ITER Spectroscopic diagnostics D-T fusion requires lowest

ITER Spectroscopic diagnostics Robin Barnsley and ITER team

- Overview of ITER

- Generic activities

- Diagnostic integration

- Plasma emission modelling

- Stray light modelling

- Neutronics modelling

- Spectroscopic diagnostics

- Visible - VUV - X-ray spectroscopy - X-ray Camera

Leading magnetic confinement device is the Tokamak

- Closed magnetic field minimizes particle losses

- Combined toroidal and poloidal fields produce helical field, that stabilizes +ve/-ve charged particle drifts.

- Plasma current induced by inner poloidal coils.

- Additional heating from neutral beams and RF/microwave

- Self-heating from fusion alphas

1 Magnetic confinement fusion research D-T fusion requires lowest temperature Ein + D + T -> (4He + 3.5 MeV) + (n + 14.1 MeV) Breeding T from Li: nslow + 6Li -> 4He + T nfast + 7Li -> 4He + T + nslow Energy multiplication, Q: Q = Pout / Pin Breakeven, Q=1 Palpha + Pneutron = Pin

Ignition: self-heating Palpha = Pin

IAEA Trieste, 24 March 2015, R Barnsley

AUG JET

ITER

Scaling to ITER from previous experiments

Physics performance can be extrapolated better than factor 2.

Technological developments ongoing for:

- First wall: blanket and divertor modules.

- Material properties under heavy neutron irradiation.

Participating Teams: China, Europe, India, Japan, Russia, South Korea, USA. Construction site: St Paul les-Durance, Provence, France Goals: Develop and demonstrate the physics and technology required for a fusion power plant.

ITER overview

ITER (www.iter.org) - Superconducting Tokamak

- Single-null divertor

- Elongated, triangular plasma

- Additional heating from RF, and negative-ion neutral-beams and

500 Pfus(MW)

10 Q (Pfus/Pin)

80+ P (MW)

40-90 Paux (MW)

1.85, 0.5 ,

5.3 Bt (T)

15(17) IP (MA)

850 VP (m3)

2 a (m)

6.2 R (m)


ITER cross-section


A Q=10 scenario with (ELMy H-mode ): Ip=15MA, Paux=40MW, H98(y,2)=1 Te

Ti

ne

10nHe

Zeff

fHe

q

e

keV

%

MA

/m2

1019m

-3 m

2s-1

Project Requirement on Diagnostics is for Measurements

Diagnostic grouping A- Magnetics systems B- Neutrons systems C- Optical systems

D- Bolometry systems E- Spectroscopy systems

F- Microwave systems G- Operational systems

First real published spectrum 1814 Fraunhofer absorption lines in solar spectrum

Fraunhofer had set up optical glass manufacture, and saw solar lines as a universal standard for calibration and demonstration of quality Measured wavelengths of hundreds of lines – (how the features appear) Measured the shape of the spectrum – peaks in the green Pre-dates photography, discovery of electron He was not concerned with origin of the lines - this took another 50 years….

Common features of spectrometers

Collection: Gather, focus or collimate the radiation Selection: A component that disperses the signal into a spectrum Prism, Grating, Crystal Detection: Conversion into a useable signal Mostly electronic detectors such as CCD Rejection: Techniques to reduce noise and background Stray-light baffling Neutron shielding Data analysis


Early Spectroscope – Kirchhoff and Bunsen 1860s Exhibits most features of modern spectrometer

Low background source – Bunsen burner

Input slit and lens collimation

Dispersion – prism

Position measurement – mirror below prism

Detector – Human eye

Stray light baffling

L b–

Adjustment

11

Workshop on ITER Diagnostics –Garching, July 1989

Many plasma properties can be measured by spectroscopy

Monitoring impurities – impurities dilute the fuel and cause radiative losses Detect impurities: Characteristic wavelengths Impurity concentrations: Line intensities Measuring plasma parameters Ion temperature: Doppler broadening of lines Plasma bulk motion: Doppler shift of lines Magnetic fields: Zeeman effect Electric fields: Stark effect

13

Measurement requirements relating to spectroscopy

MEASUREMENT PARAMETER CONDITION RANGE or COVERAGE RESOLUTION ACCURACY

10. Plasma Rotation

VTOR 1-200 km/s 10 ms a/30 30 %

VPOL 1-50 km/s 10 ms a/30 30 %

12. Impurity Species Monitoring

Be, C rel. conc. 1x10-4-5x10-2 10 ms Integral 10 % (rel.)

