Absorber Introduction & Overview Jim Hylen LBNE Hadron Absorber - - PowerPoint PPT Presentation

Absorber Introduction & Overview

Jim Hylen LBNE Hadron Absorber Core Advanced Conceptual Design Review 20 January 2015

Outline

• Summary of Requirements
• Brief description of absorber design

– Details of radiation & energy deposition are in Nikolai’s talk – Details of FEA are in Brian’s talk – Details of mechanical design/remote handling are in Vladimir’s talk

• High level strategy
• Beam accident assumptions & protection systems
• Suitability of Aluminum as choice for the core material

1/20/2015 Jim Hylen | Absorber Introduction & Overview 2

Requirements summary (see requirements document LBNE-doc-10148

• n review web site for details)

Physics requirement:

• Stop leftover beam particles (absorb protons that miss the target or get scattered

from it as well as hadrons, electrons, and photons generated in interactions of the primary beam with the target). Well defined end of the space where mesons decay to neutrinos.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 3

Proton beam hits target, producing shower of mesons Fraction of mesons decay to neutrinos (and muons) Hadron monitor detects charged particles (protons, mesons, muons) Absorber stops leftover protons, mesons and a fraction of the muons Monitor of muons penetrating absorber

Absorber hall & support rooms

1/20/2015 Jim Hylen | Absorber Introduction & Overview 4

Decay pipe to go here Absorber to go here RAW room Muon monitoring to go here

Absorber hall layout for one level, showing support room detail (can zoom in for dimensions etc.)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 5

Beam

Water-cooled Hadron monitor steel core

Spoiler Water-cooled Al Mask blocks Aluminum core, 1st 9 blocks are sculpted

Overall absorber:

• Poured concrete volume:

24,000 ft3

• Steel shielding:

2,500 ton

• Aluminum: 39 ton

“Absorber core”:

• Spoiler block
• 5 Aluminum mask blocks
• 9 Sculpted Al blocks
• 4 Solid Al blocks
• 4 Central steel blocks

which share the features:

• Water cooled
• Individually hung on

removable modules

• Each 1 foot thick

6

Hadron monitor Insertion/extraction tower

4 m diameter Decay pipe

MARS Energy Deposition & Radiological, N. Mokhov

Need to accommodate hadron monitor, used to check beam alignment

Need to accommodate muon monitor(s), used to monitor beam stability

muons are useful for monitoring neutrino beam, as they come from

the same meson decays that produce the neutrinos Spaces for muon monitors after absorber,

used by near detector group

1/20/2015 Jim Hylen | Absorber Introduction & Overview 7

Absorber protection system would use a small chamber in one of these slots

Operational & Radiological Requirements

• Operate with minimal maintenance during the lifetime of the experiment – lifetime
• f Beamline Facility assumed 30 years with 20 years of actual beam operation
• Support beam power up to 2.4 MW in a proton energy range from 60 to 120 GeV
• Sustain beam-energy deposition under expected normal operation conditions as

well as handle beam accident situations that may occur with some reasonable probability – Under Normal Operation, up to 30% of beam energy ends up in absorber – Set requirement to take 2 consecutive accident pulses of 100% beam power

• Are including monitoring to limit accident to one 100% pulse
• Provide radiation protection to people and ground-water

– Keep below the allowable limits:

• prompt radiation in beam-on accessible areas
• residual radiation in areas that will be accessed for maintenance/repair
• radio-activated air releases
• activation of subsurface soil and groundwater
• Serviceability – have the ability to replace core blocks in case of unforeseen

failures

1/20/2015 Jim Hylen | Absorber Introduction & Overview 8

Core serviceability

9

Cut-away to show gun drilled water cooling loop

Each core block is removable and replaceable

Images from: Vladimir Sidorov

Steel Al

Storage of dead radio-activated equipment

1/20/2015 Jim Hylen | Absorber Introduction & Overview

State of design

• The core is the most challenging part of absorber design, and is where our effort

has been concentrated. Its advanced conceptual design is the focus of this

• review. We believe that the analysis that will be presented shows this to be a very

viable design, and is ready to proceed to detailed mechanical design.

