UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Collision induced dissociation and mass spectrometry with the TITAN Multiple-Reflection Time-of-Flight… Jacobs, Andrew 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_september_jacobs_andrew.pdf [ 10.16MB ]
Metadata
JSON: 24-1.0380561.json
JSON-LD: 24-1.0380561-ld.json
RDF/XML (Pretty): 24-1.0380561-rdf.xml
RDF/JSON: 24-1.0380561-rdf.json
Turtle: 24-1.0380561-turtle.txt
N-Triples: 24-1.0380561-rdf-ntriples.txt
Original Record: 24-1.0380561-source.json
Full Text
24-1.0380561-fulltext.txt
Citation
24-1.0380561.ris

Full Text

Collision Induced Dissociation andMass Spectrometry with the TITANMultiple-Reflection Time-of-FlightMass-SpectrometerbyAndrew JacobsB.Sc., Benedictine College, 2017A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinThe Faculty of Graduate and Postdoctoral Studies(Physics)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)August 2019© Andrew Jacobs 2019The following individuals certify that they have read, and recommend tothe Faculty of Graduate and Postdoctoral Studies for acceptance, the thesisentitled:Collision Induced Dissociation and Mass Spectrometry with the TITANMultiple-Reflection Time-of-Flight Mass-Spectrometersubmitted by Andrew Jacobs in partial fulfillment of the requirements forthe degree of Master of Science in Physics.Examining Committee:Dr. Jens Dilling, Department of Physics and AstronomySupervisorDr. John Behr, Department of Physics and AstronomySupervisory Committee MemberiiAbstractTo better understand nuclear structure, high-precision mass spectrometry ofradioactive nuclei is required. However, as the nuclei of interest move furtheraway from stability, the half-lives and production yields drop. Furthermore,the cocktail beams are frequently contaminated with unwanted isobars andmolecules which can obscure the species of interest and impede the mea-surement. To overcome these obstacles, the Multiple-Reflection Time-of-Flight Mass-Spectrometer (MR-ToF-MS) was commissioned at TRIUMF’sIon Trap for Atomic and Nuclear science (TITAN). The device can purifybeams through the process of mass selective re-trapping and provide fast,precise, and high-sensitivity mass measurements. These two modes of oper-ation can be used sequentially to conduct mass measurements of exceedinglycontaminated beams. Furthermore, the technique of collision induced dis-sociation (CID) has been investigated and developed to allow for furthersuppression of molecular contamination. With CID, the molecular contam-ination can be suppressed by one to two orders of magnitude. This devel-opment was undertaken following a Ne experiment in which a large amountof molecular contamination was found. However, the measurement capabil-ities of the MR-ToF-MS was demonstrated during the mass measurementcampaign of neutron-rich 24−26Ne in which relative uncertainties of approx-imately 10−7 were achieved. These measurements began to approach the“Island of Inversion” for the Ne isotope chain, and they further motivatethe investigation of this region in the future, ultimately ending with themeasurement of 31Ne.iiiLay SummaryTo understand the forces inside a nucleus, the mass must be known to ahigh degree of precision. This field of research is called ‘mass-spectrometry’.However, these measurements can be difficult to perform due to contami-nation. To clean out the contamination two processes have been developed.The first processes is called ‘re-trapping’ which purposefully throws away thecontamination. The second process is called ‘collision induced dissociation’which breaks apart unwanted molecules. As a result of these developments,mass-spectrometry of Ne has been performed.ivPrefaceThe research presented in the thesis was done as part of the TITAN collab-oration with contributions from collaborators at Justus-Liebig-Universita¨t-Gießen and GSI Darmstadt. In particular:• MR-ToF-MS commissioning was done by M. P. Reiter, S. Ayet SanAnde´s, C. Hornung, C. Will, and A. Finlay.• The planning and coordination of the Ne mass-measurement experi-ment was done by M. P. Reiter.• The data collection during the Ne experiment was performed by my-self, E. Dunling, J. Flowerdew, L. Graham, B. Kootte, A. A. Kwiatkowski,Y. Lan, E. Leistenschneider, M. Lykiardopoulou, V. Monier, I. Mukul,S. F. Paul, M. P. Reiter, J. L. Tracy, M. Vansteenkiste, and C. Will.• The analysis of the Ne data was done by myself and I. Mukul.• The planning and coordination of the CID experiment was done bymyself and S. F. Paul with assistance from K. Jayamanna.• The data collection during the CID experiment was performed by my-self, E. Dunling, D. Fusco, Z. Hockenbery, C. Izzo, B. Kootte, Y. Lan,E. Leistenschneider, M. Lyiardopoulou, I. Mukul, S. F. Paul, M. P.Reiter, and J. L. Tracy.• The analysis of the CID data was done by myself.Additionally, the results of both the Ne and CID experiments are to befeatured in an upcoming publication by myself as lead author and othercontributing authors.vTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiLay Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . viList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ixList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii1 Introduction to Mass Spectrometry . . . . . . . . . . . . . . 11.1 Motivation for Mass Spectrometry . . . . . . . . . . . . . . . 11.2 Methods of Mass Spectrometry . . . . . . . . . . . . . . . . . 71.2.1 Magnetic Deflection Mass Spectrometry . . . . . . . . 71.2.2 Nuclear Reactions for Mass Measurements . . . . . . 71.2.3 Storage Ring Methods for Mass Measurements . . . . 81.2.4 Penning Trap Mass Spectrometry . . . . . . . . . . . 101.2.5 Time-of-Flight Mass Measurements . . . . . . . . . . 131.2.6 Comparison of Methods for Nuclear Mass Spectrome-try . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141.3 Production of Radioactive Ion Beams . . . . . . . . . . . . . 141.3.1 The In-Flight Production and Separation Method . . 151.3.2 The Isotope Separation On-Line (ISOL) Method . . . 151.3.3 The Need for Collision Induced Dissociation . . . . . 162 TRIUMF’s Ion Trap for Atomic and Nuclear Science (TI-TAN) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.1 The Radio-Frequency Quadrupole (RFQ) Cooler Buncher . . 19viTable of Contents2.2 The Multiple-Reflection Time-of-Flight Mass-Spectrometer (MR-ToF-MS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.3 The Electron Beam Ion Trap (EBIT) . . . . . . . . . . . . . 222.4 The Measurement Penning Trap . . . . . . . . . . . . . . . . 233 The Multiple-Reflection Time-of-Flight Mass-Spectrometerin Depth and Recent Upgrades . . . . . . . . . . . . . . . . . 253.1 Principles of Time-of-Flight Mass Spectrometry . . . . . . . 253.1.1 Basic Time-of-Flight Mass Spectrometry . . . . . . . 273.1.2 Single Reflection Time-of-Flight Mass Spectrometry . 333.1.3 Multiple-Reflection Time-of-Flight Mass Spectrome-try . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.2 The RFQ Systems in the MR-ToF-MS . . . . . . . . . . . . 353.3 The Analyzer System of the MR-ToF-MS . . . . . . . . . . . 383.4 Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . 393.4.1 Mass Measurements in the MR-ToF-MS . . . . . . . 403.4.2 Mass Selective Re-Trapping in the MR-ToF-MS . . . 423.5 Recent Upgrades to the MR-ToF-MS System . . . . . . . . . 443.5.1 Improved Differential Pumping System . . . . . . . . 453.5.2 100 Hz Repetition Rate Operation Mode . . . . . . . 453.5.3 MagneToF Detector System . . . . . . . . . . . . . . 503.5.4 RF Frequency Modification . . . . . . . . . . . . . . . 503.5.5 Improved Switching Operation . . . . . . . . . . . . . 534 Collision Induced Dissociation (CID) Using the MR-ToF-MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564.1 CID in Analytic Chemistry . . . . . . . . . . . . . . . . . . . 564.2 CID at TITAN . . . . . . . . . . . . . . . . . . . . . . . . . . 574.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 595 Neon Isotope Mass Measurements and Results . . . . . . . 645.1 Neon Isotope Mass Measurement Data . . . . . . . . . . . . 645.1.1 Fitting Procedure to Extract Mass Values . . . . . . 685.1.2 Propagation of Errors . . . . . . . . . . . . . . . . . . 685.1.3 Final Mass Results and Comparison with the AtomicMass Evaluation (AME2016) Literature Values . . . . 715.2 Shortcomings of the Ne Mass Measurements . . . . . . . . . 736 Summary, Conclusions, and Outlook . . . . . . . . . . . . . . 75viiTable of ContentsBibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77viiiList of Tables1.1 A summary of the key fields of research involving nuclear massspectrometry and the necessary relative uncertainties. . . . . 61.2 A comparison of key values for the different methods of massspectroscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.1 A summary of maximum number of turns and subsequentresolving powers for common calibrants at 100 Hz. . . . . . . 505.1 The reported values for the Ne mass measurements with theirrespective calibrants. . . . . . . . . . . . . . . . . . . . . . . . 71ixList of Figures1.1 The comparison of the Semi-Emperical Mass Formula withmeasured results. . . . . . . . . . . . . . . . . . . . . . . . . . 31.2 An example of a shell closure at N = 50 using the two neutronseparation energy. . . . . . . . . . . . . . . . . . . . . . . . . 41.3 S2N near the N = 20 Island of Inversion using values fromthe AME2016. . . . . . . . . . . . . . . . . . . . . . . . . . . 51.4 The routes of the four primary astrophysical processes fornucleosynthesis. . . . . . . . . . . . . . . . . . . . . . . . . . . 61.5 The triple alpha decay chain for 169Pt, 170Au, and 170mAu. . 81.6 The two different modes of operation for mass measurementsin a storage ring. . . . . . . . . . . . . . . . . . . . . . . . . . 91.7 An drawing of a hyperbolic Penning trap geometry with anaxial magnetic field. . . . . . . . . . . . . . . . . . . . . . . . 111.8 An example spectra generated using the TOF-ICR technique. 121.9 A schematic drawing of the process of an FT-ICR mass mea-surement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131.10 The various methods of RIB production, including hybridcombinations of in-flight and ISOL. . . . . . . . . . . . . . . . 161.11 Measured Ne yield compared to total beam current at massesA = 23-27. As Ne yields drop, reduction of molecular con-tamination becomes crucial. . . . . . . . . . . . . . . . . . . . 172.1 The ISAC-I and ISAC-II Experimental Halls with the TITANexperimental set-up circled in red. . . . . . . . . . . . . . . . 192.2 The TITAN experimental set-up with all traps and possibleroutes for ions labelled. . . . . . . . . . . . . . . . . . . . . . 202.3 The DC potential of the TITAN RFQ for accumulation andejection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.4 A plot of x and y stable trajectories in an RFQ without gascooling as a function of a and q. . . . . . . . . . . . . . . . . 232.5 A diagram of the TITAN EBIT in operation. . . . . . . . . . 24xList of Figures3.1 The TITAN MR-ToF-MS with significant features labeled. . . 263.2 A simple schematic of a Time-of-Flight Mass-Spectrometer inoperation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273.3 A diagram depicting the formation of the first order time focus. 293.4 A two-step acceleration scheme for a TOF analyzer. . . . . . 303.5 The visualization of an ion bunch with a velocity distributiondepicted as ions with a spatial and temporal distribution andno velocity distribution. . . . . . . . . . . . . . . . . . . . . . 313.6 A diagram of the transform of a primary time focus to asecondary location after the reflection of an ion bunch. . . . . 343.7 A schematic drawing depicting both open and closed loop sys-tems for Multiple-Reflection Time-of-Flight Mass-Spectrometers. 363.8 The four MR-ToF-MS RFQs highlighted in red. . . . . . . . . 373.9 The three different modes of operation using the MR-ToF-MSswitchyard. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383.10 The TITAN MR-ToF-MS analyzer with key electrodes labeled. 393.11 A schematic sketch of the different stages of an MR-ToF-MScycle and their corresponding trap and/or analyzer potentials. 403.12 The efficiency and resolving power of the TITAN MR-ToF-MS as a function of number of turns. . . . . . . . . . . . . . . 413.13 The separation power and re-trapping efficiency of the TITANMR-ToF-MS as a function of trap depth at different numbersof turns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433.14 A scan of re-trapping time versus count rate used to determinethe efficiency and the resolving powers. . . . . . . . . . . . . . 443.15 The MR-ToF-MS’s vacuum system and its coupling to theTITAN beamline. . . . . . . . . . . . . . . . . . . . . . . . . . 463.16 The new pressures in various sections of the MR-ToF-MS asa function of injected He gas pressure. . . . . . . . . . . . . . 473.17 The normalized transmission of 39K at a various number ofisochronous turns before and after the installation of the newturbo-pump. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483.18 The charge exchange half-lives of 22Ne+ and 40Ar+ comparedto 23Na+. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493.19 The new MagneTOF detector mounted to its base prior toinstallation in the MR-ToF-MS. . . . . . . . . . . . . . . . . . 513.20 A 2D scan of injection steering voltages for both an MCP andMagneTOF detector. . . . . . . . . . . . . . . . . . . . . . . . 523.21 The MR-ToF-MS RF generator box with one modified coilsuch that the inductance is reduced by ≈1/2. . . . . . . . . . 53xiList of Figures3.22 Diagram of the HV switches used at the TITAN MR-ToF-MS. 543.23 A Gaussian fit of the histogram of systematic variation dueto the switching of the second electrostatic mirror after im-provements to the switching were made. . . . . . . . . . . . . 554.1 The three step general schematic for performing a measure-ment with CID. . . . . . . . . . . . . . . . . . . . . . . . . . . 574.2 The electric potential and beam energy superimposed overthe path from the TITAN Cooler-Buncher to the MR-ToF-MS. 584.3 A comparison of spectra using the MR-ToF-MS’s two differ-ent modes of operation. . . . . . . . . . . . . . . . . . . . . . 604.4 A scan of the injection energy for A = 78 beam. . . . . . . . 614.5 The scan of injection energy (with maximum energy trans-ferred) and the resulting rates of different atomic and molec-ular species for A = 76. . . . . . . . . . . . . . . . . . . . . . 624.6 The relationship between the bond energies of various moleculesand their suppression factor. . . . . . . . . . . . . . . . . . . . 635.1 Mass spectra taken with the TITAN MR-ToF-MS at mass =24 u. Re-trapping (lower panel) was employed to suppress24Mg in favor of measuring 24Ne. . . . . . . . . . . . . . . . . 655.2 The mass measurement spectra from MR-ToF-MS for 24−26Newith other species labelled. . . . . . . . . . . . . . . . . . . . 665.3 The resulting mass values from 24Ne mass measurements atdifferent MR-ToF-MS settings (corresponding to differing filenumbers). Additionally, the different fitting methods arecompared. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675.4 A mass spectra taken with MR-ToF-MS with the three dif-ferent fitting functions used in the Ne data analysis. . . . . . 695.5 The difference between AME2016 mass values and the newTITAN measurements. . . . . . . . . . . . . . . . . . . . . . . 725.6 The background contamination of beam from a FEBIAD ionsource for A = 23 - 32. . . . . . . . . . . . . . . . . . . . . . . 