Be, C influx 4x1016-2x1019 /s 10 ms Integral 10 % (rel.)

Cu rel. conc. 1x10-5-5x10-3 10 ms Integral 10 % (rel.)

Cu influx 4x1015-2x1018 /s 10 ms Integral 10 % (rel.)

W rel. conc. 1x10-6-5x10-4 10 ms Integral 10 % (rel.)

W influx 4x1014-2x1017 /s 10 ms Integral 10 % (rel.)

Extrinsic (Ne, Ar, Kr) rel. conc. 1x10-4-2x10-2 10 ms Integral 10 % (rel.)

Extrinsic (Ne, Ar, Kr) influx 4x1016-8x1018 /s 10 ms Integral 10 % (rel.)

28. Ion Temperature Profile

Core Ti r/a < 0.9 0.5 - 40 keV 100 ms a/10 10 %

Edge Ti r/a > 0.9 0.05 - 10 keV 100 ms 50 mm 10 %

32. Impurity Density Profile

Fractional content, Z<=10 r/a < 0.9 0.5 - 20 % 100 ms a/10 20 %

r/a > 0.9 0.5 - 20 % 100 ms 50 mm 20 %

Fractional content, Z>10 r/a < 0.9 0.01 - 0.3 % 100 ms a/10 20 %

r/a > 0.9 0.01 - 0.3 % 100 ms 50 mm 20 %


PBS System Range Function PA Status

55E4 Divertor imp monitor 200 – 1000 nm

Impurity species and influx, divertor He density, ionisation front position, Ti.

Yes PDR prep

55E2 Ha system Visible region ELMs, L/H mode indicator, nT/nD and nH/nD at edge and in divertor. Yes PDR prep

55E3 VUV spectr. – main 2.3 – 160 nm Impurity species identification. Yes PDR prep

55EG VUV spectr. – divertor 15 – 40 nm Divertor impurity influxes, particularly Tungsten Yes PDR prep

55EH VUV spectr. – edge 15 - 40 nm Edge impurity profiles Yes PDR prep

55ED X-ray spectr. – survey 0.1 – 10 nm Impurity species identification Yes PDR prep

55EI X-ray spectr. – edge 0.4 – 0.6 nm Impurity species identification, plasma rotation, Ti.

Yes PDR prep

55E5 X-ray spectr.-core 0.1 – 0.5 nm Yes Hand-over

55E7 Radial x-ray camera 1 – 200 keV MHD, Impurity influxes, Te Yes PDR prep

55EB MSE Visible region q (r), internal magnetic structure Yes Hand-over

55E1 Core CXRS Visible region Ti (r), He ash density, impurity density profile, plasma

rotation, alphas.

No CDR prep

55EC Edge CXRS Visible region Yes PDR prep

55EF BES Visible region Beam-attenuation and fluctuations. No CDR Oct 2012

55E8 NPA 0.01- 4 MeV nT/nD and nH/nD at edge and core. Fast alphas. Yes PDR closed

55EA LIF Visible Divertor neutrals No Pre- CDR held

55E Hard X-ray Monitor 100keV – 20MeV Runaway electron detection IO CDR closed


Diagnostics are highly integrated


Port Integration Activities

PCSS

ISS

PP

Rails

Bioshield Inner Outer

GRS reservation

LFS H alpha

LENPA

Cryostat

RGA

• Joint team working closely together to integrate diagnostics and services • Port Integration very important driver for diagnostics – 2014 big progress • Driver for all port systems-Interfaces