• Some effort has also been directed at civil construction requirements, to allow cost

estimates, and radiological issues that drive the civil construction have been seriously investigated. Some of that will be presented, although not the focus of this review.

• The designs of the outer part of the absorber, the water systems, and the

monitoring systems are at a conceptual level; real work on those systems will not start until after the core design passes review. They will then be subject to their

• wn reviews after design and optimization efforts occur.
• However, the concepts for the absorber protection monitoring follow fairly

straightforwardly from NuMI experience, and will be shown in this talk.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 10

Conditions used as input to FEA

LBNE will have two phases, 1.2 MW beam, then upgrade to 2.4 MW beam; but because of difficulty to upgrade absorber, design absorber from start for 2.4 MW

There will be different configurations of targets and horns in LBNE over its lifetime, in particular:

– The current baseline target/horn (using NuMI target/horn design) is financially driven; we know we can improve the beam by re-optimizing components – The current target/horn will only take 1.2 MW; we know we will re-design these components for 2.4 MW

Want to leave flexibility to accommodate those future configurations. Thus we have designed to meet “worst case” scenarios:

– Have assumed shortest reasonable target (two interaction lengths)

• Giving worst non-interacting proton spot at absorber for normal operation

– Have assumed beam can totally miss target, hitting absorber directly

• Giving worst accident condition at absorber, although current baffle/target design

does not allow such a severe condition

– Have assumed tightest focus of beam at absorber (optics + beam-spot-at- target) for accident condition

1/20/2015 Jim Hylen | Absorber Introduction & Overview 11

Beam parameters (see LBNE Beamline Operating Parameters LBNE-doc-10095

• n review web site for details on phases and energy dependence)
• If not noted, results presented for 120 GeV proton beam (worse than 60 GeV).

Will also present results that show design is OK for 60 GeV proton beam. (In fact, absorber could take MUCH higher beam power at 60 GeV).

• Beam parameters for various studies were picked for worst case upgrades:

– 2.4 MW beam is 1.5e14 POT/spill, 1.2 second repetition at 120 GeV – Beam spot size at target is 1.7 mm RMS for 1.2 MW design, may be up to 3 mm for 2.4 MW designs; used 2.4 mm for accident study since with our optics this gives smallest spot at absorber (7 mm RMS) for beam accident condition – Baffle hole diameter: 17 mm at 1.2 MW, but may be up to 30 mm for 2.4 MW, so have used 30 mm for range of possible beam accidents – Normal operation modeled with fin-type two-interaction-length target where:

• 0.3% of beam misses target entirely, giving 9 mm central spot on absorber (7 kW)
• 13% of beam protons multiple scatter through target, giving 5 cm central spot on

absorber (~300 kW)

• The 2.4 MW target will probably include spoiler feature (described later for 1.2 MW

target) and be longer, so these beam assumptions are conservative

Designing to above parameters leaves significant flexibility for future configurations

1/20/2015 Jim Hylen | Absorber Introduction & Overview 12

Using experience with NuMI

• LBNE core (gun-drilled water lines in aluminum blocks) based on NuMI

absorber, which has been operating for a decade without any problems.

• NuMI upstream decay pipe window is same material (Al 6061-T6) as

absorber core, has seen same integrated proton-flux-density as LBNE absorber spoiler will, and has no problems so far.

• Remote handling of core blocks based on T-block hanging system used

successfully for last decade in NuMI target hall to swap horns and targets.

• Beam Position Monitor (BPM) input to NuMI Beam Permit System,

checking that beam is directed toward the target, has been used for last decade.

• Thermocouples in absorber core watching for overheating have been an

input to NuMI Beam Permit System for last decade.