735.7 The attempted 27Ne measurement with the re-trapping win-dows highlighted . . . . . . . . . . . . . . . . . . . . . . . . . 74xiiAcknowledgementsI would like to thank all the members of the TITAN collaboration whohave supported and helped me during my Master’s thesis. In particular, Iwould like to thank J. Dilling and A. A. Kwiatkowski for their guidance andsupervision, M. P. Reiter for his instruction and guidance pertaining to theMR-ToF-MS, J. Bergmann for his software support, and T. Murboeck forhis contributions to my thesis writing.xiiiChapter 1Introduction to MassSpectrometryThe nuclear or atomic mass is one of the fundamental properties of such amany-body quantum system. The primary motivation for the determinationof the nuclear mass is to extract fundamental knowledge of the interactionsat work. This leads to better understanding of the nuclear forces and al-lows for the creation of better theories or models which can predict nuclearproperties. For example, mass measurements can help refine the approachesused in nuclear Shell Model calculations [1]. Additionally, nuclear masses areused in calculations for isotopic abundances following astrophysical events[2, 3]. Moreover, it is also possible to use atoms or nuclei as probes forbasic research such as testing fundamental symmetries and determining theunitarity of the Cabibbo-Kobayashi-Maskawa (CKM) matrix [4]. Below isa table which indicates the required precisions of mass measurements to beapplicable to the relevant fields of research. To reach these various levelsof precision, different measurement techniques can be used. These varioustechniques are classified in the experimental category of mass spectrometry.Section 1.2 gives an overview of how different types of mass spectrometryare performed. Lastly, section 1.3 will described the processes of creatingthe beams of exotic nuclei on which experiments are performed.1.1 Motivation for Mass SpectrometryWhen the field of nuclear physics was in its infancy, it was found that massesof isotopes seemed to scale with integer values of the mass of Hydrogen [5].However, evidence eventually emerged that indicated a deficit in the ex-pected mass existed [6]. Only once the neutron was discovered could thisdeviation be explained, and a more complete understanding of the compo-sition of the nucleus was attained [7]. However, the mass of an isotope isnot simply the addition of its constituent parts. Because E = mc2, theenergy of the bonds between the nucleons (through the strong interaction)11.1. Motivation for Mass Spectrometryand the electrons to the nucleus (through the Coulomb interaction) must betaken into account when determining the total mass of an atom. The finalequation for the mass of a particular atom is:m(A,Z) = Zmp + (A− Z)mn − EBn/c2 − EBe/c2 (1.1)where Z is number of protons, A is number of nucleons (protons and neu-trons), EBn is nuclear binding energy, and EBe is electron binding energy.However, EBe is on the order of eV for outer electrons while nuclear massesare on the order GeV/c2. Thus, it will be ignored.With the masses of the proton and neutron known to a high precision(δm/m ≈ 10−11), the remaining challenge is to calculate the binding energyof a particular nucleus by creating a theory with predictive power; in otherwords: predict the total mass of any atom, with any combination of protonsand neutrons. The first attempt at modeling the nuclear mass was the Bethe-Weizsa¨cker Semi-Empirical Mass Formula which is based on the liquid dropmodel [8]. As explained, the key to this model is a prediction of the nuclearbinding energy. To model the binding energy, a spherical volume of thenucleus is considered and various corrections are then added until the finalformula takes the form:EB(A,Z) = avA− asA2/3 − ac Z2A1/3− aa (N −A)24A− δA1/2(1.2)where av is the volume term, as is the surface term, ac is the Coulomb re-pulsion term, aa is the proton-neutron asymmetry term, and δ is the pairingterm. To determine these values, the function is fit using various measuredmasses with different A and Z.While the Semi-Empirical Mass Formula is able to reproduce many largescale trends in nuclear binding energies, there are still missing effects. Thisis most noticeable when Emeasured − Etheory is plotted versus the neutronnumber N for many isotopes as in Figure 1.1. There is a clear emergence ofstructure that the theory fails to predict. This eventually led to the formu-lation of the Shell Model [9]. In the Shell Model, nuclei with a characteristic“magic” number of protons and/or neutrons experience higher stability. Oneway to identify neutron shell closures is by plotting the two neutron sepa-ration energy versus neutron number. The two neutron separation energyis chosen, as opposed to the single neutron separation energy, because itaverages over pairing. As a result, it creates a smooth trend that is easierto analyze The two neutron separation energy is defined as:21.1. Motivation for Mass SpectrometryFigure 1.1: The difference between the measured binding energy of isotopesand the binding energy predicted by the Semi-Empirical Mass Formula.Note the sharp deviations at specific values such as N = 28, 50, and 82.Figure from [10].S2N (N,Z) = EB(N,Z)− EB(N − 2, Z) (1.3)and is the equivalent of the amount of energy necessary to removed twoneutrons from a nucleus. A sharp drop in the separation energy is indicativeof a shell closure such as can be seen in Figure 1.2.For stable isotopes (i.e. non-radioactive) and isotopes near stability (i.e.a long radioactive half-life), the Shell Model produces good predictions thatalign well with experimental data. However, far from stability, this is notalways the case. There are now documented cases in which new neutron shellclosures arise at N = 32 and 34 [12–14], and circumstances when shell effectsfade such as the N = 20 Island of Inversion (a region of the nuclear chartin which single particle energies invert). For a typical shell closure (i.e. thefilling of the canonical shells), there is a steep drop in the S2n after the shellcloses. Thus, strong evidence for the disappearance of a shell closure can befound where the sharp decrease is mitigated. An example of this can be seenin Figure 1.3. The majority of the borders of the N = 20 Island of Inversion31.1. Motivation for Mass SpectrometryFigure 1.2: An example of a shell closure at N = 50. The two neutronseparation energy (S2N ) is plotted as a function of the neutron number. Lo-cations with a sharp decline in S2n indicate a shell closure. (Figure modifiedfrom [11]).41.1. Motivation for Mass SpectrometryFigure 1.3: S2N near the N = 20 Island of Inversion using values from theAME2016 [11].are well defined with Mg as the upper boundary (with Al as a transitionstate) and Ne as the suspected lower boundary. However, there are stillsubstantial uncertainties of the masses of neutron-rich Ne which results inuncertainty in the demarcation of the Island of Inversion. Thus, there is aneed for high precision mass measurements of Ne to better understand thisregion of the nuclear chart.In addition to probing the structure of nuclei, isotopic mass measure-ments also play a crucial role in nuclear astrophysics. In 1957 Burbidge,Burbidge, Fowler, and Hoyle published the seminal review in the field inwhich the main processes of isotopic production were outlined [15]. Theseprocesses are: the slow neutron capture (s-process), the rapid neutron cap-ture (r-process), the rapid proton capture (rp-process), and stellar burning.The various considered paths the processes take can be superimposed on anuclear chart of isotopes (Figure 1.4). Because they remain at or near sta-bility, the s-process and stellar burning are relatively well understood. How-51.1. Motivation for Mass SpectrometryFigure 1.4: The routes of the four primary astrophysical processes for nucle-osynthesis. They are: Stellar Burning (yellow), s-process (red), rp-process(blue), and the r-process (green).ever, both the r-process and rp-process follow paths on the nuclear chartfar from stability in explosive events which are less well understood. Tobetter define and reproduce the observed isotopic abundances, the massesof the isotopes through which the processes travel must be known to a highprecision (δm/m ≈ 10−7). However, not all masses in each process mustbe known to the same precision, and some nuclei hold more weight thanothers. The current understanding of mass dependence in the rp-process ishighlighted in [16] while the r-process sensitivities can be found in [17].Field δm/mNuclear Structure 10−6 - 10−8Astrophysics 10−7Fundamental Physics 10−8 - 10−11Table 1.1: A summary of the key fields of research involving nuclear massspectrometry and the necessary relative uncertainties. Nuclear structureand fundamental physics have a large variation in necessary precision dueto different sub-fields (e.g. nuclear Shell Model and nuclear halos are bothin the field of nuclear structure).61.2. Methods of Mass Spectrometry1.2 Methods of Mass SpectrometryThe following section gives a brief overview of various techniques used inatomic or nuclear mass spectrometry. Historically, many techniques weredeveloped for the measurement of stable isotopes; however, some techniqueshave been modified to perform measurements of radioactive isotopes. Aradioactive isotope is a nucleus that is unstable and, as a result, will decayinto a different species. Because these species decay over time, they are notnaturally occurring. Thus, they must be produced before mass spectrometrycan be performed. The production techniques will be discussed in section1.3.1.2.1 Magnetic Deflection Mass SpectrometryThe first method of mass spectrometry, which was implemented by Aston inhis initial discoveries [5], was magnetic deflection. This utilizes the simplerelation that, in a magnetic field, a charged particle experiences a force of~F = q(~v × ~B). Using this formula in conjunction with the equation forcentripetal force ~Fc =mv2r the mass to charge ratio of a non-relativistic ioncan be determined with:mq=Brv(1.4)where m is mass, q is charge, B is magnetic field strength, r is radius, andv is velocity. Thus, for an ion with a known charge and velocity travelingthrough a well defined magnetic field, the radius of its curvature can bemeasured and the mass can be determined.1.2.2 Nuclear Reactions for Mass MeasurementsIn a nuclear reaction, the mass can be determined using the concept of Q-value, which is the difference between final and initial states (i.e. bindingenergies) or between initial and final kinetic energies such that:Q =∑fB(Nf , Zf )−∑iB(Ni, Zi) =∑iKi −∑fKf . (1.5)Thus, if all kinetic energies and all but one mass are known, the mass ofthe unknown isotope can be determined. From this, the binding energycan be deduced. This is particularly useful when measuring the masses ofisotopes far from stability via α-decay (Figure 1.5) and β+/−-decay chains.71.2. Methods of Mass SpectrometryFigure 1.5: The triple alpha decay chain for 169Pt, 170Au, and 170mAu from[11].An example of utilizing a β+ decay chain is the measurement of 100Sn [18].However, it is absolutely necessary that all excited states populated by thereaction are identified and measured; otherwise, large systematic deviationswill exist between the measured and true values.1.2.3 Storage Ring Methods for Mass MeasurementsFor high-resolution and high-accuracy direct (non-reaction) mass measure-ments, there are several options. To conduct measurements with a highenergy (≈ 10− 1000 MeV/u) beam of radioactive ions, a storage ring is em-ployed. To determine the mass of ions, revolution frequencies in the storagering are measured and converted to mass via:∆ff= − 1γ2t∆(m/q)m/q+∆vv(1− γ2γ2t) (1.6)where f is revolution frequency, γ = (1 − β2)−1/2, β = v/c, m is mass,q is charge, and γt is transition energy. However, for mass measurementsto be possible, the second term in the equation above must go to zero.The two different modes of operation that achieve this are Schottky MassSpectrometry (SMS) and Isochronous Mass Spectrometry (IMS) for whichconceptual sketches are shown in Figure 1.6 [19].To perform SMS, a cool beam (low velocity spread between ions) is81.2. Methods of Mass SpectrometryFigure 1.6: The two different modes of operation for mass measurements ina storage ring from [19].91.2. Methods of Mass Spectrometryrequired. After the hot beam (high velocity spread) is injected into thestorage ring, an electron cooler is used to cool the beam. This method usesmomentum transfer between the hot ions and a collinear electron beam.This leads to the ions eventually reaching an equilibrium. Once this isachieved, the velocity spread of the ions is effectively zero such that ∆vv → 0.The induced current from the stored ions is then measured on two parallelplates. The measured signal is Fourier transformed to extract the revolutionfrequencies. This technique of cooling and then measuring takes on the orderof 1 to 10 s. As a result, typically only long lived (half-life > 1 s) speciescan be measured using SMS.Performing a mass measurement in IMS mode does not involve the pre-viously mentioned cooling process but can use a still hot beam instead. Inthis mode of operation, γt is adjusted by manipulating the ion optics in thestorage ring such that γt → γ and 1 − γ2γ2t → 0. Thus, for a given m/q,the frequency is independent of the velocity (i.e. isochronous) [20]. Thetime-of-flight of the ions is measured for many revolutions, and from that,the revolution time is determined. The measured revolution time can beconverted into frequency, and the mass is determined. Because this methoddoes not require cooling, isotopes with half-lives on the order of microsec-onds can be measured.1.2.4 Penning Trap Mass SpectrometryFor measurements of low-energy RIB, Penning traps provide both the high-est resolving power (R = m/∆m) and lowest relative uncertainty (δm/m).Resolving power refers to the technical capability of the technique to distin-guish between two different masses. Thus, the value ∆m refers to the widthof the mass peak. Relative uncertainty refers to the ratio of the absolutemass uncertainty to the mass and is dependent on many different parame-ters (e.g. peak width, statistics, and systematic uncertainty). Historically,the Penning trap was first developed for high precision measurements of theelectron g-factor [21], but it is now a common instrument in nuclear massspectrometry. As can be seen in Figure 1.7, a Penning trap uses a homoge-neous magnetic field along with an electric field generated by ring and endcap electrodes to generate a geometry and conditions to confine charged par-ticles (e.g. protons, electrons, or ions). Once a charged particle is confined,it rotates with a well described and understood motion inside the trap. Therevolution frequency of the particle is proportional to its mass. Therefore, toperform a mass measurement, the cyclotron frequency (νc) of the particle inthe trap must be measured, and the charge to mass ratio can be calculated101.2. Methods of Mass SpectrometryFigure 1.7: An drawing of a hyperbolic Penning trap geometry with an axialmagnetic field. The ring and end cap electrodes are indicated in red andblue respectively. Figure from [22].with:νc =mqB2pi(1.7)where m is mass, q is charge, and B is magnetic field strength. However,due to difficulties in precisely determining B (δB/B < desired δm/m), itis typically more common to measure the frequency ratios of two species todetermine the mass of the unknown species. The final mass to charge ratiowould then be:mq=νc,calνcmcalqcal(1.8)where