TRITON

1
TRITON

• Johannes Schwieters
• Thermo Electron (Bremen)

2
Design goals for TRITON

• Develop new high resolution optics for a new
  multicollector platformnext generation of
  MC-instruments (TIMS ICPMS)
• Break the 10 ppm external precision barrier
• Resolve the putative 142Nd anomaly 2ppm
  resolution
• Get rid of detector biases
• Cup factors
• Gain bias
• Smaller samples
• Increase sensitivity
• Improve Signal/Noise
• No me too !

3
The project name Neptune fountain in Bremen
NEPTUNE
TRITON
The Neptune - Fountain in Bremen
4
The Finnigan TRITON
Introduced 1998 Goldschmidt Conference,Toulouse,
France
5
The Finnigan NEPTUNE
Introduced 1999 Goldschmidt Conference,Boston,
US
6
Layout of the ion optics
Laminated magnet
Zoom lens
Focus quad
Source lens stack
Collector array
21 sample turret
RPQ-SEM
D
2.D
Mass dispersion 818mm
7
Classical approach Magnification close to 11
Faraday Cup
Source
M1
Magnet lens
High probability for secondary particles to
escape the cups cup factors !.
8
Effect of large magnification ion optics
Faraday Cup
Secondary particles are released close to
cup entrance.
Source
M1
Magnet lens
81 cm mass dispersion M2
Faraday Cup
Source
Magnet lens
9
The new Faraday Cups

• Plug-in design
• Machined from solid graphite
• Deepest Faraday cups

50 V dynamic range
Ions
10
Variable Multicollector
17 mass range 6Li7Li/ 42Ca-48Ca 202Hg-204Pb 23
8U
Beam guide for fast IC or FAR switching
Internal position encoding
11
Zoom optics for enhanced multidynamic
measurements
Ion Source
Focus
Zoom Quadrupoles
Multi-Collector
Dispersion
RPQ
Magnet
12
Multidynamic measurements the peak overlap
problem
Sr double collector
87Sr
88Sr
86Sr
87Sr
13
Static measurements The amplifier calibration
bias problem

• Classical concept External gain calibration
• High precision constant current source is
  sequentially connected to all amplifiers
  Precision of gain calibration assumption 4
  ppm /channel
• Consequence for Nd and Sr (three isotopes)
• The gain calibration bias limits the external
  precision to about

? 7ppm
But, design goal is 2 ppm !
14
Amplifier switching scheme
15
Example 143/144 (2x7 blocks)
16
1
V
F
2
V
F
3
V
F
V
4
F
5
V
F
6
V
F
7
V
F
8
V
F
V
9
F
Relay Matrix
Amplifier and V/F converter
relay open
relay closed
Front end processor
evacuated and thermostated housing
17
Result 142Nd/144Nd anomaly resolved !
Data from Caro et al. IPG-Paris
18
142Nd/144Nd resolution test Bremen demo lab
30.5 ppm
15.8 ppm
4.0 ppm
19
142Nd/144Nd resolution test
20
TRITON 143Nd/144Nd instrumental reproducibility
Error bars represent SE of 10 separate runs with
less than 5 ppm external reproducibilityMerck
standard
21
NEPTUNE 143Nd/144Nd instrumental reproducibility
Error bars represent SE of 9 blocks, each 100
data 8s Merck
22
TIMS and MC-ICPMS

• The mass bias generated in the TIMS source and
  the ICP-source are completely different
• Mass Bias in ICP ca. 10x larger than TIMS
• ICP gives higher efficiency for heavier isotopes
  while it is vice versa for TIMS
• People tend to use the same correction algorithms
  for both types of ion sources !
• Does this work ?

23
The mouth of truth
24
Accuracy of Nd-data (ICPMS) using different
fractionation laws
25
Fundamentals Current amplifiers
Resistor
Integration Digitization (UF-converter)
Faraday
Selection of high resistor values to
achieve smallest S/N !
R resistor value tm integration time T
temperature (Kelvin) U voltage at output
26
Fundamentals Current amplifiers
Noise level for different integration times
5 µV ca. 320 cps
27
Noise limit for small samples
TRITON low noise amplifier system working at the
limits Samples prepared on tungsten
filaments Using TaCl activator.
Data are from Bob Cliff, Leeds, UK
28
Why Multi-Ion-Counting (MIC) ?
1 mV Faraday signal
ca. 60.000 cps on IC
signal/noise
29
TRITON/NEPTUNE multicollector with MIC

• plug-in MIC detectors identical in size and
  inter- changeable with Faraday Cups
• Option Detector packages can be mounted onto the
  sides of Faraday cups
• Up to 8 MIC channels plus 9 Faraday cups can be
  installedsimultaneously
• Unit mass spacing possible up to U- mass range

30
Direct Single Particle Analysis MIC-TIMSUranium
isotopes (5-10µm particles)

• Nuclear safeguards and monitoring applications
  Detection of anomalous 233, 235, 236U from
  nuclear reactions

UO2 standard particle
Picture provided by Kristofor Ingeneri, IAEA,
Vienna
1 µm
31
Uranium particles Measurement and Evaluation
Strategy
TIMS Particles loaded directly onto V-shaped
filaments
20 s
1 s
1 s
1 s

• Static acquisition followed by rapid in-run
  cross calibration sequence by peak jumping of the
  major beam across the array

1 s
32
Results from Single Particle analysis (MIC-TIMS)
235U/238U
234U/238U
236U/238U
0.3 (1 RSD)

• Run duration 15 minutes
• 238U varies from 70 000 cps to15 000 cps
• All precisions within counting statistics !

0.3 (1 RSD)
Calibration Stability (for 238U and 235U ICs)
33
Summary Innovations in Multicollector Technology

• Ion optics
• Extended ion optical magnification mass
  dispersion (81 cm)
• Zoom optics for enhanced multidynamic
  measurements
• Variable multicollector
• Extended mass range 17
• In-situ high precision position control
• Variable in detector type integrated solution
  for MIC
• New plug-in Faraday cups improved cup
  performance
• Amplifier system
• Extended dynamic range of Faraday cup electronics
  (50 V _at_1011 Ohms)
• Virtual amplifier eliminates gain biases
• Ultimate precision and accuracy for static
  measurements

34
A different view of the instruments

• picture

35
Improvements on high precision
36
Anomaly in ISUA rocks clearly resolved
Data from Caro et al. , IPG-Paris