• Muon monitor has been used at NuMI, but not as input to Beam Permit

System; some modifications/upgrades required to make this operate smoothly and reliably, and will be discussed.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 13

NORMAL OPERATION

1/20/2015 Jim Hylen | Absorber Introduction & Overview 14

Strategy part 1 – keep temperature low

At the previous core review (Nov. 2013), design had peak temperature in Aluminum core of nearly 170 C. Concern was expressed that Aluminum can fatigue, creep, lose temper over time at high temperature, and a more conservative design was desired.

LBNE absorber design features to keep temperature low:

• Spoiler layer initiates particle shower, then air space between it and the rest of the core

blocks lets shower spread out, reducing energy density before further showering

• Sculpted core blocks (thin at center, thicker at water line)

– Provide increased heat transfer conductance to water lines per amount of material at the center – Give more air space for shower to spread out (Aluminum dT < 1 C per pulse)

• Water lines are drilled near center of core blocks, reducing transport dT (major contribution)
• Multiple water lines are drilled in each block (also provides redundancy, could run 1 line off)
• Water flows at high enough velocity (2.5 m/s) to give good heat transfer coeff. (dT < 15 C)
• Water flows at high rate (80 gpm /block), gives low water dT in block ( dT max. = 2 C )
• Low temperature water supply (10 C) (will dehumidify air to keep dew point below this)

The combination reduces aluminum maximum normal operating temperature to 88 C with the worst target configuration we imagine using. We believe this is a robust design.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 15

ACCIDENTS

1/20/2015 Jim Hylen | Absorber Introduction & Overview 16

Strategy part 2 – beam accidents

The absorber continually takes ~ 30% of the beam power. In a beam accident condition, this could go to 100%, and the beam can be concentrated in a smaller spot

• n the absorber if beam does not hit the target.

LBNE design features to limit beam accidents:

• Target baffle limits probability and area of possible beam accident
• Have multiple systems to pull beam permit in non-normal conditions
• Target with wide upstream “spoiler” fins, to cover baffle hole, spread spot

– Although we did not take credit for the spoiler fins in the accident FEA Two worst case beam accidents, where FEA will be presented:

– Full beam hits center of absorber, where core is already at maximum temperature from normal operation – Full beam hits absorber where gun-drilled water lines are

Two other scenarios

– Beam not exactly centered on target for long term running

• (1/4 mm offset -> 0.3% of beam that misses target -> 0.33%, acceptable)

– Target gradually deteriorates, letting more protons through

• Thermocouples in absorber core can catch a 10% increase (NuMI experience)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 17

How would an accident pulse reach the absorber? How probable is it?

Baffle hole 17 mm diameter Beam profile, 1 sigma, 2 sigma Target 10 mm wide Geometry of 1.2 MW target and baffle

Beam would have to thread the hole between target and baffle Or target would have to move (limited if in horn)

12/19/2011 18

• Tgt. Vert. Pos. Therm. / Jim Hylen

NuMI experience: (beam, target, baffle hole were all x0.7 of size of this design) Two or three accidents hit baffle, none into the baffle hole. One target bent, allowing 20% more power to absorber. One target gradually deteriorated, 10% more power to absorber Guesstimate: probably zero, maybe 1, accident threading hole during LBNE lifetime. Gradual target deterioration likely.

Base design target now puts some material in hole between target and baffle (but not credited in FEA)

Target based on NuMI Low-Energy-Target (NT) for LBNE at 1.2 MW (LT1.2)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 19

1st few fins are larger diameter (26 mm) than baffle hole (17 mm)

54 micro-radian multiple scattering for 120 GeV protons, 12 mm spot size at absorber

This doubles the scattering for worst-case accident of beam missing fin, reducing central proton flux at absorber

• ver cases presented

(details in Nikolai’s talk) Increasing size of 1st few fins will not affect pion production

Target and baffle

Why not just make the target larger or the baffle hole smaller?