the subscript ‘cal’ refers to the calibrant quantities. The cyclotronfrequency can be described by the eigenmotion frequencies added in quadra-ture such that:ν2c = ν2+ + ν2− + ν2z (1.9)where ν+/− are the radial motions and νz is the axial motion. In an idealPenning trap, this relation simplifies to νc = ν+ + ν−. Currently there existseveral different techniques for measuring νc that can be employed dependingon the experimental constraints and desired precision.For the so-called Time-of-Flight ion-cyclotron resonance technique, oneapplies a quadrupole RF excitation to ions in the Penning trap. If this111.2. Methods of Mass SpectrometryFigure 1.8: An example spectra generated using the TOF-ICR techniquefrom [22]. The data is fit using the theoretical lineshape from [23].frequency matches the cyclotron resonance, the ionic magnetron motion isconverted to cyclotron motion. This results in an increase in radial energyEr. The ions are ejected from the trap after an excitation time of Trf , andthe time-of-flight (TOF) to a detector outside the magnetic field is measured.As the ejection ion traverses through the magnetic field gradient, the radialenergy is converted to axial kinetic energy. Hence, if νrf = νc, a decreasein TOF should be detected. To better measure this decrease in TOF, νrf isscanned, and the TOF is determined at each value [22]. The data can thenbe fit using the theoretical lineshape as described in [23] (Figure 1.8).For the Fourier Transform ion-cyclotron resonance method, the ions re-main in the trap and are not released. Instead, this technique directly mea-sures the ion motions in the trap. This is done by a pair of pickup electrodesmeasuring the induced image current of the ion on the electrodes. The in-duced current is amplified, and a Fast Fourier transformation is applied.121.2. Methods of Mass SpectrometryFigure 1.9: A schematic drawing of the process of an FT-ICR mass measure-ment modified from [24]. The ion revolves inside the trap while the imagecharge induces a current on the pickup electrodes. A FFT is then appliedto the signal, and a mass spectrum is generated.This generates a spectra with a peak at the cyclotron resonant frequency.A schematic of the process can be seen in Figure 1.9 [24]. However, becausesingle ions are used, the induced currents are extremely low (≈ fA). As aresult, special precautions and techniques are applied to achieve practicalsignal-to-noise ratios for measurements.1.2.5 Time-of-Flight Mass MeasurementsTo conduct mass spectroscopy of very short lived species found in low energybeams, the Time-of-Flight technique can be used. In general, ions withthe same kinetic energy will have different velocities that are dependent ontheir mass-to-charge ratios. If all species travel over the same distance with aknown energy, the mass-to-charge ratio can be determined by measuring thetime it takes to traverse a well defined distance. Previous examples of thistechnique include the TOFI spectrometer at Los Alamos [25] or the SPEGspectrometer at GANIL [26]. To increase the distance the ions travel whiledecreasing the length of the beamline, they can be reflected by electrostaticmirrors many times using a technique that is known as Multiple-ReflectionTime-of-Flight Mass-Spectrometry. This process can reach a resolving power131.3. Production of Radioactive Ion BeamsMethod Resolving Power Min. Uncert. Meas. Time Ions Needed Range CleaningSMS ≈ 5× 105 ≈ 5× 10−8 ≈ 1− 10 s ≈ 1 Large N/AIMS ≈ 2× 105 ≈ 10−7 ≈ 10− 100 µs ≈ 1 Moderate NoTOF-ICR ≈ 106 to 107 ≈ 10−9 ≈ 0.01− 10 s ≈ 100− 1000 Small YesFT-ICR ≈ 106 to 107 ≈ 10−7 ≈ 1− 100 s ≈ 1− 100 Very Small N/AMR-ToF-MS ≈ 5× 105 ≈ 10−7 ≈ 10 ms ≈ 10− 100 Moderate YesTable 1.2: A comparison of key values for the different methods of massspectroscopy. Many values are interdependent (e.g. increasing measure-ment time increases resolving power) so a range is given for many values.Additionally, these values are given for only radioactive isotope measure-ments, including stable isotope measurements would drastically increase thevariation in values.of ≈ 250,000 in less than 10 ms flight time [27]. An in depth discussion of thetheory behind Time-of-Flight Mass-Spectrometry along with its applicationat TITAN will be given in Chapter 3.1.2.6 Comparison of Methods for Nuclear MassSpectrometryEach modern method of mass spectrometry has different strengths andweaknesses. To help summarize them, the values of key parameters ofeach method are given in Table 1.2. Again, resolving power is definedas R = m/∆m or mass over peak width. Minimum uncertainty refers tothe lowest relative uncertainty one can expect for a measurement with theparticular technique. The minimum measurement time is defined by thetime needed to complete a measurement with acceptable relative uncer-tainty (which is dependent on the scientific motivation of the measurement,see Table 1.1) while the upper limit is defined by the time it takes to reacha minimum uncertainty. The number of ions needed sets the lower limit onthe amount of ion required to reach within an order of magnitude the mini-mum uncertainty. Range refers to how wide a variety of masses that can beaccepted and measured at the same time (i.e. within the same measurementcycle). Lastly, cleaning refers to the method’s ability to remove unwantedcontaminants, if present, so that it does not appear in a final measurement.1.3 Production of Radioactive Ion BeamsTo be able to perform measurements of radioactive species for applicationsin mass spectrometry, the radioactive isotopes must first be produced, ion-141.3. Production of Radioactive Ion Beamsized, and separated from contamination. There are two commonly usedmethods for generating radioactive ion beams (RIB): the so-called in-flightproduction and separation method and isotope separation on-line (ISOL)method. Both of these methods have advantages and disadvantages whichwill be discussed, along with greater detail of the production processes, inthe following sections. A diagram of the path to creation of RIB at in-flightand ISOL facilities is given in Figure 1.10.1.3.1 The In-Flight Production and Separation MethodThe in-flight method [28] utilizes typically high energy heavy ion beams (≈100 MeV/u - 1000 MeV/u) and thin (≈ 1− 10 mm) targets. The beam im-pinges on the target and radioactive isotopes are generated. In this way, thereaction products created in the target receive significant forward momen-tum from the projectile and continue to fly forward. The beam of productisotopes passes through a selection stage, typically called a fragment separa-tor, where magnetic rigidity (Bρ) and energy degradation (∆E) are used toseparate the various product species in the beam [29]. The separation selectsspecies according to m/q, and hence, care must be taken for charge-state se-lection. After this, the beam can be sent to experiments (e.g. storage rings)or can be decelerated in a gas stopping cell and sent to low energy exper-iments (e.g. Penning traps). In addition to providing isotopically selectedbeams, the key strengths of this production method are the insensitivity tochemical effects in the target and the omission of an ion source. Both aredue to the creation of charged nuclei from the initial reaction with essen-tially no chemical reaction, such as charge-exchange, in the target due tothe high energy and very short reaction time [30].1.3.2 The Isotope Separation On-Line (ISOL) MethodThe ISOL method employs light projectiles (e.g. protons or neutrons) onthick targets (where thick refers to the targets ability to stop the reactionproducts). The induced nuclear reactions are fission fragmentation and spal-lation. After production, the products diffuse out of the target and througha transfer line before being ionized using a variety of ion sources. The ionizedspecies (typically singly charged) are extracted using electrostatic potentials.The resulting low energy beam is sent through a dipole magnet with slits toselect a mass unit of interest. The beam is delivered to experiments at lowenergies or undergoes post-acceleration typically to 3-15 MeV/u [28]. TheISOL method allows for the maximization of isotope production in particu-151.3. Production of Radioactive Ion BeamsFigure 1.10: The various methods of RIB production, including hybrid com-binations of in-flight and ISOL [32].lar parts of the nuclear chart by selecting targets and reaction mechanismsthat are well matched to the species of interest. However, the time it takesfor reaction productions to diffuse out of the target (≈ 1-10 ms) restrictsthe accessible nuclei due to half-life. Additionally, target chemistry pre-vents highly refractory elements from diffusing out of the target altogether.Lastly, the need for various ion sources adds an additional degree for diffi-culty in RIB production. The key advantages of the ISOL method are highselectivity and excellent beam quality [31].1.3.3 The Need for Collision Induced DissociationAt both in-flight and ISOL facilities, beams may contain large amounts ofcontamination. At ISOL facilities, the extreme conditions in the target (i.e.high temperatures, voltages, and radiation) lead to the formation of manystable and radioactive molecules. The contamination of the beam becomes amore prevalent problem when using powerful ion sources such as the ForcedElectron Beam Induced Arc Discharge (FEBIAD) plasma ion source, whichis capable of ionizing noble gasses [33]. To better understand the extent ofthis feature, Figure 1.11 shows yields of radioactive Ne isotopes superim-posed on molecular background contamination. Molecular species can alsobe generated at in-flight facilities. Here, they are created in gas stoppingcells when decelerating the high energy beam [34]. Thus, there is need for atechnique to reduce contamination when necessary. The process of CollisionInduced Dissociation (CID) is the approach considered in this thesis andhas previously been characterized at LEBIT at NSCL [34]. Chapter 4 willaddress CID development and characterization at TITAN in greater detail.161.3. Production of Radioactive Ion BeamsFigure 1.11: Measured Ne yield compared to total beam current at masses A= 23-27. As Ne yields drop, reduction of molecular contamination becomescrucial.17Chapter 2TRIUMF’s Ion Trap forAtomic and Nuclear Science(TITAN)The ion trap system TITAN is located at TRIUMF National Laboratoryin Vancouver, Canada. The experimental set-up is installed in the Iso-tope Separator and Accelerator (ISAC) experimental hall where low- andmedium-energy nuclear physics experiments are performed using RIB. TheRIB is created by the ISOL method discussed in section 1.3.2. A schematicoverview of the two ISAC halls can be found in Figure 2.1 [35]. To createthe radioactive isotopes, a proton beam from TRIUMF’s main cyclotron,with energies up to 500 MeV and currents up to 100 µA, impinges on athick target (usually a stack of carbide or metallic foils). The generatedatomic species diffuse out of the target to an ion source. Because ionizationis element specific, the ion source is selected for every specific experimentbased on the desired beam. Examples of commonly used ion sources are:hot surface ionization, FEBIAD [33], or resonant ionization laser ion source[36]. The resulting beam has an initial energy of a few tens of keV andcan be sent to experimental systems such as TITAN. Otherwise, it can befurther accelerated for medium and high energy experiments using a seriesof linear accelerators [37].This chapter will serve as an introduction to the traps found at TI-TAN [38]. The four traps that are currently part of the TITAN experi-ment are, in order of appearance, the Radio-Frequency Quadrupole (RFQ)Cooler-Buncher, the Multiple-Reflection Time-of-Flight Mass-Spectrometer(MR-ToF-MS), the Electron Beam Ion Trap (EBIT), and the MeasurementPenning Trap (MPET). These traps can be used in differing combinationsdepending on the required specifications such as: desired precision, the pos-sible measurement time (as dictated by the species of interest’s half-life),and the amount of contamination (i.e. the amount of unwanted species inthe beam compared to the desired species). An overview of the experimental182.1. The Radio-Frequency Quadrupole (RFQ) Cooler Buncherset-up can be seen in Figure 2.2.Figure 2.1: The ISAC-I and ISAC-II Experimental Halls with the TITANexperimental set-up circled in red.2.1 The Radio-Frequency Quadrupole (RFQ)Cooler BuncherThe first trap at TITAN is the RFQ Cooler-Buncher. The two primaryfunctions of this linear Paul Trap are cooling and bunching of the incomingDC ion beam from ISAC. This step is necessary so that the beam can betransmitted to the subsequent ion traps as a bunch with a small energyspread and at high efficiency. Additionally, the DC beam is delivered to theRFQ at an energy of 20 to 60 keV, but the required beam energies for thevarious subsequent traps range from 1.3 keV to 5 keV. In order to adjst tothe desired energy, the energy of the extracted bunched beam is reducedusing a pulsed drift tube after ejection from the RFQ.192.1. The Radio-Frequency Quadrupole (RFQ) Cooler BuncherFigure 2.2: The TITAN experiment with all traps and possible routes forions labelled [38, 39].The RFQ consists of four rods with 24 segments each. The segmentsare used to create a smooth electric potential to allow for the accumulationof ion bunches on the ejection end of the RFQ. Additionally, the electrodesegments at the end of the RFQ can be switched to eject the bunched beam.A diagram of the electrodes and typically applied potentials can be seen inFigure 2.3 [40]. To cool the ions as they are accumulated in the RFQ, aHe buffer gas is used. This provides collisional cooling. The RF and DCpotentials are applied to the rod segments with the circuit seen in Figure2.3 to generate the desired radial confinement and longitudinal drift.The radial potential of the system is defined as:202.1. The Radio-Frequency Quadrupole (RFQ) Cooler BuncherΦ(~r, t) =UDC + VRF cos(ωRF t)4(x2 − y2)2R20(2.1)where UDC is the DC potential, VRF is the RF potential, ωRF is the RFfrequency, R0 is the distance from the center of the trap to the rods, and xand y are the Cartesian coordinates in the trap. The electric field can bedefined as:~E =(−xy)2(UDC + VRF cos(ωRF t))R20. (2.2)From this, the equations of motion can be written as:x¨+emR20(UDC + VRF cos(ωRF t)x = 0 (2.3)y¨ − emR20(UDC + VRF cos(ωRF t)y = 0 (2.4)z¨ = 0. (2.5)Additionally, the so-called Mathieu parameters can be introduced as:au = ax = −ay = 4eUDCmωRFR20(2.6)qu = qx = −qy = 2eVRFmωRFR20(2.7)η =ωRF t2(2.8)and allows for equations 2.3 and 2.4 to be written as:d2udη2+ [au − 2qu cos(η − η0)]u = 0 (2.9)where η0 is the initial phase and u is x or y (as shown in equations 2.6 and2.7) [41]. This can be used to plot the values of stable x and y trajectoriesas a function of a and q as can be seen in Figure 2.4 [42]. The values forwhich x and y stability overlap are those which allow for ion transport.Additionally, it can be seen that, for a given value of a and q, there is alow mass cut-off. This is highlighted by a zoomed in portion of the plot fora and q < 1. Thus, the UDC and VRF values can be chosen for a specificmass and frequency such that it acts as a mass filter. The TITAN RFQ is212.2. The Multiple-Reflection Time-of-Flight Mass-Spectrometer (MR-ToF-MS)operates with predetermined values for UDC ; therefore, VRF and ωRF mustbe carefully chosen to allow for the transport of the desired species.Figure 2.3: The DC potential of the TITAN RFQ for