1103/ PhysRevLett.98.070802 7. T. Rosenband, D.B. Hume, P.O. Schmidt, C.W. Chou, A. Brusch, L. Lorini, W.H. Oskay, R.E. Drullinger, T.M. Fortier, J.E. Stalnaker, S.A. Diddams, W.C. Swann, N.R. Newbury, W.M. Itano, D.J. Wineland and J.C. Bergquist, “Frequency ratio of Al+ and Hg+ single-ion optical clocks; metrology at the 17th decimal place”, Science 319, 1808 (2008). doi: http://dx.doi.org/10.1126/science.1154622 8. C.J. Campbell, A.G. Radnaev, A. Kuzmich, V.A. Dzuba, V.V. Flambaum and A. Derevianko, “Single-ion nuclear clock for metrology at the 19th decimal place”, Phys. Rev. Lett. 108, 120802 (2012). doi: http://dx.doi.org/10.1103/ PhysRevLett.108.120802 9. B.R. Beck, J.A. Becker, P. Beiersdorfer, G.V. Brown, K.J. Moody, J.B. Wilhelmy, F.S. Porter, C.A. Kilbourne and R.L. Kelley, “Energy splitting of the ground-state doublet in the nucleus 229 Th”, Phys. Rev. Lett. 98, 142501 (2007). doi: http://dx.doi.org/10.1103/PhysRevLett.98.142501 10. E. Peik and C. Tamm, “Nuclear laser spectroscopy of the 3.5 eV transition in Th-229”, Europhys. Lett. 61, 181 (2003). doi: http://dx.doi.org/10.1209/epl/i2003-00210-x 11. E.V. Tkalya and L.P. Yatsenko, “Creation of inverse population in the 229 Th ground-state doublet by means of a narrowband laser”, Laser Phys. Lett. 10, 105808 (2013). doi: http://dx.doi.org/10.1088/1612-2011/10/10/105808 12. M.S. Safronova, U.I. Safronova and C.W. Clark, “Relativistic all-order calculations of Th, Th+, and Th2+ atomic properties”, Phys. Rev. A 90, 032512. doi: http://dx.doi.org/10.1103/PhysRevA.90.032512 13. W.D. Phillips, “Nobel Lecture: Laser cooling and trapping of neutral atoms”, Rev. Mod. Phys. 70, 721 (1998). doi: http://dx.doi.org/10.1103/RevModPhys.70.721 14. P.O. Schmidt, T. Rosenband, C. Langer, W.M. Itano, J.C. Bergquist and D.J. Wineland, “Spectroscopy using quantum logic”, Science 309, 795 (2005). doi: http://dx.doi. org/10.1126/science.1114375 15. J.P. Home, M.J. McDonnell, D.J. Szwer, B.C. Keitch, D.M. Lucas, D.N. Stacey and A.M. Steane, “Memory coherenceof a sympathetically cooled trapped-ion qubit”, Phys. Rev. A 79, 050305 (R) (2009). doi: http:// dx.doi.org/10.1103/PhysRevA.79.050305 16. K. Zimmermann, M.V. Okhapkin, O.A. Herrera-Sancho and E. Peik, “Laser ablation loading of a radiofrequency
ion trap”, Appl. Phys. B. 107, 883 (2012). doi: http://dx.doi. org/ 10.1007/s00340-012-4884-1 17. O.A. Herrera-Sancho, M.V. Okhapkin, K. Zimmermann, C. Tamm, E. Peik, A.V. Taichenachev, V.I. Yudin and P. Głowacki, “Two-photon laser excitation of trapped 232 Th+ ions via the 402 nm resonance line”, Phys. Rev. A 85, 033402 (2012). doi: http://dx.doi.org/10.1103/ PhysRevA.85.033402 18. J.W. Hager, “A new linear ion trap mass spectrometer”, Rapid Commun. Mass Spectrom. 16, 512, (2002). doi: http://dx.doi.org/10.1002/rcm.607 19. C.J. Campbell, A.V. Steele, L.R. Churchill, M.V. DePalatis, D.E. Naylor, D.N. Matsukevich, A. Kuzmich and M.S. Chapman, “Multiply charged thorium crystals for nuclear laser spectroscopy”, Phys. Rev. Lett. 102, 233004 (2009). doi: http://dx.doi.org/10.1103/ PhysRevLett.102.233004 20. R.E. Honig and D.A. Kramer, “Vapor pressure data for the solid and liquid elements”, RCA Rev. 30, 285 (1969). 