• Protons hitting baffle can cause systematic error in neutrino flux since

resulting pions focus differently through the horns

• Increasing target width further down the target can cause some

reduction in neutrino flux, by absorbing pions trying to get out of the sides of the target But future target designs (like beryllium ball design) may naturally put more material in that space.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 20

Beam Accident condition target-system baffle prevents beam

hitting absorber outside aluminum core

1/20/2015 Jim Hylen | Absorber Introduction & Overview 21

Absorber

Core blocks are 152 cm x 152 cm, accident pulses at 44 cm radius are well away from edges of Al blocks

Last /H trim; 3 inch aperture Midpoint of baffle; 30 mm aperture max. for 2.4 MW beam

• Ref. LBNE-doc-7317

Last V/H trim; 3 inch aperture

Accident pulse input

For FEA, two conditions

• Full pulse hits water line directly
• Full pulse hits center of core after core gets to

2.4 MW normal operating temperature Beam permit system should stop beam after 1 bad pulse, but have set a requirement to survive a 2- pulse accident

1/20/2015 Jim Hylen | Absorber Introduction & Overview 22

Three independent systems provide redundancy to pull beam permit quickly in non-normal condition

• Beam position monitors upstream of

target – Pull beam permit after 1 beam spill if proton trajectory is off, missing target

• Muon monitor after absorber combined with Toroid proton monitor before target

– Can pull beam permit after 1 beam spill, if muon response is not proportional to number of protons in spill

• Thermocouples in absorber core

– with thermocouple on-axis in shower, provides fast response (detect Energy deposition in thermocouple itself)

Hardware Safety Systems

1/20/2015 Jim Hylen | Absorber Introduction & Overview 23

Beam Position Monitor system

NuMI beam permit system is operating with +/- 1.5 mm BPM window (pulls permit if proton beam center is more than 1.5 mm off nominal trajectory) Permit is pulled before the next pulse. If LBNE uses the same tolerance, pulse at the edge of this allowed window would send somewhat more beam tail (2% of beam) to absorber NuMI has had 2 horizontal BPMs and 2 vertical BPMs in final straight section; lesson learned from NuMI is that one of these can drift, and Autotune system will then re-direct beam to follow new trajectory. Require LBNE be protected against this, so that one bad BPM cannot compromise the system; redundant BPMs are a simple and cheap solution.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 24

Using muon monitor in beam permit

The general points of operation can be demonstrated with NuMI data

• Note, cannot prevent 1st accident pulse, but can stop beam after 1 pulse.
• For each pulse, measurement is made of:

– Protons down primary beam line (using beam toroid monitor) – Muon flux after absorber (using small chamber centered on beamline)

• If response of these two devices are not proportional, pull beam permit for

subsequent pulses The concept is that if the protons did not hit the target, they will not produce mesons that can decay in the decay pipe to produce muons that penetrate the absorber.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 25

NuMI data at ~50% of normal NuMI beam power, showing drop in muon monitor response when beam partially misses target

1/20/2015 Jim Hylen | Absorber Introduction & Overview 26

Demonstration of drop in NuMI muon monitor response when beam misses target, with beam ~ 2% of full power

1/20/2015 Jim Hylen | Absorber Introduction & Overview 27

Beam hits baffle instead, baffle acts as target

Still plenty

• f statistics

per pulse in single pad of NuMI muon monitor

Majority of beam threading hole between target and baffle

Correlation between toroid POT measurement and muon monitor response, (few weeks of NuMI data)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 28

Mis-associated spills, ACNET lumberjack not good enough! Beam Permit System will have to properly time/associate the readouts Possible cut band, pull beam permit if muons disagree by more than ~10% of full power beam (would work even with NuMI gas flow chambers; can be tighter using other technology) POT/spill (e12, from toroid) Muon Monitor Response

Pull permit OK Pull permit

Improvements desired compared to NuMI for Beam Permit

• n Muon monitor response
• NuMI Muon monitor response frequently jumps for some hours by ~ 5% when the

gas supply bottles are changed; also response varies with atmospheric pressure; don’t want these variations which are features of a flowing gas system – LBNE beam permit muon monitor should ideally be sealed or vacuum system, such as sealed gas ionization tubes or Silicon/Diamond detector or SEM. Plan to test a new chamber in NuMI muon alcove (next slide) – Slow drifts (ageing) of chambers can be compensated by changing permit

• limits. Muon monitors are in negligible residual radiation areas, and can be

replaced easily if they drift too far. LBNE permit monitor only needs to be < 1% the size of the NuMI muon monitors – not very expensive. Thus requirements for chamber are actually not especially challenging.