accumulation (black)and ejection (red) taken from [40].2.2 The Multiple-Reflection Time-of-FlightMass-Spectrometer (MR-ToF-MS)The MR-ToF-MS is the newest trap to be added to the TITAN experiment.It was designed and built at the University of Gießen [43, 44], and is basedon the MR-ToF-MS at GSI Darmstadt [27]. The trap was commissionedonline in 2017, and the accuracy was verified using several joint RIB massmeasurements with the MPET [13, 14]. Chapter 3 will provide a moredetailed description of the TITAN MR-ToF-MS.2.3 The Electron Beam Ion Trap (EBIT)The uncertainty of a measurement of an ion’s cyclotron frequency in a Pen-ning trap is inversely proportional to the charge q of the measured ions(Equations 1.7 and 1.8). Therefore, for mass measurements which requirea very high degree of precision, an increase in charge-state q will lead to222.4. The Measurement Penning TrapFigure 2.4: A plot and x and y stable trajectories in an RFQ without gascooling as a function of a and q. Additional detail is given to the stabletransport region near q = 1 to highlight the light mass cut-off. Figure takenfrom [42].improved precision. The increase is accomplished using a charge breeder intandem with a Penning trap mass measurements [45, 46]. Additionally, theEBIT can be used for other techniques such as in-trap decay spectroscopyand in-trap decay and recapture, which allows access to progenitor species[47].In the EBIT, to trap the ions, a series of electrodes is used to createa trapping potential in the longitudinal direction while a superconductingmagnet creates a radially confining field. To create ions with higher chargestates, a high energy (10s of keV) high current (100 mA) electron beambombards the trapped ions. When the electron beam travels into the mag-netic field, it is compressed due to the magnetic field before interacting withthe ion bunch. The electron beam passes out the opposite side where it isstopped on the collector [48]. A schematic diagram of the set-up can befound in Figure 2.5.2.4 The Measurement Penning TrapThe principles of Penning trap operation have been briefly discussed in sec-tion 1.2.4. The Measurement PEnning Trap (MPET) at TITAN is a hy-perbolic Penning trap that uses a 3.7 T magnet. The MPET employs the232.4. The Measurement Penning TrapFigure 2.5: A diagram of the TITAN EBIT in operation. Indicated are theelectron-gun (left), the trap electrodes and superconducting coils (center),and electron collector (right). The lower half of the figure depicts the axialconfinement (from the trap electrodes) and the radial confinement (fromthe magnetic field) of the ions as they are bombarded by the electron beam.Figure taken from [48].ToF-ICR technique and has performed measurements at relative uncertain-ties on the order of δm/m ≈ 10−9 [22]. A schematic diagram of a hyperbolicPenning trap is presented in Figure 1.7 and an example of a ToF-ICR spec-trum in Figure 1.8.24Chapter 3The Multiple-ReflectionTime-of-FlightMass-Spectrometer in Depthand Recent UpgradesAs mentioned in the previous chapter, the Multiple-Reflection Time-of-Flight Mass-Spectrometer (MR-ToF-MS) is the newest addition to the TI-TAN experimental set-up [39]. In this chapter, the foundational principlesof Time-of-Flight Mass Spectrometry will be discussed, and a more detaileddescription of the individual components of the MR-ToF-MS will be given.Additionally, an introduction to the different modes of operation will bepresented. Lastly, recent upgrades to the device will be introduced, and thesystematic improvements will be presented. A general schematic overviewof the TITAN MR-ToF-MS with key features highlighted can be found inFigure 3.1.3.1 Principles of Time-of-Flight MassSpectrometryTime-of-Flight Mass Spectrometry (ToF-MS) exploits the fact that ions withthe same energy but different masses will have different velocities [49]. Ifthose ions accelerate to energy K over distance Lacc and travel over a knowndistance Ldrift, they will spread out in time due to their velocity difference.This can be seen in Figure 3.2. However, because it is difficult to determinethe exact energy of a ion, it is best to introduce a pair of species, carryout a relative measurement, and compare their respective time-of-flights. Ifthe mass of one species is known, the mass of the unknown species can bederived.253.1. Principles of Time-of-Flight Mass SpectrometryFigure 3.1: The TITAN MR-ToF-MS with significant features labeled. Fig-ure modified from [32].263.1. Principles of Time-of-Flight Mass SpectrometryFigure 3.2: A simple schematic of a Time-of-Flight Mass-Spectrometer inoperation. Shown on the left are three different species in one ion bunch.After acceleration, the species separate into different bunches in the centraldrift region before impinging on a detector on the right.3.1.1 Basic Time-of-Flight Mass SpectrometryIn the simplest case of ToF-MS, ions, with charge q, are accelerated by astatic electric potential difference U to a final energy K that is mass indepen-dent. The ions travel a distance at a constant energy. Using a set distanceLdrift and classical mechanics, the drift time tdrift can be derived:K = qU =12mv2 (3.1)→ v =√2qUm(3.2)Ldrift = v tdrift (3.3)→ tdrift = Ldrift√m2qU. (3.4)As expected, the TOF is dependent on the mass-to-charge ratio (tdrift ∝√m/q). Thus, ions with a smaller mass-to-charge ratio (in general, lighterions) have shorter drift times. However, not only the tdrift needs to be ac-counted for but the acceleration time tacc as well. If the acceleration isconstant over the distance Lacc, tacc can be found using classical electrostat-ics and non-relativistic mechanics:273.1. Principles of Time-of-Flight Mass SpectrometryF = ma = qU→ a = qUm(3.5)Lacc =12at2acc → tacc =√2Lacca(3.6)tacc =√2mLaccqU. (3.7)The ion’s kinetic energy can be defined as K = qULacc such that tdrift andtacc are defined in terms of K. The total time of flight then comes to:ttotal = tacc + tdrift =√2mKqU+ Ldrift√m2qK. (3.8)However, this equation is based on ideal conditions and assumes: no energyspread, no spatial distribution of ions, perfect linear acceleration, and noaberrations in the drift region. In reality, all of these issues play a non-negligable role in ToF-MS and will be discussed in the following sections.Spatial Distribution Effect on ToF-MSAs described above, the final kinetic energy of an ion is dependent on Lacc.A spatial distribution of ions before the initial acceleration will then resultin a distribution of kinetic energies. If we consider ∆` to be the deviationfrom the reference starting position (L0acc) and K0 to be the reference kineticenergy and using the above equations and substituting L0acc −∆` for Lacc,we get:K = qU(L0acc −∆`) = (1−∆`L0acc)K0 = (1 + δ)K0 (3.9)where −δ = ∆`/L0acc. Additionally, it can be seen from Equation 3.9, thatδ = K/K0−1. This is the relative deviation for the kinetic energy ratio. Mostcommonly, it is treated as a parameter when assessing the aberrations in aTime-of-Flight spectra. Using the parameterization for K, Equation 3.8 cannow be written as:ttotal = Ldrift√m2K0(1 + δ)+√2mK0(1 + δ)qU. (3.10)Factoring out√2m/K0 gives:283.1. Principles of Time-of-Flight Mass SpectrometryFigure 3.3: A diagram depicting the formation of the first order time focus.Because the ions which start further from the detector have a higher energy,they catch up to the leading ions with a lower energy. Figure modified from[32].ttotal =√2mK0(Ldrift2√(1 + δ)+ L0acc√(1 + δ)). (3.11)If a Taylor expansion is done, the first order dependence on δ disappearsfor T(δ) = ttotal - t0total when Ldrift = 2L0acc. This is referred to as the primarytime focus. Practically speaking, this occurs because ions that need to travelfurther have a higher energy, while ions that are initially closer to the timefocus have lower energy. Thus, the time focus is where the faster ions catchup to the ions that were initially leading. This is depicted in Figure 3.3.When using the first order approximation, the time-of-flight of the primarytime focus is:ttotal = 2L0acc√2mK0. (3.12)However, time-of-flight aberrations still exist such as the higher orderterms of the previously mentioned Taylor expansion of Equation 3.11. Tocounteract this, a two-step acceleration of the ion bunch can be done. Thisgives another degree of freedom and creates a primary time focus at:293.1. Principles of Time-of-Flight Mass SpectrometryFigure 3.4: A two-step acceleration scheme for a TOF analyzer. With fixedvalues of La1 and La2, the values for K can be adjusted such that the secondorder TOF aberration is eliminated. Figure modified from [32].L = 2La1(K0Ka1)3/2− 2La2√K0Ka111 +√Ka1/K0(3.13)where K0 = Ka1+Ka2. By carefully selecting a set geometry with fixedspatial positions for La1 and La2, the ratio Ka1/Ka2 can be adjust to shift thefirst order time focus. Additionally, the ratio can be adjusted to eliminatethe second order time-of-flight aberration [49]. As a result, the position ofthe time focus becomes mass independent, but the time-of-flight to the timefocus remains mass dependent. A schematic of two-step acceleration can befound in Figure 3.4Velocity Distribution Effect on ToF-MSIn addition to an intrinsic spatial distribution of accelerated ions, there isalso a velocity distribution at the moment the ions are accelerated. This isdue to the thermal energy distribution of the ions. In order to reach hightransport efficiency, the motion perpendicular to the direction of acceler-ation requires ion optical focusing elements to correct the deviations andcreates spatial aberrations. Motion parallel to the direction of acceleration303.1. Principles of Time-of-Flight Mass SpectrometryFigure 3.5: The visualization of an ion bunch with a velocity distributiondepicted as ions with a spatial and temporal distribution and no velocitydistribution. The red arrow depicts the ‘turn-around-time’ of an ion with aninitial velocity anti-parallel to the direction of acceleration. Figure modifiedfrom [32].creates an additional forward component in the ion’s time-of-flight. Motionanti-parallel to the direction of acceleration creates a reduction in the for-ward component in the ion’s time-of-flight. Over the ensemble, this wouldtranslate to an energy distribution proportional to the original spread. Tomodel the velocity distribution in a simple manner, the ion’s starting posi-tion and time can be adjusted to re-create the different velocities. For ionswith a positive velocity (i.e. travelling in the direction of the acceleration),their starting time and position are moved back so that t0 → t0 −∆t andz0 → z0 −∆z. For ions with a negative velocity (i.e. ions moving in the op-posite direction of acceleration), their starting time is increased while theirstarting position is decreased so that t0 → t0 + ∆t and z0 → z0−∆z. Theseshifts can be seen in Figure 3.5In quantitative terms, the change in position, ∆z, is dependent on ve-locity squared, mass, charge of the ion, and the strength of the acceleratingfield U such that:∆z = −mv2z02qU. (3.14)The change in time has similar dependencies, except ∆t ∝ v, with the313.1. Principles of Time-of-Flight Mass Spectrometryequation being:∆t = −mvz0qU. (3.15)∆t is referred to as the ‘turn-around time’ and defines the minimum peakwidth for an ion bunch as a function of the velocity. This so-called turn-around-time can be seen physically in Figure 3.5 by tracing the amount oftime needed to invert the initial velocity of an accelerated ion. This is inturn dependent on the temperature of the ion bunch and can be used todefine the maximum resolving power of the system [49]. To determine theresolving power, it is assumed that the population of ions is thermalized totemperature T, and the velocities follow a Gaussian distribution centered atzero. To observe such a distribution, approximately fifty ions of the samespecies need to be observed. For that distribution,σv =√kBTm(3.16)and FWHM = 2√2ln(2)σ. Using these two equations in combination withEquation 3.15 yields:∆tFWHM =2√2ln(2)mkBTqU. (3.17)Because it is not possible to correct for ion velocities when accelerating, thisvalue determines the peak’s time width in an otherwise ideal TOF analyzer.Resolving power can be defined in terms of either mass-to-charge ratioor TOF such that:Rm =m/q∆m/q=Rt2=ttof2∆t. (3.18)Using Equation 3.18, the equation for the primary time focus using singlestage acceleration (i.e. acceleration done with a single potential U as seenin Equation 3.12), and the equation for turn around time spread (Equation3.17) a simple approximation for resolving power can be calculated as:Rm =√K04ln(2)kBT(3.19)where K0 is the reference kinetic energy. Thus, to increase resolving power,higher extraction fields and colder ion bunches need to be used. However,this technique has its limits. To reach a resolving power of 1000 for an ion323.1. Principles of Time-of-Flight Mass Spectrometrybunch at room temperature, the ion energy would need to be K0 ≈ 65 keV.This potential strength would increase the aberrations in a real TOF massspectrometer, negatively effecting the resolving power.3.1.2 Single Reflection Time-of-Flight Mass SpectrometryTo reach resolving powers on the order of 1000, the flight path of the ionsneeds to be extended. However, there needs to be a way for the path toremain isochronous (i.e. the TOF is energy independent, comparable to anIMS Storage Ring discussed in section 1.2.3). To accomplish this, an electricfield can be used as a ‘mirror’ to reverse the direction of the ions. This istypically realized using a hyperbolic grid configuration. As a result, ionswith higher kinetic energies will travel further into the hyperbolic field andtake more time to turn around. However, because the ions have a higherenergy, they will eventually catch up to the slower ions. As a consequence,a new time focus is formed. To find the new time focus, the time-of-flightfrom the primary time focus can first be calculated as:tmir =√mK0(L1 + L2√1 + δ+ 4L0mir√1 + δ). (3.20)Using a first order approximation (similar to the one performed on Equation3.11), the reflected time focus is found to be at:L2 = 4Lmir − L1. (3.21)Figure 3.6 shows this concept schematically. Thus, by varying the potentialof the reflecting field, the location of the new time focus can be adjusted.Additionally, a two step reflection potential (i.e. two different potentialsapplied to the mirror) can be used to eliminate the second order TOF aber-ration (analogous to the two step extraction acceleration discussed above)[49].3.1.3 Multiple-Reflection Time-of-Flight MassSpectrometryWhile the use of a single reflection can improve the resolving power by afactor of≈ 2, significantly higher resolving powers are required for mass spec-trometry in nuclear physics (see section 1.1). One effective way to increaseresolving power is by increasing the total time-of-flight of the ion bunch. Toachieve a resolving power of 100,000 with an ion bunch with width 10 ns333.1. Principles of Time-of-Flight Mass SpectrometryFigure 3.6: A diagram of the transform of a primary time focus to a sec-ondary location after the reflection of an ion bunch [32].and irrespective of other aberrations, Equation 3.18 can be re-arranged tofind:tTOF = 2∆t · Rm = 2 ms. (3.22)Instead of creating a beamline tens to hundreds of meters long, it is far moreefficient to fold the flight path on top of itself using a pair of electrostaticmirrors. Additionally, this technique reduces the through-put and requiresbunched operations. This allows for the flight path to be extended whilemaintaining a compact system. To calculate the resolving power based onthe number of loops done between the mirrors (referred to as ‘turns’), thetime-of-flight is modified to: tTOF = t0 + Ntturn where N is number of turnsand tturn is the time one turn takes. Additionally, the peak width is modifiedto: ∆t2 = ∆t20 + N2∆t2turn where ∆to is the peak width due to turn aroundtime and injection aberrations, and ∆tturn is the peak widening caused byaberrations during turns. The mass resolving power is then:Rm =t0 + Ntturn2√∆t20 + N2∆t2turn. (3.23)If the ions were to fly for an infinite amount of time, the idealized maximum343.2. The RFQ Systems in the MR-ToF-MSresolving power can be calculated as:limN→∞Rm =tturn2∆tturn. (3.24)However, aberrations and voltage instabilities prevent this value from beingreached in real systems. Additionally, there can exist operational constraintsdependent on the device’s design which limit the number of turns possible.To conduct Multiple-Reflection Time-of-Flight Mass-Spectrometry, thereare two ways to construct the device [44]. 