Page 17

Port Plugs: Final Design Reviews Completed • UPP Assembly: GUPP+DFW+DSM

• FDR completed

• EPP Assembly: GEPP+DFW+DSM • FDR completed

All Diagnostic Port Plugs have • Common Design and shared Procurement • This saves costs and time for the project

• Tender started • Modular structures

• Welded (Electron Beam mainly) • ESPN

Page 18

Mock-up of a large diagnostic mirror Cleaning by RF plasma discharge

(Basel University)

Sketch of the mock-up. The blue dots are removable samples

The Aluminum foil protects the samples during commissioning

phase

Mirror restoration is important for many diagnostics

Above: Measured spectrum, Fitted spectrum

Below: C5+ CX from Deuterium heating beam. Electron impact excitation of C5+ in edge plasma. C5+ CX from edge neutrals. Background lines Be+ 5271 & C2+ 5305

Charge-exchange recombination spectroscopy gives local measurement

Based on charge exchange between heating neutral beam and fully stripped impurity ions. eg D + C6+ > D+ + C5+* > C5+ + h

Line width > Ion temp. Line shift > plasma motions. Intensity > impurity concentrations

D0+C6+ -> D+ + C5+ (n = 8 -> n = 7)

Peak CX cross-section is around 60 keV – typical of heating beams on current machines.

ITER has 1 MeV heating beams, and requires a diagnostic neutral beam for CXRS


Lower view: FM M2 M4 M3 L1 L2 L3 L4

Port plug optical design

Integration of CXRS-edge

Port plug mechanical design

Fibre budle design

1 2 3 4 5

6

Ion temperature high res. in the pedestal

Plasma rotation Impurity concentrations:

Helium ash and fast alphas Low Z impurity profiles Zeff profile

Upper view optimized for r/a > 0.85 - Improved spatial resolution - Improved recess for first mirror

Motional Stark Effect

Stark effect Line Split (MSE-LS): Δλ ~ |vNB x B| Stark effect Line Polarization angle γ (MSE-LP):

View of DNB added during CDR preparation - Overall sensitivity comparable with HNB - Gives more flexibility and availability

Line-shift measurement preferred - Potentially more robust measurement - Smaller optics - Tests ongoing - eg JET

Stray light modelling

• In-vessel reflections are important for all visible systems

• LightTools is now well-established – Quantitative results – Sources built based on plasma modelling – Surface reflectivities can be modelled or measured – Directly uses CAD models of in-vessel components – Generates image-quality results – Benchmarking comparison with JET in progress

CAD model of vessel Model of SOL and divertor sources

Stray light modelling with LightTools


2.5 Mrays

-13

-12

-11

-10

-9

-8

Log

( Ill

umin

ance

[W/m

m2 ]

2.5 Mrays

-13

-12

-11

-10

-9

-8

Log

( Ill

umin

ance

[W/m

m2 ]

No reflections With Reflections

Use of LightTools for design of H-alpha system

Divertor signal

Scrape off Layer (SOL)

Signal near FW

Signal near FW =

SOL signal +

reflected signal from divertor

See next slide for detailed image of reflections on FW

Scrape off Layer (SOL) Signal near FW is required from diagnostic measurements


Strategies for stray light mitigation led to re-design of H-alpha system

Reflected Light Profile (dark spots are technical holes

with suppressed stray light level)

W/m2

EP#11 EP#12 UP#2

Image features on inner wall

Tangential view most features

highest contrast

Divertor view to quantify divertor

emission


X-ray and VUV cover a large part of the electromagnetic spectrum


X-ray VUV

Typical plasma region

Core plasma Outer plasma

Wavelength 0.1 – 10 nm 2.4 – 160 nm

Input optics Direct views Gazing incidence mirrors

Windows Beryllium windows possible Not possible

Requires vacuum extension

Dispersion Crystal Grating

Detectors CCD, Active Pixel Channel-plate, CCD

Comparison between X-ray and VUV There is an overlap in the physics and the measurement