21. S. Aronson and A.B. Auskern, “Vapor pressure measurements on thorium nitrides”, J. Phys. Chem. 70, 3937 (1966). doi: http://dx.doi.org/10.1021/j100884a033 22. L.R. Churchill, M.V. DePalatis and M.S. Chapman, “Charge exchange and chemical reactions with trapped Th3+”, Phys. Rev. A, 83, 012710 (2011). doi: http://dx.doi. org/10.1103/PhysRevA.83.012710 23. V.I. Troyan, P.V. Borisyuk, R.R. Khalitov, A.V. Krasavin, Y.Y. Lebedinskii, V.G. Palchikov, S.S. Poteshin, A.A. Sysoev and V.P. Yakovlev, “Generation of thorium ions by laser ablation and inductively coupled plasma techniques for optical nuclear spectroscopy”, Laser Phys. Lett. 10, 105301 (2013). doi: http://dx.doi.org/10.1088/16122011/10/10/105301 24. A.A. Sysoev and A.A. Sysoev, “Can laser ionization TOF mass spectrometry be a promising alternative to LA ICP and GD mass spectrometry in the field of elemental analysis of solids?”, Eur. J. Mass Spectrom. 8, 213 (2002). doi: http://dx.doi.org/10.1255/ejms.494 25. A.A. Sysoev, “Time-of-flight analysers with sector fields: advances and prospects”, Eur. J. Mass Spectrom. 6, 501 (2000). doi: http://dx.doi.org/10.1255/ejms.380 26. A.A. Sysoev, A.A. Sysoev, S.S. Poteshin, V.I. Piatakhin, I.V. Shchekina and A.S. Trofimov, “Direct sampling time-offlight mass spectrometers for technological analysis”, Fresenius J. Anal. Chem. 361, 261 (1998). doi: http://dx.doi. org/10.1007/s002160050876 27. R.N. Evtukhov, S.F. Belykh and I.V. Redina, “Transition and rare-earth metal ion sources”, Rev. Sci. Instrum. 63, 2463 (1992). doi: http://dx.doi.org/10.1063/1.1142912 28. M.B. Wise and M.R. Guerin, “Direct sampling MS for environmental screening”, Anal. Chem. 69, 26A (1997). doi: http://dx.doi.org/10.1021/ac971504r 29. A.J. White, M.G. Blamire, C.A. Corlett, B.W. Griffiths, D.M. Martin, S.B. Spencer and S.J. Mullock, “Development of a portable time-of-flight membrane inlet mass spectrometer for environmental analy-
12
Multisectional Linear Ion Trap and Novel Loading for Electron and Nuclear Transition Spectroscopy
sis”, Rev. Sci. Instr. 69, 565 (1998). doi: http://dx.doi. org/10.1063/1.1148695 30. R.A. Ketola, J.T. Kiuru, V. Tarkiainen, T. Kotiaho and A.A. Sysoev, “Comparison of analytical performances of a micro-array quadrupole instrument and a conventional quadrupole mass spectrometer equipped with membrane inlets”, Rapid Commun. Mass Spectrom. 17, 753 (2003). doi: http://dx.doi.org/10.1002/rcm.973 31. B. Brkic, N. France, A.T. Clare, C.J. Sutcliffe, P.R. Chalker and S. Taylor, “Development of quadrupole mass spectrometers using rapid prototyping technology”, J. Am. Soc. Mass Spectrom. 20, 1359 (2009). doi: http://dx.doi.org/10.1016/j.jasms.2009.03.025 32. A.T. Clare, L. Gao, B. Brkic, P.R. Chalker and S. Taylor, “Linear ion trap fabricated using rapid manufacturing technology”, J. Am. Soc. Mass. Spectrom. 21, 317 (2010). doi: http://dx.doi.org/10.1016/j.jasms.2009.10.020
33. V.I. Troyan, A.V. Krasavin, P.V. Borisyuk, V.G. Palchikov,
A.A. Sysoev and S.S. Poteshin, “A method for production and detection of thorium-229 and thorium-232 with different charge states”, Accepted Russian patent proposal 2013154617 from 09.12.2013. 34. V.I. Troyan, P.V. Borisyuk, O.S. Vasiliev, A.V. Krasavin, S.S. Poteshin, A.A. Sysoev, D.M. Chernyshev, S.I. Donchenko and V.G. Palchikov, “Quadrupole Paul ion trap in the multi-charged thorium ions optical spectroscopy system for development of nuclear frequency standard”, Meas. Tech. 57, 777 (2014). doi: http://dx.doi.org/10.1007/s11018014-0535-8 35. C.J. Campbell, A.G. Radnaev and A. Kuzmich, “Wigner crystals of 229 Th for optical excitation of the nuclear isomer”, Phys. Rev. Lett. 106, 223001 (2011). doi: http:// dx.doi.org/10.1103/PhysRevLett.106.223001