1.) In an open-path system, ionsare injected and undergo a fixed number of turns before exiting the otherside of the analyzer. 2.) In a closed system, the ions are injected andundergo a variable number of turns before one of the mirrors is opened, andthe ions exit the analyzer. Switching the mirrors in a closed system adds anadditional degree of complexity, but the gained flexibility can compensatefor this. A diagram comparing the two options can be found in Figure 3.7.3.2 The RFQ Systems in the MR-ToF-MSFigure 3.8 highlights the four different RFQs that are needed for the fulloperation of the MR-ToF-MS. The four RFQs needed are: the input RFQ,the transport RFQ, the RFQ switchyard, and the output RFQ. The fourrods used in the RFQs are manufactured with resistive material (carbondoped Polyphenylene Sulfide) to create a smooth DC potential. Potentialminima are needed for trapping, and apertures are placed on both ends ofthe RFQs to provide the necessary potentials. The concept behind stable iontransport has already been discussed in section 2.1, and the same principlesapply here.To generate a symmetric RF for the MR-ToF-MS RFQs, an LC circuitdriven with a square wave signal. The inductance of the circuit is definedby the toroidal coils used in the system while the capacitance is determinedby the coaxial cables, the vacuum feed-throughs, and the electrodes. Thus,to change the resonance frequency of the circuit, either the inductance ofthe torus can be changed, and/or the capacitance of the coaxial cables canbe adjusted. This is discussed in further detail in section 3.5.4.Like the TITAN RFQ, the RFQs in the MR-ToF-MS are filled witha He buffer gas. This is necessary for cooling of injected ions, cooling ofthe re-trapped ions (see section 3.4.2), and ultimately leads to the efficienttransport of ions. The pressures at which the RFQs operate is 1− 5× 10−2mbar while the desired pressure in the analyzer is 1 × 10−7 mbar. Thus,353.2. The RFQ Systems in the MR-ToF-MSFigure 3.7: A schematic depicting both open and closed loop systems forMultiple-Reflection Time-of-Flight Mass-Spectrometers. Figure modifiedfrom [32].363.2. The RFQ Systems in the MR-ToF-MSFigure 3.8: The four MR-ToF-MS RFQs highlighted in red. (modified from[32].373.3. The Analyzer System of the MR-ToF-MSFigure 3.9: The three different modes of operation using the MR-ToF-MSswitchyard along with the three different potentials (Low, Mid, and High)needed to operate. Figure from [32].there is the need for differential pumping. This aspect of MR-ToF-MS hasundergone a recent upgrade and is addressed in section 3.5.1.In addition to the traditional RFQs (input, transport, and output), anovel six port RFQ switchyard was designed and built based on a previousbent RFQ system [50, 51]. The pseudopotential (potential produced by theRF voltages) provides radial confinement while different potentials appliedon the ends of the rods allow for pass-through, bending, and merging modesof operation. To operate in the different modes, three different potentialsare needed as is highlighted in Figure 3.9 [32]. This RFQ switchyard allowsfor the merging of outside beams with beam from an internal ion source.Additionally, the RFQ switchyard allows for isobarically cleaned beam frommass-selective re-trapping (see section 3.4.2) to be sent from the MR-ToF-MS to the rest of TITAN.3.3 The Analyzer System of the MR-ToF-MSThe main function of the analyzer is to allow long Time-of-Flight mea-surements to be performed. The system is show in Figure 3.10 with keycomponents highlighted. The analyzer of the MR-ToF-MS consists of ninemain electrodes (four mirror electrodes on each side and a grounded drifttube in the middle). The inner pair of electrodes (E4 and E6) serve as focus-ing lenses while the outer three pairs of electrodes (E1-3 and E7-9) create areflective potential. The operational voltages of each electrode were deter-mined through simulation such that first- and second-order aberrations totime-of-flight were eliminated and third-order aberrations were minimized383.4. Modes of OperationFigure 3.10: The TITAN MR-ToF-MS analyzer with key electrodes labeled(from [32]).(such as those discussed in section 3.1).In addition to the drift tube in the center of the analyzer, there are twopairs of deflecting electrodes. These four electrodes are called the mass rangeselectors (MRS). By applying a deflection voltage when the ion of interest isnot passing through the drift tube and switching off when the ion of interestenters the drift tube, the injected ion bunch can be cleaned of non-isobariccontaminants. This MRS cleaning is normally performed for the first 30-50turns of a measurement cycle to provide sufficient cleaning. However, onlya single pair of electrodes is needed to provide adequate deflection steering.This allows for the other pair of electrodes to be re-purposed if the needarises.When injecting beam into the analyzer, the ion bunch first passes througha series of lenses and steerers to ensure the beam is correctly aligned. Tocheck beam alignment, the voltages of the steering electrodes can be scannedwhile monitoring the rate of transmission and the peak properties. Opti-mum settings are for maximum transmission and minimum peak width,which typically leads to a symmetric peak. An example of employing asteering electrode scan can be found in section 3.5.3.3.4 Modes of OperationThe standard operation of the system is done at 50 Hz. For this, an ionbunch is sent to the MR-ToF-MS input RFQ from the TITAN RFQ every20 ms. For specific cases, a doubled frequency can be used, and a 100 Hzmode of operation has been developed for the Ne beamtime. Typically, theion bunch is transported from the input RFQ through the switchyard andtransfer RFQ to the pre-trap. In the pre-trap, the ions undergo a cooling393.4. Modes of OperationFigure 3.11: A schematic sketch of the different stages of an MR-ToF-MScycle and their corresponding trap and/or analyzer potentials. 1.) Theinjection trap kicks the ions into the analyzer. 2.) The entrance mirrorcloses and the ions reflect back and forth separating by mass. 3a.) Theentrance mirror is opened and the injection trap is closed to re-capture theion of interest. 3b.) The exit mirror is opened, and the ions impinge on adetector creating a time-of-flight spectrum.stage before being moved to the main trap. The ions are then injected intothe analyzer where a mass measurement or mass selective re-trapping cyclewill be completed. An overview of the various MR-ToF-MS cycles can befound in Figure 3.11.3.4.1 Mass Measurements in the MR-ToF-MSWhen conducting a mass measurement, the ions undergo two step accelera-tion from the injection trap and enter the analyzer through the first electro-static mirror. The potential of the first electrostatic mirror is subsequentlyraised to trap the ions in the analyzer. The ions then undergo a sufficient403.4. Modes of OperationFigure 3.12: The efficiency and resolving power of the TITAN MR-ToF-MSas a function of number of turns.number of turns such that they are separated by mass into distinct bunchesand reach a maximum resolving power (≈ 250,000). Once enough turns havebeen completed, the potentials of the second mirror are lowered to allow theions to exit the analyzer and impinge on the detector. Single ion counting isperformed to create a time-of-flight spectrum which, with a calibrant mass,can be converted to a mass spectrum. As a general rule, heavier speciesrequire longer flight times to reach a maximum resolving power. However,as the flight time increases, the efficiency of the system drops due to im-perfections and residual gas interactions. A plot of efficiency and resolvingpower as a function of the number of turns can be seen in Figure 3.12. Ata high number of turns (> 300), the resolving power saturates at ≈ 250,000and the efficiency goes to ≈ 70%.413.4. Modes of Operation3.4.2 Mass Selective Re-Trapping in the MR-ToF-MSTo perform mass selective re-trapping, the injection, capture, and time-of-flight separation steps are the same. However, the necessary resolving powerfor re-trapping is normally significantly lower than mass measurements. Asa result, it is common for re-trapping cycles to only use 30 to 50 turns beforelowering the potentials of the first electrostatic mirror. The trap potentialsare then closed at a pre-determined time so that the desired isotope is turn-ing around in the trap at the time of switching. The re-trapped ions are thenre-cooled in the buffer gas. They can then be re-injected into the analyzer inmass measurement mode. Alternatively, the ion bunch can be transportedout of the MR-ToF-MS, via the switchyard and output RFQ, and sent tothe rest of TITAN as a purified beam with unwanted isobaric contaminantsremoved or strongly suppressed.When considering the efficiency of re-trapping, there is a dependenceon both the number of turns the ions undergo in the analyzer, and thepotential depth of the injection trap. However, the achievable resolvingpower is also dependent on the trapping potential. This creates a depen-dence where higher potentials result in higher efficiencies but lower resolvingpower. Thus, a balance between efficiency and separation power must befound to maximize the impact of re-trapping. This need for balance can beseen in Figure 3.13, and optimization is done on a case by case basis.To determine the re-trapping parameters, the time at which the trapis closed can be scanned while measuring the count rate of the species ofinterest. Plotting the measured values results in a curve with steep sides anda flat, plateaued top. The efficiency can be calculated using the maximumtransmission during re-trapping compared to the count rate without re-trapping. To calculate the re-trapping resolving power, both the FWHMand rise/fall time of the curve is measured. With the re-trapping time-of-flight known (i.e. the time at which the trap closes) the resolving power, R,is calculated using:R =t2 ·∆t . (3.25)As can be seen in Figure 3.14, for 30 turns during the re-trapping cycle,the FWHM resolving power is 8,000 while the edge resolving power is 20,000.The FWHM resolving power can be thought of as how wide the re-trappingwindow is, while the edge resolving power describes how sharp the boardersof the re-trapping window are. Thus, increasing the FWHM resolving powerreduces the size of the window, and increasing the edge resolving power423.4. Modes of OperationFigure 3.13: The separation power and re-trapping efficiency of the TITANMR-ToF-MS as a function of trap depth at different numbers of turns.433.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.14: A scan of re-trapping time versus count rate used to determinethe efficiency and the resolving powers. The window is characterized bysharp edges quantified through ’edge’ resolving power and a plateaued topwith a width defined with FWHM resolving power.creates a sharper drop in transmission after leaving the window. As canbe seen in Figure 3.13, the best way to increase the resolving power ofre-trapping is to increase the number of turns. This allows for creatingincreasingly narrow windows which in turn produces cleaner beams.3.5 Recent Upgrades to the MR-ToF-MS SystemThe MR-ToF-MS’s key feature is the ability to perform fast, high-precisionmeasurements with very high sensitivity. The main aspects that contributeto this are high efficiency, high flexibility, and high accuracy. The MR-ToF-MS has undergone a series of upgrades to further improve these features.443.5. Recent Upgrades to the MR-ToF-MS System3.5.1 Improved Differential Pumping SystemAs mentioned in section 3.2, the RFQs of MR-ToF-MS are filled with Hebuffer gas for the cooling of ion bunches before injecting into the analyzersection. However, a low vacuum (< 10−7 mbar) is desired in the analyzersection such that the mean free path of the ions (i.e. the distance the ionscan travel before scattering) is longer than the total path length. Thisprevents ion-He scattering which results in losses in efficiency. In order toachieve this, strong differential pumping is necessary to separate the twodifferent regions and provide the required pressure. The schematic of thedesired vacuum is found in Figure 3.15 [32]. This differential pumping isrealized by employing an additional turbo-pump (TwisTorr 704 FS) whichwas installed at the boundary of the RF Trap and analyzer. The locationof the new turbo-pump is highlighted in Figure 3.15.The overall transmission efficiency depends on the He gas pressure intwo different ways. First, gas pressure must remain high enough such thatinjected beam from the Cooler-Buncher can be captured and cooled effi-ciently. Second, the pressure must not be too high and, as a result, causethe pressure in the analyzer to be too high. Thus, there is a balancing pointof maximum efficiency. By adding an additional turbo-pump, the balancingpoint has shifted such that the overall efficiency has increased. This is be-cause more He gas can be injected into the RFQs, allowing more injectedbeam to be captured, while keeping the pressure in the analyzer low. Thisdescription of the systematic shift in pressures can be seen in Figure 3.16.As a result of the shift in optimum pressure, there is also a shift in thetransmission efficiency in the analyzer. To measure this, the count rate isdetermined at various turns and subsequently normalized to one turn. Theresult is plotted in Figure 3.17 where, at a high number of turns, there is a50% improvement in transmission.3.5.2 100 Hz Repetition Rate Operation ModeFor some mass measurements, it is crucial to reduce the amount of timenecessary to complete the measurement. For example, when the radioactivehalf-life is below 20 ms (the time it takes to complete one measurementcycle). Another reason is the charge exchange half-life of elements with highionization potentials (e.g. the noble gases). An example of the loss of 22Neand 40Ar compared to 23Na attributed to charge exchange with contaminantsin the He buffer gas can be seen in Figure 3.18. To increase the transportefficiency of these species, the number of interactions with contaminants in453.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.15: The MR-ToF-MS’s vacuum system and its coupling to theTITAN beamline. The newly installed turbo-pump is highlighted in red.Figure modified from [32].463.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.16: The new pressures in various sections of the MR-ToF-MS as afunction of injected He gas pressure.473.