The techniques are very different


0.01 0.1 1.0 10 100 1000

10 mm

1 mm

0.1 mm

10 um

Wavelength (nm)

Pra

ctic

al w

indo

w th

ickn

ess

Visible VUV

X-ray

Window thickness is a very strong function of wavelength For Vacuum-ultraviolet there is no practical window

Hence the term VUV, and the need for vacuum extension

1 um

Glass Beryllium

Polymer


0.01 0.1 1.0 10 100 1000

90

10

1

0,1

Wavelength (nm)

Pra

ctic

al g

razi

ng a

ngle

(deg

rees

)

Visible

VUV

X-ray

Reflective optics become increasingly difficult at shorter wavelengths


Diffraction Grating – 2d effect

Many natural phenomena, such as bird feathers, show this effect

Constructive interference occurs when the path-length is equal to multiple of a wavelength. Short wavelengths require grazing incidence – reduced aperture and difficult alignment Small groove spacing- difficult to manufacture Practical limit > ~1 nm


Constructive interference occurs when the path-length is equal to multiple of a wavelength. Works well when wavelength is similar to atomic or molecular spacing. Large apertures are possible Highly perfect crystals offer high spectral resolution Practical limit < ~5 nm

Crystal Diffraction – 3d effect

Natural Beryl

Many natural crystals are hard to improve on for x-ray spectroscopy


Impurity emission is modelled using a wide range of plasma scenarios - ADAS atomic database

- SANCO impurity transport code. - M O’Mullane


ADAS/SANCO Modelled emission of VUV spectral lines

Modelling of emission along lines of sight for imaging VUV spectrometer

VUV lines 10 – 100 nm mostly in the outer plasma

Modelled Ionization balance for impurities relevant for the Core X-ray Spectrometer


Radial X-Ray Camera Conceptual Design Review 21/02/2012 Page 35

Subsystem VUV Core survey VUV Edge imaging VUV Divertor

PBS 55.E3 55.EH 55.EG

Function Impurity species

identification. Impurity profile

Divertor impurity influxes, particularly Tungsten

Wavelength range (nm)

2.4 – 160 17 – 32 15 - 32

Resolving power (λ/δλ)

~500 ~500 ~500

Gratings 5 1 1

Implementation

Slot in Eq 11 port-plug

10 x 100 mm^2

Collimating mirrors in port-cell

Slot in Up18 port-plug

Field mirror in port-plug

Collimating mirror in port-cell

Slot in Eq 11 port-plug

Field mirror in port-plug

Collimating mirror in port-cell

ITER VUV spectroscopy subsystems

Choice of 20 nm for short-wavelength range makes the imaging mirror relatively insensitive to deposition of impurities

0 5 10 15 20 25 30 35 400

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9Mirror reflectivity at 15 incidence (Comparison of gold with thick layers of plasma impurities)

Wavelength (nm)

Ref

lect

ivity

AuBeCW

Reflectivity for VUV mirror at 15 deg grazing angle Shows sharp cut-off at critical wavelength


5-channel Main Plasma Survey Spectrometers, with shielding concept for MCNP analysis

To plasma

Collimating mirrors

Slits

Spectrometers

Detectors

Shielding


KSTAR VUV Spectrometer Test 2012 Campaign (F-Port) CR Seon, MS Cheon, S Pak & HG Lee

87 deg.