• Readout of NuMI front ends to ACNET can get stale data, making POT to Muon

comparison invalid; for LBNE the front ends providing data to the beam permit system will need features that clear stale data and provide heart-beats, providing reliable data

• At least two channels should be provided to give redundancy.
• Need controls experts to design permit for high reliability

1/20/2015 Jim Hylen | Absorber Introduction & Overview 29

Absorber protection muon monitor technology

• Most appealing candidate is a sealed ionization chamber, and would like to

assess (and test at NuMI) the model that was used for CNGS muon monitoring. As LBNF forms, this is definitely an area to ask for help from CERN.

• Believe that a specially built SEM would work, but development costs and need

for vacuum utility make this less desirable than the sealed ionization chamber. This technology provides an alternate solution if the CERN devices would be non- linear (saturate) at our intensities.

• May also explore silicon or diamond detectors. Matevz Cerv recently came from

CERN and installed diamond detectors in NuMI muon alcoves. A problem with the polycrystalline diamond detectors we have is that if beam is off for some amount

• f time the detectors take some time after beam returns to stabilize. This makes

them unsuitable for protection monitors. Do not know if there is a solution to this (single crystal diamond? Silicon? A separate radiation source? …)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 30

Thermocouple system

• Have not studied optimum deployment of thermocouples, but think they would be

in either solid Aluminum block 2 or 3 or 4 (or maybe all three) where: – Shower has spread out, so an array does not need to be as densely instrumented – Still quite large signal (temperature jumps), but not in the most stressful location – Geometry of blocks are suitable for replaceable thermocouples (sculpted blocks are not suitable); Vladimir will show a mechanical mounting concept

• Also, have not optimized array density, although will show a minimal concept.
• Although we have significant experience with thermocouples (such as at shower

maximum in the NuMI absorber, in the target pile baffles, and recently in the target position monitor) doing some prototyping in the NuMI target pile would be a very useful exercise to demonstrate operation in the predicted environment and fine- tune the design.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 31

Thermocouple system input to beam permit

Can run in One mode, or Two modes of operation running simultaneously

• Absolute temperature limit (simple and robust)

– Pull beam permit if ~ 5C above normal operating temperature – Protects against slow deterioration of target, or extra beam tails, or beam partially missing target continuously – This limit would also pull permit after 1 to 5 full accident pulses, depending on how close to normal operating temperature the core is when accident pulses happen, and where pulses are directed – Also pull permit if accident pulse breaks the thermocouple, yielding non- physical read-back

• Compare temperature sample before/after pulse (Optional mode)

– Pull permit if dT > 1 C

• (dT=0.6 C normal, 20 C full accident head on)

– Will pull beam permit in 1 accident pulse Note in both cases, thermocouples are in the shower; there is no delay time for heat to transmit through aluminum, e.g. to edge of block; the thermocouple material itself heats up, although it also rapidly equilibrates to surrounding material

1/20/2015 Jim Hylen | Absorber Introduction & Overview 32

Thermocouple system input dT per pulse, on-axis 100% accident

dT (C) Upstream dT (C) Downstream Full Al Block 2 22.8 20.4 Full Al Block 4 10.0 5.6 Center of Full block 2 Temperature plot for accident pulses

Multiply scale by 1.2 to get kJ/cm3 per pulse

Sample thermocouple layout

Wires to connection on top

• f block

Edge of Al block 44 cm radius of possible beam accidents Thermocouples

Example of an array that if installed on solid AL block 2 would pull the permit in a single pulse for a half-intensity accident pulse anywhere on reachable face of absorber