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.17: The normalized transmission of 39K at a various number ofisochronous turns before and after the installation of the new turbo-pump.483.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.18: The charge exchange half-lives of 22Ne+ and 40Ar+ comparedto 23Na+. Due to the noble gases’ short charge exchange half-lives, it is idealfor the ions to spend as little time in the He buffer gas as possible.the buffer gas must be reduced. This is realized by reducing the cycle timeof the MR-ToF-MS or increasing the purity of the He buffer gas. In a firstattempt the MR-ToF-MS is switched to a 100 Hz repetition rate and hencereduce the interaction time in the system. The change involves adjustingthe timings of the trigger system and carrying out a re-tune of the analyzervoltages due to the change in duty cycle.Once the analyzer has been re-tuned, there is no difference in efficiencyversus number of turns compared to the 50 Hz repetition rate. However, itresults is a loss in maximum resolving power for heavier isotopes (A > 39).In the new mode of operation, the maximum flight time is 4.9 ms. Thisreduces the maximum number of turns that can be done in a single cyclewhich effects the resolving power. This effect is highlighted in Table 3.1.493.5. Recent Upgrades to the MR-ToF-MS SystemIsotope Maximum Number of Turns Resolving Power39K 390 200,00085Rb 270 130,000133Cs 200 120,000Table 3.1: A summary of maximum number of turns and subsequent resolv-ing powers for common calibrants at 100 Hz. As can be seen in this table,ideal mass measurements would be for isotopes lighter than 39K to reachMR-ToF-MS’s full potential of over 200,000 resolving power.3.5.3 MagneToF Detector SystemThe MR-ToF-MS system was originally equipped with a Microchannel Plate(MCP) installed as the detector [32]. However, it was found that there wasa strong injection-steering dependent variation in detection efficiency on theMCP. This lead to the tuning of the analyzer being further complicated.To simplify the tuning procedure, a new MagneTOF detector was installed.This detector was chosen due to its high efficiency, narrow pulse width, androbust nature (e.g. long lifetime, low noise, and no need to store in vacuum[52]. The detector mounted to its base is picture in Figure 3.19.After the installation of the MagneTOF detector, the tuning of the beamwas simplified. It was found that there was no longer a position sensitivedetection efficiency on the detector. To demonstrate this, the injection steer-ing voltages were scanned over their optimum values, and the relative countrates were plotted. The plots can be seen in Figure 3.20. Additionally, theoverall detection efficiency was increased from ≈ 60% to 80% when comparedto the maximum of the MCP.3.5.4 RF Frequency ModificationAs mentioned in section 3.2, the resonant frequency of the RF generators arecurrently adjusted manually. In its normal mode of operation, the MR-ToF-MS can efficiently transport a wide range of masses (≈ 35 - 170 u). However,for particularly light masses, the resonant frequency has to be adjusted (seesection 2.1 for the theory). This involves changing the capacitance and/orthe inductance of the circuit. In this particular case, both the capacitanceand inductance were reduced. To reduce the capacitance, the SHV (safehigh voltage) cables from the generator to the vacuum feedthroughs wereshortened reducing the capacitive load. To reduce the inductance of thegenerator, one of the wound coils was modified such that only half of the503.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.19: The new MagneTOF detector mounted to its base prior toinstallation in the MR-ToF-MS.513.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.20: Top: A 2D scan of injection steering voltages with MCP in-stalled. Bottom: A 2D scan on injection steering voltages with the Mag-neTOF installed. Note that the scales of both plots are identical. Thisdemonstrates the lack of position sensitive efficiency on the MagneTOF, andconsequentially, the reduced need for fine tuning of the injection steering.523.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.21: The MR-ToF-MS RF generator box with one modified coil suchthat the inductance is reduced by ≈1/2.coil is used. Figure 3.21 shows the interior inductance coils of one of theRF generators. After the reduction in inductance and capacitance, the MR-ToF-MS RFQs were able to transport ions with a mass as low as 19 u(corresponding, for example, to H3O+).3.5.5 Improved Switching OperationThe point at which the MR-ToF-MS is susceptible to systematic errors isduring the switching of voltages after the isobaric bunches have separated.This is most prominent when lowering the voltages of the second electrostaticion mirror to allow ions to leave the analyzer (as discussed in section 3.4.1).Studies showed this relative systematic uncertainty previously to be on alevel of 3 × 10−7 [13, 14, 39]. This number is determined from varying theopening time of the second mirror, recording the scatter in time-of-flight ofthe ion of interest, and fitting the Gaussian distribution of the ToF scatter.The ultimate cause of the error during switching is the ringing of voltagesafter switching. The ringing of a voltage is characterized by the outputoscillating above and below the desired value after switching. This can be533.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.22: Diagram of the HV switches used at the TITAN MR-ToF-MS.The red resistors were replaced with 510 Ω pulse withstanding resistors.seen by connecting the voltage to an oscilloscope via a voltage divider. Inprinciple, there should be a smooth exponential decay in voltage. However,fluctuations in the voltage were observed after switching. This provides astrong indication that ringing is occurring. To prevent this, the switchingspeed needs to be reduced.The switch consists of two RC circuits, one on high voltage, and one onlow voltage, with the capacitance to ground. The switching is controlledby a pair of IGBTs (insulate-gate bipolar transistor) and logic boxes. Aschematic of the circuit can be seen in Figure 3.22. In an RC circuit, the timeconstant is defined as τ = RC and can be used to describe the rise and falltime of the switch. The ringing of the voltages indicates that the switchingis happening too fast, so the time constant must be increased. In this case,it would be difficult to add more capacitance due to physical limitations onthe circuit board, so the resistance was changed from 200 Ω to 510 Ω usingresistors which fulfill the requirement of being pulse withstanding.Changing resistors achieved a doubling of the switching time which re-sulted in no negative effect on the ions. The initial τ was ≈ 2× 10−8 s andthe new τ is ≈ 6 × 10−8 s. Additionally, the ringing after switching is nowsignificantly lower. Ultimately, this leads to a lower systematic uncertainty.The same process as described above was again used to determine the sys-tematic uncertainty, and has now been improved from 3 to 1× 10−7 as canbe seen in Figure 3.23.543.5. Recent Upgrades to the MR-ToF-MS SystemFigure 3.23: A Gaussian fit of the histogram of systematic variation due tothe switching of the second electrostatic mirror after improvements to theswitching were made. This demonstrates an improvement by a factor of 3when compared to the previous switching.55Chapter 4Collision InducedDissociation (CID) Using theMR-ToF-MSCollision induced dissociation (CID) is a common technique in analyticchemistry. It is a relatively unexplored application in the field of experi-mental nuclear physics. Some investigations have been done at LEBIT [34]and JLU-Gießen [53] in which CID is used for beam purification, but a fullcharacterization of CID at TITAN has not been done yet.4.1 CID in Analytic ChemistryIn analytic chemistry, CID is a well understood technique that is used tostudy the thermodynamic properties of molecular ions. The process of CIDinvolves the excitation of rotational and vibrational states in ionic moleculesdue to collisions with an neutral atomic buffer gas. The energy of the col-lisions is varied to gain insight into the bond strengths of the molecules[54]. The general scheme for conducting CID is a three step process ofpre-separation, dissociation, and post-separation. First, a beam of molec-ular and atomic ions undergoes separation such that only one mass unitis selected. In controlled experiments, this is normally a single molecularspecies, but this is not the case for RIB molecular contamination. Then,the remaining beam undergoes dissociation using one of a variety of meth-ods. The resulting dissociation products undergo another separation andare analyzed. A general schematic for this can be found in Figure 4.1.Within the general three step scheme, there exists a variety of methodsfor performing CID. The processes are classified as follows. High energyCID uses beams with an energy of ≈ 1 - 10 keV. To perform the first stepof separation, a dipole magnet is used. After the dipole, there is a chamberfilled with gas in which the collisional dissociation takes place. The CIDproducts then travel through an electrostatic bender before being counted564.2. CID at TITANFigure 4.1: The three step general schematic for performing a measurementwith CID. Initially, the ions undergo a pre-separation so that there is a singlemass unit (green). The ions then undergo a dissociation process creatingions of different masses. A post-separation can then take place to againisolate a single mass unit (red).on a detector [55]. For low energy CID, beams with ≈ 1 - 100 eV areused. The most common method used at this energy is the Quadrupole-Quadrupole-Time-of-Flight technique. The initial separation is done by aquadrupole mass filter. The remaining ions are injected into a gas-filled RFQwhere dissociation occurs. The products are then injected into a time-of-flight mass spectrometer (in most cases a reflectron) for identification [56].Additionally, low energy CID can be performed on trapped ions insteadof beams. This is done through the Quadrupole Ion Trap technique (alsoreferred to as resonance excitation). In this technique, the ions are trappedin an RFQ and excited in the buffer gas. The excitation from the RF causescollisions with the buffer gas. The excitation amplitude and time can bevaried to probe the properties of the molecular bonds [57].4.2 CID at TITANThe method for performing CID at TITAN closely resembles the Quadrupole-Quadrupole-Time-of-Flight technique that is described in [56]. The steps areas follows: First, a single mass unit of radioactive or stable beam is deliveredto the TITAN Cooler-Buncher at 20 keV as a mixture of atomic and molec-ular ions. The continuous beam is converted to bunched beam and ejected574.2. CID at TITANFigure 4.2: The electric potential and beam energy superimposed over thepath from the TITAN Cooler-Buncher to the MR-ToF-MS. By adjusting theinitial potential of the pulsed drift tube before it switches to ground, theenergy at which the beam enters the input RFQ of the MR-ToF-MS. Imagemodified from [58].towards the MR-ToF-MS. As the bunched beam is traveling through a drifttube, it is pulsed down in energy from 20 keV to ≈ 1.3 keV by a switchingdrift tube. By changing the potential of the pulsed drift tube before switch-ing to ground, the kinetic energy of the beam entering the MR-ToF-MS inputRFQ can be varied. However, due to the maximum voltage allowed (so thatsparking is prevented) on the RFQ aperture (which allows for capture ofthe beam), the maximum injection energy is ≈ 1.45 keV. The MR-ToF-MSoperates at 1.35 kV, so the maximum kinetic energy of the beam is ≈ 100eV. Thus, the CID done at TITAN falls into the low-energy regime. Theenergy of the beam as it travels through the TITAN experiment as well asthe electric potentials it interacts with can be found in Figure 4.2584.3. ResultsOnce the beam undergoes CID in the input RFQ, it is transportedthrough the system to the analyzer. ToF-MS is then performed on thebeam as is described in section 3.4.1. For the characterization of CID, twodifferent versions of MS are employed. The so-called broadband mode ispurposefully employed to shorten the flight path of the ions by using 3 turnsor less. This allows for the observation of many mass units simultaneouslysuch that the mass of the initial molecule can be observed along with themasses of the breakup products. In a subsequent step, the so-called highresolution mode is used to increase the length of the flight path by increas-ing the number of turns to 300 or more. This mode of operation focuses ona single mass unit while ignoring species at different mass units. This al-lows for the separation and identification of individual atomic and molecularspecies. Resulting spectra show the outcomes generated by the two modesof operation (Figure 4.3).To characterize CID at TITAN, the Off-Line Ion Source (OLIS) deliveredstable beam with a mix of atomic and molecular species at two different massunits [59]. At A = 78 u, a beam with one atomic and one molecular specieswas measured as a first case. At A = 76 u, a beam with several atomic andmolecular species was measured as a second, more complex case. For bothcases, an initial spectrum was taken in the broadband mode of operation.This was to identify the relevant mass units to focus on (i.e. the mainmasses supplied by CID or molecular pickup). Once the key mass unitswere identified, high resolution spectra were taken to identify the individualspecies at each mass unit. Additionally, while in high resolution mode, theinjection energy was scanned, and the transmission rate of each species wasmeasured. A relationship between the breakup or the creation of individualspecies and the injection energy can then be determined.4.3 ResultsAs previously mentioned, the beam at A = 78 consisted of one atomicspecies, 78Kr+, and one molecular species, CH2O2S+. When scanning in-jection energy in high resolution mode for A = 78 u, it was found that thetransmission rate for 78Kr+ remained constant while the rate for CH2O2S+decreased at high injection energy. Next, the molecular dissociation productwas identified as CHOS+ by using the broadband and high resolution modesof operation in tandem. The injection energy was scanned again, and therate of transmission increased as injection energy increased. The transmis-sion rate of each species as a function of injection energy can be seen in594.3. ResultsFigure 4.3: A comparison of spectra using the MR-ToF-MS’s two differentmodes of operation. Bottom: A broadband view of A = 76 beam injectedinto the MR-ToF-MS. The pickup and breakup products can be seen fromA = 60 to A = 81. Top: A high resolution spectra at A = 76. From thisspectra, individual atomic and molecular species can be determined.604.3. ResultsFigure 4.4: A scan of the injection energy for A = 78 beam. For higherinjection energies, it can be seen that the 78Kr+ rate remains constant, theCH2O2S+ rate decreases, and the CHOS+ rate increases.Figure 4.4. The stable transmission of 78Kr+ indicates that losses do notoccur in the input RFQ due to an increase in the injection energy (up to acertain point, as discussed in the last section, when ion energy exceeds thetrapping potential). This means that the decrease in CH2O2S+ is due toCID. The increase in transmission of CHOS+ further reinforces this point.For the more complicated beam at A = 76, many different atomic andmolecular species were identified in the high resolution mode. Additionally,it