• Spectrometer table on the F-port deck

• 3 m - long Vacuum Extension Tube

• Two Gate Valves

• One Bellows

• Collimation Mirror Set

1. Cylindrical 10 cm x 5 cm, R.O.C. = 13.5 cm

2. Convex 10 cm x 5 cm, R.O.C. = 700 cm

Vacuum extension

VUV spectrometer on the optical

table

First Measurement of KSTAR Plasma Impurity

Fe XVI

Fe XVI

C III

He II

Fe XV

O VI

C IV

He II

O VI

Metal Lines

The three ITER x-ray spectrometer subsystems

Subsystem Main plasma x-ray survey X-ray Core imaging X-ray Edge imaging

Function Impurity species identification and

monitoring

Core ion temperature, rotation, impurity profile

Edge ion temperature, poloidal rotation, impurity profile

Wavelength range (nm) 0.05 – 10 0.2 – 0.4 0.2 – 0.5

Resolving power (λ/δλ)

Below 2.5 nm ~1000, Above 2.5 nm ~ 100 ~8000 ~8000

Implementation

Slot in E11 port-plug, Diffracting optic in port-cell

Slot(s) in E09 port-plug, Diffracting optics inside port-plug

Slot in U09 port-plug, Diffracting optic behind the port-plug

XRCS Edge (IN-DA)

XRCS Core (US DA)

XRCS Survey (IN-DA)

JET vacuum beam-line has features required for ITER

Valve on RH flange Thick hydro-formed bellows

ITER equivalent: Port-cell Interspace Port-flange

Shielding supported independently of sight-tube

ws

e


Long-wavelength channels Short-wavelength channels

To neutron dump Optional beryllium window Optional Mylar window

From plasma

ITER Core X-ray Survey Spectrometer – see S Varshney

ArXVII spectrum from NSTX – Manfred Bitter

3000

2000

1000

03.94 3.95 3.96 3.97 3.98 3.99 4.00

Phot

on C

ount

s / C

hann

el

Wavelength (Å)

(c) w

x y q r a k

n >

3 sa

tellit

es z

j

n

High-resolution x-ray spectroscopy

Ti: Doppler broadening

Vtor/pol: Doppler shift

Te Dielectronic satellite ratio

ne Forbidden line ratio z/(x+y) (sometimes)

Zeff Continuum

imp Impurity injection

nimp Absolute calibration


High resolution imaging crystal spectrometers Recent advances in active pixel detectors such as Pilatus and Medipix have enabled

a new generation of imaging crystal spectrometer. The technique has moved quickly from demonstration, to routine production of a wide

range of new physics results The ITER design has been based on this principle since 2003

The ITER design has been based on this pri

The astigmatism of off-axis spherical crystal allows two different foci Image on 2-d detector contains - wavelength in dispersion plane

- Spatial profile perp. to dispersion

plane

Fast 2-d active pixel detectors enable imaging crystal spectrometer

Bottom

Core

Top

Crystal

Detector

C-Mod Plasma (Height =72 cm)

Advances in detector technology enable new measurement capability CERN-led Medipix 3 – in development

Active pixel detector - Each pixel has analog pulse processing, thresholds, and digital counter - 256 x 256 array. Pixels 55 um square - Multiple enrgy windows - 1 us pulse-process time per pixel - Radiation-tolerant to ~1014 neutron/cm2 Diagnostic applications - X-ray spectroscopy and imaging - Particle detection and spectroscopy - Fast visible and VUV framing (with MCP) - Neutron and gamma spectroscopy

27 September 2004 Michael Campbell

MEDIPIX2 Hybrid Pixel DetectorMEDIPIX2 Hybrid Pixel Detector

Detector and electronics readout are optimized separately

27 September 2004 Michael Campbell

Charge sensitive preamplifier with individual leakage current compensation2 discriminators with globally adjustable threshold3-bit local fine tuning of the threshold per discriminator1 test and1 mask bitExternal shutter activates the counter13-bit pseudo-random counter1 Overflow bit