1/20/2015 Jim Hylen | Absorber Introduction & Overview 34

Comment on extent of Thermocouple system

• The central thermocouples are all that are needed if one is willing to rely on the

(already redundant) muon and BPM systems to catch off-trajectory pulses. So it is a judgment call as to whether it is worth deploying the extra thermocouples, and also the optional before/after pulse comparison electronics. Can make that decision during preliminary design studies.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 35

Other hardware protection systems

• LBNE Absorber WHAT IF failure mode analysis LBNE-doc-10189

(on the review web site) contains information about mitigations of risk, impact of and recovery from various failure modes

– Think they are fairly straightforward – Don’t plan to discuss unless there are questions from the committee

1/20/2015 Jim Hylen | Absorber Introduction & Overview 36

Core Material Suitability

1/20/2015 Jim Hylen | Absorber Introduction & Overview 37

Desirable material properties for core

• Good resistance to thermal impulse (combination of heat capacity, thermal

expansion coefficient, modulus of elasticity and yield strength, ductility)

• Low density (to spread out shower, decrease energy deposition density)
• High thermal conductivity (to conduct heat to water cooling lines)
• High radiation damage resistance
• High corrosion resistance
• Tolerant of high temperatures
• Low toxicity (avoid mixed waste if possible)
• Good creep and fatigue resistance
• Reasonable expectation that it would last the life of the facility
• Conducive to modular design, to facilitate replacement of failed section if

necessary

• Low cost

38 1/20/2015 Jim Hylen | Absorber Introduction & Overview

Core Material – Aluminum 6061-T6

Aluminum

• Forms a protective solid oxide layer as it oxidizes, preventing further oxidation;

(will discuss NuMI experience with Aluminum window) can be in air

• Allows for simple modular construction and the possibility for replacement – no

need for encapsulation

• Gun-drilled water lines provide simple connection of cooling water to core
• Has excellent thermal conductivity
• Is very radiation resistant, fine at expected 1 DPA
• Is not toxic; no mixed waste, non-toxic if somehow becomes particulate
• Reasonable cost
• Has worked well for NuMI (we have experience, absorber and DK window)

Issues for Aluminum

• Need lower temperature compared to other materials (avoid creep and fatigue)

– Addressed by spreading out shower (spoiler + sculpting)

• Less resistance to thermal impulse than some other materials

– However, more ductile than Be or graphite, can plastically deform in accident and still function perfectly OK; have checked accident conditions

39

Slide from: Jim Hylen 1/20/2015

Jim Hylen | Absorber Introduction & Overview

Al 6061-T6 experience: NuMI Decay Pipe upstream window

• Photo 2007 (after 2 yrs operation) Photo 2012 (after 7 yrs operation)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 40

Air side of window shown (other side is helium) (Note lighting was different). Central ~1cm radius spot where beam hits window appears to build up an Aluminum Oxide layer; no further deterioration noted in later photo.

Al 6061-T6 experience: NuMI Decay Pipe upstream window

How does the NuMI decay pipe window compare to what the LBNE absorber spoiler will see? Left-over protons are multiple-scattered by target (4-radiation-lengths -> 0.24 milli-radians). The integrated proton flux thus scales by N_POT / L^2.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 41

NuMI to date LBNE design lifetime units Ratio Integrated Protons-on-target 2e21 39e21 POT 20 Distance L, target to (NuMI window, LBNE spoiler) 45 220 m 4.9 L^2 m^2 24 Flux ratio N_POT / L^2 LBNE spoiler / NUMI window 0.8 NuMI window has accumulated more integrated flux than spoiler will see ! NuMI window is 1/16” thick, is continuously under 0.1 bar pressure differential, and is still helium tight (target pile air upstream, decay pipe helium downstream)