was found that there were dissociation products at many different massunits from the broadband mass spectrum. Again, the injection energy wasscanned, and the transmission rates of the different species were recorded.The upper limit of the energy transferred due to collisions can be calculated,assuming purely elastic collisions, using:ET =mHemion +mHeEinj (4.1)614.3. ResultsFigure 4.5: The scan of injection energy (with maximum energy transferred)and the resulting rates of different atomic and molecular species for A = 76.where ET is energy transferred, mHe is the mass of He, mion is the mass of theinjected ion, and Einj is the injection energy. These calculated values are usedas a secondary x-axis on the top part of Figure 4.5. When plotted together, itcan be seen that the overall transmission rates of different molecular speciesdecline at differing rates and amounts. This is to be expected due to thedifference in molecular binding energies from one species to another.To further explore the difference in breakup rates, a plot of calculatedbond energy as a function of suppression factor can be created. Suppressionfactor (S) can be defined as:S =Rate(Einj = 10eV)Rate(Einj = 100eV). (4.2)Where Rate(Einj) referes to the count rate of a species at a given injectionenergy. This value essentially represents the ratio of CID ‘turned off’ to CID‘turned on’ where larger values indicate a greater decrease in transmission.624.3. ResultsFigure 4.6: The relationship between the bond energies of various moleculesand their suppression factor.As can be seen in Figure 4.6, there is a clear relation between bond energyand the suppression factor.For future mass measurements, the most common mode of operationwill be with ‘CID on’ (i.e. inject ions at ≈ 100 eV). This is to ensurethat the suppression of molecular contamination is maximized. This newlydeveloped technique should then allow for future measurements in beamswith increasing amounts of molecular contamination. This issue will beseen in the next chapter.63Chapter 5Neon Isotope MassMeasurements and ResultsTITAN took part in an ISAC beam development as part of a neutron-richNe mass measurement campaign. For this experiment, 500 MeV protonsimpinged on a UCx (Uranium Carbide) target. A Forced Electron BeamInduced Arc Discharge (FEBIAD) ion source was used. This ion source iscapable of ionizing elements with electron binding energies of >8 eV (suchas Ne)[33]. However, a consequence of such an ion source is the ionizationof additional atomic and molecular species, resulting in a large contamina-tion. Figure 1.11 indicates the level of background compared to the ob-served Ne yields. Even after employing the ISAC separator technique, ahigh background persists. To counteract this, mass-selective re-trapping (aswas mentioned in section 3.4.2) was employed.5.1 Neon Isotope Mass Measurement DataMass spectra for 24−26Ne were successfully taken in an experimental cam-paign at ISAC using the MR-ToF-MS system. The background found ateach mass provided calibrant isotopes. To enable the measurements, re-trapping was carried out for each Ne isotope. An example of re-trappingbeing implemented is shown in Figure 5.1. In the particular case of 24Ne,the 24Mg contaminant was suppressed by a factor of 104 relative to 24Newhile the 12C2 was purposefully not suppressed so that a suitable calibrantwas still available. The full spectra for A = 24 - 26 with re-trapping on isplotted in Figure 5.2 with different species labeled.In an effort to further reduce the statistical uncertainties, mass measure-ments of the same species were performed at 380, 400, 420, and 440 turns.Each spectra underwent an independent fitting procedure (which will be dis-cussed in the next section). After obtaining a mass value for each spectra,a weighted average was then taken for each of the three types of fits. Aweighted average is permitted due to the independence of each spectra. If645.1. Neon Isotope Mass Measurement DataFigure 5.1: Mass spectra taken with the TITAN MR-ToF-MS at mass =24 u. Re-trapping (lower panel) was employed to suppress 24Mg in favor ofmeasuring 24Ne.655.1. Neon Isotope Mass Measurement DataFigure 5.2: The mass measurement spectra from MR-ToF-MS for 24−26Newith other species labelled. Panel 1 shows mass 24. Panel 2 shows mass 25.Panel 3 shows mass 26.665.1. Neon Isotope Mass Measurement DataFigure 5.3: The resulting mass values from 24Ne mass measurements at dif-ferent MR-ToF-MS settings (corresponding to differing file numbers). Ad-ditionally, the different fitting methods are compared. The results from theGaussian fit are plotted in dark blue. The results from the Lorentzian fitare plotted in red. The results from the Inverse Polynomial fit are plottedin light blue. As can be seen in the plot, the values are in good agreementwithin 1σ.each measurement was taken with the same settings, this would not havebeen possible. An example of this is plotted in Figure 5.3. To get the fi-nal values, the three weighted values from the three fits are averaged andthe systematic uncertainty of δm/m = 3 × 10−7 is added to the statisticaluncertainty. Additional analysis details will be provided in the subsequentsections.675.1. Neon Isotope Mass Measurement Data5.1.1 Fitting Procedure to Extract Mass ValuesWhen fitting the peaks found in the mass-spectra, three different functionswere employed. They are Gaussian, Lorentzian, and Inverse Polynomial.This was done to further understand the ideal peak shape produced by theTITAN MR-ToF-MS. The Gaussian function takes the form:y = y0 +Aσ√pi/2e−2(x−xc)2σ2 (5.1)where y0 is the offset, xc is the peak centroid, A is the peak area, and σ isthe peak width. Full width half max maximum (FWHM) can be calculatedusing FWHM = σ√ln(4). The Lorentzian function used is:y = y0 +2Apiw4(x− xc)2 + w. (5.2)Again, y0 is the offset, xc is the peak centroid, and A is the peak area.However, w now defines the FWHM. Lastly, the Inverse Polynomial functionis defined as:y = y0 +A1 +A1(2x−xcw )2 +A2(2x−xcw )4 +A3(2x−xcw )6(5.3)where y0, xc, A, and w have the same definitions as in the Lorentzian func-tion. However, A1, A2, and A3 add additional degrees of freedom to betterfit the tails of the peak. For symmetric, nearly Gaussian peaks (i.e. distri-butions with a centroid described by a Gaussian function but non-Gaussiantails), A1 ≈ A2 ≈ A3 ≈ 1. For all fits, width (σ or w) was fit as a commonvalue for all peaks in the same spectra such that all peaks have the sameshape. An example of the three different fits can be found in Figure 5.4. Itis quite apparent that the Gaussian fit underestimates the tails of the peakwhile the Lorentzian fit overestimates the tails. The Inverse Polynomial fitproduces tails between the other two fits but still fails to precisely fit thetails. To solve this problem, an Exponentially Modified Gaussian functionis in development for fitting MR-ToF-MS spectra based on previous workdone at GSI and JLU-Gießen [60]. However, the choice of the model doesnot have a strong effect on the final result as can be seen in Figure 5.4.5.1.2 Propagation of ErrorsAfter fitting all peaks and getting the centroid values for the calibrants andspecies of interest, a final calculation is done to derive the mass value forthe species of interest:685.1. Neon Isotope Mass Measurement DataFigure 5.4: A mass spectra taken with MR-ToF-MS with the three differentfitting functions used in the Ne data analysis. All peaks are fit with acommon width for each function so that they have the same peak shape.695.1. Neon Isotope Mass Measurement Datamfatomic =mmionic × (Mlitatomic −me)Mmionic+me (5.4)where mfatomic is the atomic mass of the isotope of interest, mmionic is themeasured ionic mass of the isotope of interest, Mlitatomic is the literatureatomic mass value of the calibrant, Mmionic is the measured ionic mass ofthe calibrant, and me is the mass of the electron. To determine the finaluncertainty of mfatomic , the uncertainty of each term is accounted for.The calibrant for each measurement and me have known uncertaintiesfrom literature. For the measured values, there are multiple sources of un-certainty that need to be evaluated and added in quadrature. When fittinga peak, the centroid has a fit uncertainty of δxc. The statistical uncertaintyis defined by the peak width (FWHM) and number of counts (N) such that:δmstat =FWHM√N. (5.5)Thus, the uncertainty of each independent isotope of interest with a finalcalibration is:δmfatomic =√δx2ccal + δx2cIoI+ δm2statcal + δm2statIoI. (5.6)These individual uncertainties are then plotted with their respectivemasses, and a weighted mean is taken for each mass unit and fit type. Theweighted mean (χ) is calculated using:χ =1Σ 1δm × Σ mδm(5.7)and the uncertainty is derived using:δχ =√Σ (m−χ)2δmΣ 1δm. (5.8)To extract the final mass, the weighted mean of each fit function isaveraged such that:mavg =χGauss + χLorentz + χInvPoly3. (5.9)The calculation of the final uncertainty uses the δχ and previously deter-mined δmsystematic (calculated to be 3× 10−7 as discussed in section 3.5.5).This leads to:705.1. Neon Isotope Mass Measurement Dataδmavgmavg=√δm2sys +(δχGaussχGauss)2+(δχLorentzχLorentz)2+(δχInvPolyχInvPoly)2. (5.10)Due to the many measurements at different settings for each isotope, theuncertainty from the weighted means contributes very little. Thus, thedominant source of uncertainty is the systematic component of 3×10−7. Thisdemonstrates the importance of the reduction of the systematic uncertaintyas was discussed in section 3.5.5.5.1.3 Final Mass Results and Comparison with the AtomicMass Evaluation (AME2016) Literature ValuesTo report the final values in Table 5.1, the atomic masses were converted toMass Excess (ME) using:ME = (matomic −A)× 931494.028keVc2. (5.11)Additionally, the mass ratio of isotope of interest and calibrant is reportedin the table as mmcal . This will allow for the recalculation of the masses ifany calibrant mass changes.Species Calibrant ME (keV/c2) mmcal24Ne 12C2 -5954.0 (6.9) 0.99973367 (31)25Ne 12C13C -2023.8 (8.0) 0.99977893 (34)26Ne 12C14N 456.4 (8.0) 0.99990063 (33)Table 5.1: The reported values for the Ne mass measurements with theirrespective calibrants. Both Mass Excess (ME) and mass ratio are reportedas atomic masses.To compare the three measured masses with literature, they are plottedas A versus AME2016 − TITAN in Figure 5.5. The AME is a standardizedevaluation program where all available published data is accumulated anda documented procedure is applied. As a result, tables are produced thatlist the Mass Excess of all known isotopes. Additionally, the uncertaintiesof the AME values are superimposed on the plot in gray. It can be seen thatall values agree within one sigma. The TITAN measurement for 24Ne willnot impact the AME mass uncertainty. However, for 25Ne and 26Ne, theuncertainties are reduced by a factor of three and two respectively.715.1. Neon Isotope Mass Measurement DataFigure 5.5: The difference between AME2016 mass values and the newTITAN measurements for 23−27Ne. The uncertainties of the AME valuesare shaded in gray. From the plot, it can be seen that the TITAN values arein good agreement with previous data and have reduced the uncertaintiesfor 25Ne and 26Ne. For 27Ne, the uncertainty drastically increases.725.2. Shortcomings of the Ne Mass MeasurementsFigure 5.6: The background contamination of beam from a FEBIAD ionsource for A = 23 - 32.5.2 Shortcomings of the Ne Mass MeasurementsAs mentioned in section 1.1, the N = 20 Island of Inversion is a region of greatinterest (see Figure 1.3) due to the weakening of the N = 20 shell. However,to reach this region, measurements of 29−31Ne are required. Reaching thisregion will prove to be exceedingly difficult due to the presence of largeamounts of molecular contamination. This can be seen in Figure 5.6 wherethe ratio of background to Ne yields approaches 1010. A measurement wasattempted for A = 27, but due to the molecular contamination, the 27Ne wasobscured (Figure 5.7). As can be seen in the figure, the re-trapping windowwas too wide to maximize the suppression of contaminants. After completingthis experiment, the results were analyzed, and further development of re-trapping optimization was done to increase the resolving power in this modeof operation (section 3.4.2).Re-trapping can, at best, provide a suppression of background on theabundance level of 104 (hence, one ion of interest to 104 ions of contaminationwith their respective peaks), and the MR-ToF-MS has a dynamic range of104. Thus, if there exists a background to ion of interest ratio of greater than735.2. Shortcomings of the Ne Mass MeasurementsFigure 5.7: The attempted 27Ne measurement. The re-trapping windowis outlined in red and blue. Outside the blue lines, re-trapping suppressesspecies by ≈ 104 while between the red and blue lines, re-trapping suppressesspecies by ≈ 101. The expected location of the 27Ne is marked by the verticalblack line.108, the measurement will remain impossible for now. As indicated in Figure5.6, this is a possibility. For the neutron-rich Ne isotopes in particular, thedominant contaminant species are molecules. Thus, the need for CID isapparent when attempting measurements of the most exotic species of Ne.74Chapter 6Summary, Conclusions, andOutlookThe mass of atoms is a fundamental property and provides knowledge of theinteractions at work. Thus, there is a need for precise mass measurements toprovide insight into nuclear structure, nuclear astrophysics, and fundamentalsymmetries. Additionally, the common modern forms of mass-spectrometryhave been compared. A brief overview of the production of RIB at both ISOLand In-Flight facilities has been given along with a discussion on sources ofunwanted beam contamination. As a result, the benefits of an MR-ToF-MSin experimental nuclear physics has been established.The TITAN MR-ToF-MS has proven to be a fast, precise, and sensitivemass-spectrometer. Additionally, a high degree of efficiency, flexibility, andaccuracy has been demonstrated. Moreover, these qualities have all beenimproved recently to enable mass measurements of increasingly rare isotopes.Some of the aforementioned improvements were utilized during a mea-surement campaign of the masses of 24−26Ne. The uncertainties of the massvalues of 25Ne and 26Ne were improved compared to the previous literaturevalues. However, the penultimate goal of measuring 31Ne was not reached.To counteract the issues faced when measuring increasingly neutron-rich Neisotopes, mass-selective re-trapping was studied and further improved, andCollision Induced Dissociation at TITAN was developed.To do this, a stable ion beam (non-radioactive) composed of both atomicand molecular ions from an off-line ion source was used. After the compo-sition of the beam was determined, the injection energy was varied, and achange in transmission rates was observed. A decrease in molecular trans-mission indicates an increase in the occurrence of CID. From this study, itwas found that molecular transmission can be reduced by one to two ordersof magnitude with an energy of Einj = 100 eV. Thus, the general mode ofoperation will be with this injection energy. However, the precise amount ishighly dependent on the molecular bond strength.With these recent developments, the TITAN MR-ToF-MS is well equippedto continue preforming high-precision mass-measurements of increasingly75Chapter 6. Summary, Conclusions, and Outlookexotic isotopes (i.e. very neutron rich or deficient). With the ability to in-duce dissociation, the MR-ToF-MS is more prepared to handle beams withincreasing amounts of molecular contamination. In particular, the MR-ToF-MS is well situated to measure more neutron-rich Ne isotopes in the future.76Bibliography[1] O. Sorlin and M. G. Porquet. Nuclear magic numbers: New features farfrom stability. Progress in Particle and Nuclear Physics, 61(2):602–673,2008.