Medipix2 Cell SchematicMedipix2 Cell Schematic

Preamp

Disc1

Disc2

Double Disc logic

Vth Low

Vth High

13 bits

Shift Register

Input

Ctest

Testbit

Test Input

Maskbit

Maskbit

3 bits threshold

3 bits threshold

Shutter

Mux

Mux

ClockOut

Previous Pixel

Next Pixel

Conf

8 bits configuration

Polarity

Analog Digital

Preamp

Disc1

Disc2

Double Disc logic

Vth Low

Vth High

13 bits

Shift Register

Input

Ctest

Testbit

Test Input

Maskbit

Maskbit

3 bits threshold

3 bits threshold

Shutter

Mux

Mux

ClockOut

Previous Pixel

Next Pixel

Conf

8 bits configuration

Polarity

Analog Digital

Lower Hybrid Wave Induced Rotation on Alcator C-Mod Measured by imaging crystal spectrometer (Ken Hill et al)

New measurement capability for non-NBI discharges

ITER Core Imaging X-ray Spectrometer

The views projected onto flux surfaces

Three sub-views with imaging crystal spectrometers Toroidal component ~25 deg.

Upper view from within port plug Inner and mid views from behind port plug

Close collimation inside port-plug keeps direct neutrons away from sight-tube components

Core imaging x-ray spectrometer

Under study - Direct neutrons closely collimated in sight-tube - Minimize sight-tube activation - Maximum use of low-activation components – eg Aluminium - Beam-dump to stop direct neutrons

27th Diagnostics ITPA, 3-7 November 2014, R Barnsley

Total core radiated power is around 50 MW – mostly x-rays

Normalized radiated power profiles of individual impurities

X-ray profile resolved into 5% energy bands

Diagnostic first wall DFW

Diagnostic shield module DSM

Port-plug Port-plug rear flange

Secondary vacuum tube

In-port detectors

Ex-port detectors

ITER Radial X-ray Camera


Separate slots for each camera module – big improvement in neutronics - practical DFW slots

Radial X-Ray Camera Conceptual Design Review 21/02/2012 Page 54

X-ray Camera - splitting fan view into several sub-views Results in improvement in neutron flux at port flange

No loss of x-ray sensitivity

1.72×108

n/cm2/s 1.3×1010 n/cm2/s

Improved Model Initial Neutronic Model

Present Global planning for ITER Diagnostics

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 ITER Construction

2021 2022

ITER Operations

2023

Concept

Complete Designs

Manufacture

Integration

Commissioning

Ready for Testing/Operation

Assembly and Installation


Summary - Spectroscopy is a very powerful technique, used

in some form by almost all ITER diagnostics

- Spectroscopic diagnostics are monitoring impurities in the main plasma, throughout the spectral range

- The systems are nearly all in procurement

- Some, such as VUV are in prototype phase

Gamma-rays and neutrons

- Nuclear reactions among fuel and light impurity nuclei - Optics large rather than complex – slits, slots and shielding - Neutron measurements for total power and reaction profile - Gammas for high energy particles – alphas, non-thermal ions

Neutron and -cameras for ITER

Radial camera - 20 Views total

- 12 ex-vessel

- 8 in-vessel – dictated by narrow port

Vertical camera - Required to detect in-out asymmetry

- Difficult to integrate

- Divertor location favoured

Instrumentation - Counters and spectrometers

- Fission chambers for neutrons

- Scintillators for gammas and neutrons

- Natural and CVD diamonds

Leicester University Skylark sounding rocket spectrometer ~1975 3-channel gridded-aperture collimators.

~5 minutes observation of solar coronal x-ray spectra


Bragg rotor X-ray survey spectrometer for JET Installed ~1993 – still operational

6-crystal rotor

4-crystal “paddle”

Anti-scatter collimators

Position encoders

Drive motors Gas proportional counters with 1.5 um polymer windows

Filters


Typical X-ray survey spectrum of JET

The X-ray spectrum is relatively uncrowded compared with VUV


"Light-bucket" detector Position sensitive detector

Collimator Scanning flat crystal Curved crystalSlit

Broadband crystal spectrometers

(Left) Scanning monochromator, with input collimator, flat crystal and “light-bucket” detector. JET .Design (Right) Polychromator, with input slit, fixed crystal and position-sensitive detector. ITER design – made possible mainly by modern detecotors


Documents

ITER Spectroscopic diagnostics D-T fusion requires lowest