[2] M. Arnould, S. Goriely, and K. Takahashi. The r-process of stellarnucleosynthesis: Astrophysics and nuclear physics achievements andmysteries, 2007.[3] H. Schatz, A. Aprahamian, V. Barnard, L. Bildsten, A. Cumming,M. Ouellette, T. Rauscher, F. K. Thielemann, and M. Wiescher. EndPoint of the rp Process on Accreting Neutron Stars. Physical ReviewLetters, 86(16):3471–3474, 2001.[4] I. S. Towner and J. C. Hardy. The evaluation of Vud and its impact onthe unitarity of the Cabibbo-Kobayashi-Maskawa quark-mixing matrix.Reports on Progress in Physics, 73(4), 2010.[5] Francis W. Aston. Mass Spectra and Isotopes. Nobel Lecture, 1922.[6] A. S. Eddington. The Internal Constitution of the Stars. The ScientificMonthly, 11(4):297–303, 1920.[7] J. Chadwick. Possible Existence of a Neutron. Nature, 129(3252):312,1932.[8] C. F. von Weizsa¨cker. Zur Theorie der Kernmassen. Zeitschrift fu¨rPhysik, 96(7):431–458, 7 1935.[9] M. Goeppert Mayer and J. H. Jensen. Elementary Theory of NuclearShell Structure. John Wiley and Sons, New York, 1955.[10] J. M. Pearson. The Quest for a Microscopic Nuclear Mass Formula.In David Lunney, Georges Audi, and H. Jurgen Kluge, editors, AtomicPhysics at Accelerators: Mass Spectrometry, pages 59–74, Dordrecht,2001. Springer Netherlands.77Bibliography[11] M. Wang, G. Audi, F. G. Kondev, W. J. Huang, S. Naimi, and X. Xu.The AME2016 atomic mass evaluation (II). Tables , graphs and refer-ences. Chinese Physics C, 41(3), 2017.[12] F. Wienholtz, D. Beck, K. Blaum, Ch. Borgmann, M. Breitenfeldt,R. B. Cakirli, S. George, F. Herfurth, J. D. Holt, M. Kowalska,S. Kreim, D. Lunney, V. Manea, J. Mene´ndez, D. Neidherr, M. Rosen-busch, L. Schweikhard, A. Schwenk, J. Simonis, J. Stanja, R. N. Wolf,and K. Zuber. Masses of exotic calcium isotopes pin down nuclearforces. Nature, 498(7454):346–349, 2013.[13] E. Leistenschneider, M. P. Reiter, S. A. San Andre´s, B. Kootte,J. D. Holt, P. Navra´til, C. Babcock, C. Barbieri, B. R. Barquest,J. Bergmann, J. Bollig, T. Brunner, E. Dunling, A. Finlay, H. Geissel,L. Graham, F. Greiner, H. Hergert, C. Hornung, C. Jesch, R. Klawit-ter, Y. Lan, D. Lascar, K. G. Leach, W. Lippert, J. E. McKay, S. F.Paul, A. Schwenk, D. Short, J. Simonis, V. Soma`, R. Steinbru¨gge, S. R.Stroberg, R. Thompson, M. E. Wieser, C. Will, M. I. Yavor, C. An-dreoiu, T. Dickel, I. Dillmann, G. Gwinner, W. R. Plaß, C. Scheiden-berger, A. A. Kwiatkowski, and J. Dilling. Dawning of the N = 32Shell Closure Seen through Precision Mass Measurements of Neutron-Rich Titanium Isotopes. Physical Review Letters, 120(6):62503, 2018.[14] M. P. Reiter, S. A. San Andre´s, E. Dunling, B. Kootte, E. Leistenschnei-der, C. Andreoiu, C. Babcock, B. R. Barquest, J. Bollig, T. Brunner,I. Dillmann, A. Finlay, G. Gwinner, L. Graham, J. D. Holt, C. Hornung,C. Jesch, R. Klawitter, Y. Lan, D. Lascar, J. E. McKay, S. F. Paul,R. Steinbru¨gge, R. Thompson, J. L. Tracy, M. E. Wieser, C. Will,T. Dickel, W. R. Plaß, C. Scheidenberger, A. A. Kwiatkowski, andJ. Dilling. Quenching of the N=32 neutron shell closure studied viaprecision mass measurements of neutron-rich vanadium isotopes. Phys-ical Review C, 98(2):024310, 2018.[15] E. Burbidge, G. Burbidge, W. Fowler, and F. Hoyle. Synthesis of theElements in Stars. Review of Modern Physics, 29(4):547–650, 1957.[16] H. Schatz and W. J. Ong. Dependence of X-ray Burst Models on Nu-clear Masses. The Astrophysical Journal, 844(2):139, 2016.[17] M. R. Mumpower, R. Surman, D. L. Fang, M. Beard, P. Mo¨ller,T. Kawano, and A. Aprahamian. Impact of individual nuclear masseson r -process abundances. Physical Review C, 92(3):035807, 2015.78Bibliography[18] C. B. Hinke, M. Bo¨hmer, P. Boutachkov, T. Faestermann, H. Geissel,J. Gerl, R. Gernha¨user, M. Go´rska, A. Gottardo, H. Grawe, J. L. Gr-bosz, R. Kru¨cken, N. Kurz, Z. Liu, L. Maier, F. Nowacki, S. Pietri,Zs. Podolya´k, K. Sieja, K. Steiger, K. Straub, H. Weick, H. J. Woller-sheim, P. J. Woods, N. Al-Dahan, N. Alkhomashi, A. Atac¸, A. Blazhev,N. F. Braun, I. T. Cˇelikovic´, T. Davinson, I. Dillmann, C. Domingo-Pardo, P. C. Doornenbal, G. De France, G. F. Farrelly, F. Farinon,N. Goel, T. C. Habermann, R. Hoischen, R. Janik, M. Karny, A. Kas¸kas¸,I. M. Kojouharov, Th. Kro¨ll, Y. Litvinov, S. Myalski, F. Nebel,S. Nishimura, C. Nociforo, J. Nyberg, A. R. Parikh, A. Procha´zka,P. H. Regan, C. Rigollet, H. Schaffner, C. Scheidenberger, S. Schwer-tel, P. A. So¨derstro¨m, S. J. Steer, A. Stolz, and P. Strme. Superal-lowed Gamow-Teller decay of the doubly magic nucleus 100Sn. Nature,486(7403):341–345, 2012.[19] Fritz Bosch. Schottky mass- and lifetime-spectrometry of unstable,stored ions. Journal of Physics B: Atomic, Molecular and OpticalPhysics, 36(3):585–597, 2003.[20] M. Hausmann, F. Attallah, K. Beckert, F. Bosch, A. Dolinskiy,H. Eickhoff, M. Falch, B. Franczak, B. Franzke, H. Geissel, Th. Ker-scher, O. Klepper, H. J. Kluge, C. Kozhuharov, K. E. G. Lo¨bner,G. Mu¨nzenberg, F. Nolden, Yu. N. Novikov, T. Radon, H. Schatz,C. Scheidenberger, J. Stadlmann, M. Steck, T. Winkler, and H. Woll-nik. First isochronous mass spectrometry at the experimental storagering ESR. Nuclear Instruments and Methods in Physics Research A,446(3):569–580, 2000.[21] H. G. Dehmelt. Experiments With an Isolated Subatomic Particle atRest. Rev. Mod. Phys., 62(3):525–531, 1989.[22] M. Brodeur, V. L. Ryjkov, T. Brunner, S. Ettenauer, A. T. Gal-lant, V. V. Simon, M. J. Smith, A. Lapierre, R. Ringle, P. Delheij,M. Good, D. Lunney, and J. Dilling. Verifying the accuracy of the TI-TAN Penning-trap mass spectrometer. International Journal of MassSpectrometry, 2012.[23] M. Ko¨nig, G. Bollen, H. J. Kluge, T. Otto, and J. Szerypo. Quadrupoleexcitation of stored ion motion at the true cyclotron frequency. Inter-national Journal of Mass Spectrometry and Ion Processes, 142(1-2):95–116, 1995.79Bibliography[24] Klaus Blaum, Jens Dilling, and Wilfried No¨rtersha¨user. Precisionatomic physics techniques for nuclear physics with radioactive beams.Physica Scripta, T152:14017, 2013.[25] D J Vieira, J M Wouters, K Vaziri, R H Kraus, H Wollnik, G W Butler,F K Wohn, and A H Wapstra. Direct mass measurements of neutron-rich light nuclei near N=20. Phys. Rev. Lett., 57(26):3253–3256, 121986.[26] H Savajols. The SPEG Mass Measurement Program at GANIL. InDavid Lunney, Georges Audi, and H.-Jrgen Kluge, editors, AtomicPhysics at Accelerators: Mass Spectrometry, pages 245–254, Dordrecht,2001. Springer Netherlands.[27] T. Dickel, W. R. Plaß, A. Becker, U. Czok, H. Geissel, E. Haettner,C. Jesch, W. Kinsel, M. Petrick, C. Scheidenberger, A. Simon, andM. I. Yavor. A high-performance multiple-reflection time-of-flight massspectrometer and isobar separator for the research with exotic nuclei.Nuclear Instruments and Methods in Physics Research, Section A: Ac-celerators, Spectrometers, Detectors and Associated Equipment, 2015.[28] Y. Blumenfeld, T. Nilsson, and P. Van Duppen. Facilities and methodsfor radioactive ion beam production. Physica Scripta, 014023(T152),2013.[29] K. H. Schmidt, E. Hanelt, H. Geissel, G. Mu¨nzenberg, and J. P. Dufour.The Momentum-Loss Achromat - A New Method for the IsotopicalSeparation of Relativistic Heavy Ions. Nuclear Inst. and Methods inPhysics Research, A, 260(2-3):287–303, 1987.[30] H. Geissel, Th Schwab, P. Armbruster, J. P. Dufour, E. Hanelt, K. H.Schmidt, B. Sherrill, and G. Mu¨nzenberg. Ions penetrating throughion-optical systems and matter - non-liouvillian phase-space modelling.Nuclear Inst. and Methods in Physics Research, A, 1989.[31] B. R. Fulton. Present and Future RIB Facilities. Journal of Physics:Conference Series, 312(052001), 2011.[32] Christian Jesch. The Multiple-Reflection Time-of-Flight Isobar Separa-tor for TITAN and Direct Mass Measurements at the FRS Ion Catcher.PhD thesis, Justus-Liebig-Univeristy Giessen, 2016.80Bibliography[33] B. E. Schultz, J. Sandor, P. Kunz, A. Mjøs, O. Kester, and F. Ames.FEBIAD Ion Source Development At TRIUMFISAC. In 9th Interna-tional Particle Accelerator Conference (IPAC ’18), pages 4730–4733,Vancouver, Canada, 2018.[34] P. Schury, G. Bollen, M. Block, D. J. Morrissey, R. Ringle, A. Prinke,J. Savory, S. Schwarz, and T. Sun. Beam purification techniques forlow energy rare isotope beams from a gas cell. Hyperfine Interactions,173:165–170, 2006.[35] G .C. Ball, G. Hackman, and R. Kru¨cken. The TRIUMF-ISAC facility:two decades of discovery with rare isotope beams. Physica Scripta,91(093002), 2016.[36] J. Lassen, P. Bricault, M. Dombsky, J. P. Lavoie, Ch. Geppert, andK. Wendt. Resonant ionization laser ion source project at TRIUMF.Hyperfine Interactions, 162(1-4):69–75, 2005.[37] G. C. Ball, L. Buchmann, B. Davids, R. Kanungo, C. Ruiz, and C. E.Svensson. Physics with reaccelerated radioactive beams at TRIUMF-ISAC. Journal of Physics G: Nuclear and Particle Physics, 38(2), 2011.[38] J. Dilling, P. Bricault, M. Smith, and H. J. Kluge. The proposed TI-TAN facility at ISAC for very precise mass measurements on highlycharged short-lived isotopes. Nuclear Instruments and Methods inPhysics Research, Section B: Beam Interactions with Materials andAtoms, 204:492–496, 2003.[39] Christian Jesch, Timo Dickel, Wolfgang R. Plaß, Devin Short,Samuel Ayet San Andres, Jens Dilling, Hans Geissel, Florian Greiner,Johannes Lang, Kyle G Leach, Wayne Lippert, Christoph Scheiden-berger, and Mikhail I Yavor. The MR-TOF-MS isobar separator forthe TITAN facility at TRIUMF. Hyperfine Interact, 235:97–106, 2015.[40] T. Brunner, M. J. Smith, M. Brodeur, S. Ettenauer, A. T. Gallant,V. V. Simon, A. Chaudhuri, A. Lapierre, E. Mane´, R. Ringle, M. C. Si-mon, J. A. Vaz, P. Delheij, M. Good, M. R. Pearson, and J. Dilling. TI-TAN’s digital RFQ ion beam cooler and buncher, operation and perfor-mance. Nuclear Instruments and Methods in Physics Research, SectionA: Accelerators, Spectrometers, Detectors and Associated Equipment,2012.81Bibliography[41] P.H. Dawson. Quadrupole Mass Spectrometry and its Applications. El-sevier Science, New York, 1976.[42] P. Schury. High Precision Mass Measurements Near N = Z = 33. PhDthesis, Michigan State University, 2007.[43] Wolfgang R. Plaß, Timo Dickel, Ulrich Czok, Hans Geissel, MartinPetrick, Katrin Reinheimer, Christoph Scheidenberger, and Mikhail I.Yavor. Isobar separation by time-of-flight mass spectrometry for low-energy radioactive ion beam facilities. Nuclear Instruments and Meth-ods in Physics Research, Section B: Beam Interactions with Materialsand Atoms, 2008.[44] Wolfgang R. Plaß, Timo Dickel, and Christoph Scheidenberger.Multiple-reflection time-of-flight mass spectrometry. InternationalJournal of Mass Spectrometry, 2013.[45] R. Klawitter, A. Bader, M. Brodeur, U. Chowdhury, A. Chaudhuri,J. Fallis, A. T. Gallant, A. Grossheim, A. A. Kwiatkowski, D. Lascar,K. G. Leach, A. Lennarz, T. D. Macdonald, J. Pearkes, S. Seeraji, M. C.Simon, V. V. Simon, B. E. Schultz, and J. Dilling. Mass measurementsof neutron-rich Rb and Sr isotopes. Physical Review C, 93(4):1–7, 2016.[46] D. Lascar, R. Klawitter, C. Babcock, E. Leistenschneider, S. R.Stroberg, B. R. Barquest, A. Finlay, M. Foster, A. T. Gallant, P. Hunt,J. Kelly, B. Kootte, Y. Lan, S. F. Paul, M. L. Phan, M. P. Reiter,B. Schultz, D. Short, J. Simonis, C. Andreoiu, M. Brodeur, I. Dill-mann, G. Gwinner, J. D. Holt, A. A. Kwiatkowski, K. G. Leach, andJ. Dilling. Precision mass measurements of Cd 125-127 isotopes and iso-mers approaching the N=82 closed shell. Physical Review C, 96(4):1–7,2017.[47] D. Lascar, A. A. Kwiatkowski, M. Alanssari, U. Chowdhury, J. Even,A. Finlay, A. T. Gallant, M. Good, R. Klawitter, B. Kootte, T. Li, K. G.Leach, A. Lennarz, E. Leistenschneider, A. J. Mayer, B. E. Schultz,R. Schupp, D. A. Short, C. Andreoiu, J. Dilling, and G. Gwinner.Improvements to TITAN’s mass measurement and decay spectroscopycapabilities. Nuclear Instruments and Methods in Physics Research,Section B: Beam Interactions with Materials and Atoms, 2016.[48] Michael Wayne Froese. The TITAN Electron Beam Ion Trap: As-sembly, Characterization, and First Tests. PhD thesis, University ofManitoba, 2006.82Bibliography[49] Mikhail Yavor. Chapter 8 Time-of-Flight Mass Analyzers. Advances inImaging and Electron Physics, 157(09):283–316, 2009.[50] W. R. Plaß, T. Dickel, M. Petrick, D. Boutin, Z. Di, T. Flecken-stein, H. Geissel, C. Jesch, C. Scheidenberger, and Z. Wang. AnRF quadrupole - Time-of-flight system for isobar-separation and multi-plexed low-energy Rare-Isotope Beam experiments. European PhysicalJournal: Special Topics, 150(1):367–368, 2007.[51] W. R. Plaß, T. Dickel, S. San Andres, J. Ebert, F. Greiner, C. Hor-nung, Christian Jesch, Johannes Lang, Wayne Lippert, Tamas Ma-joros, Devin Short, Hans Geissel, Emma Haettner, Moritz P. Reiter,Ann Kathrin Rink, Christoph Scheidenberger, and Mikhail I. Yavor.High-performance multiple-reflection time-of-flight mass spectrometersfor research with exotic nuclei and for analytical mass spectrometry.Physica Scripta, 2015(T166), 2015.[52] Dick Stresau, Kevin Hunter, Wayne Shiels, Peter Raffin, and Yair Be-nari. A New Class of Robust Sub-nanosecond TOF Detectors with HighDynamic Range. In 54th ASMS Conference on Mass Spectroscopy, Seat-tle, Washington, 2006.[53] W. Lippert. Further Development and Application of a MobileMultiple-Reflection Time-of-Flight Mass Spectrometer for AnalyticalHigh-Resolution Tandem Mass Spectrometry. PhD thesis, Justus-Liebig-University Gießen, 2016.[54] F Muntean and P B Armentrout. Guided ion beam study of collision-induced dissociation dynamics : integral and differential cross sections.Journal of Chemical Physics, 115(3):1213–1228, 2001.[55] J. H. Beynon, R. G. Cooks, J. W. Amy, W. E. Baitinger, and T. Y.Ridley. Design and Performance of a Mass-analyzed Ion Kinetic Energy(MIKE) Spectrometer. Analytical Chemistry, 45(12):1023A–1031A,1973.[56] Howard R. Morris, Thanai Paxton, Anne Dell, Jean Langhorne,Matthias Berg, Robert S. Bordoli, John Hoyes, and Robert H. Bateman.High sensitivity collisionally-activated decomposition tandem massspectrometry on a novel quadrupole/orthogonal-acceleration time-of-flight mass spectrometer. Rapid Communications in Mass Spectrome-try, 10(8):889–896, 1996.83Bibliography[57] John N. Louris, R. Graham Cooks, John E. Syka, Paul E. Kelley,George C. Stafford, and John F. Todd. Instrumentation, Applications,and Energy Deposition in Quadrupole Ion-Trap Tandem Mass Spec-trometry. Analytical Chemistry, 59(13):1677–1685, 1987.[58] E. Leistenschneider. Dawning of Nuclear Magicity in N=32 Seenthrough Precision Mass Spectrometry. PhD thesis, University of BritishColumbia, 2019.[59] K. Jayamanna, F. Ames, G. Cojocaru, R. Baartman, P. Bricault,R. Dube, R. Laxdal, M. Marchetto, M. MacDonald, P. Schmor,G. Wight, and D. Yuan. Off-line ion source terminal for ISAC at TRI-UMF. Review of Scientific Instruments, 79(02C711), 2008.[60] S. Purushothaman, S. A. San Andre´s, J. Bergmann, T. Dickel, andJ. Ebert. Hyper-EMG : A new probability distribution function com-posed of Exponentially Modified Gaussian distributions to analyzeasymmetric peak shapes in high-resolution time-of-flight mass spec-trometry. International Journal of Mass Spectrometry, 421:245–254,2017.84

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0380561/manifest

Comment

Related Items