Parameter List (page 1) (see Absorber Parameters LBNE-doc-10095

• n review web site for more details)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 42

Total beam power in absorber 742 kW For 2,400 kW beam Beam power in absorber core 512 kW 120 GeV, Normal ops. Beam energy deposited in core 2,246 kJ / pulse Full beam accident Decay pipe diameter / length 4 m / 204 m Helium filled Distance target to absorber 220 m Absorber size, excluding concrete 6.8m W x 6.4m H x 8.6m L Core blocks: 1.5m W x 1.5m H x 0.3m L (mask 2m W x 2m H) Spoiler 1 block, Al 6061-T6 Solid Mask 5 blocks, Al 6061-T6 Air holes in center Sculpted 9 blocks, Al 6061-T6 Thinner at center Solid Al 4 blocks, Al 6061-T6 Solid Fe 4 blocks, Carbon steel Nominal gap between core blocks 5 mm

Parameter List (page 2)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 43

Total water flow for core cooling 1760 gpm For 2.4 MW beam, incl. filt. Volume of water for core cooling system 1810 gallons Includes skid, etc. Water lines per block (Al / Fe) 4 / 2 X ( 19 / 4 ) blocks Flow rate per water line 20 gpm Water line diameter 1 inch Water flow velocity 2.5 m/s Water supply temperature 10 C Core blocks: Max. T normal / accident Accident is 2 100% pulses Spoiler 66 C / 146 C Mask 25 C Sculpted 88 C / 171 C / 140 C Normal / Off axis / On axis Solid Al 84 C / 121 C Solid Fe 235 C / 252 C

Backup

BACKUP

1/20/2015 Jim Hylen | Absorber Introduction & Overview 44

Backup for thermocouple system estimates

dT (C) Upstream dT (C) Downstream Center Full Al Block 2 0.63 0.58

dT per pulse, normal operation

Backup for thermocouple system estimates Solid Al On-Axis: Contour plots - temperature Steady state After 2 Accident Pulses Solid Aluminum block 2

• Maximum temperature of 121C after two pulses.

Thermocouple response time – from heat deposited

• utside, not internal (which should be even faster)

1/20/2015 Jim Hylen | Absorber Introduction & Overview 47

For ungrounded 1/16” diam. thermocouple (type we have used in NuMI) “time constant” response to dT in surrounding material is 0.3 seconds.

From Omega.com

Material selection – look at low density materials

Beryllium has the best material properties, and would be the ideal selection except for:

• Toxicity

– After irradiation, would be mixed waste, very hard to dispose of – Although benign as a solid, if some of it does become particulate and spread, it is a serious health hazard; so used only where necessary

• Cost (much higher than other options)

Albemet has not quite as good properties, with similar downsides

48

Slide from: Jim Hylen 1/20/2015

Jim Hylen | Absorber Introduction & Overview

Material selection – look at low density materials

Graphite, advantages:

• Has good resistance to thermal impulse (combination of heat capacity, thermal

expansion coefficient, modulus of elasticity and yield strength)

• Reasonable thermal conductivity (to conduct heat to water cooling lines)
• Tolerant of high temperatures, if in inert atmosphere
• Low toxicity
• Good creep and fatigue resistance

49

Slide from: Jim Hylen 1/20/2015

Jim Hylen | Absorber Introduction & Overview

Material selection – look at low density materials

Graphite, disadvantages (relative to aluminum):

• Oxidizes at room temperature in charged-particle radiation, hence may have to be

encapsulated and covered in an inert atmosphere (room temperature oxidation of graphite using radiation is in fact used in commercial applications)

– Note graphite can oxidize to a gas, which is lost, as contrasted to Aluminum, which

• xidizes to a solid film that protects the Aluminum core from further oxidation

– In particular, encapsulation complicates modularization, and modularization is desired to be able to replace any failed part of the core

• Suffers radiation damage at much lower dose than Aluminum

– LBNE dose in absorber core ~1 DPA, where graphite material properties change significantly with charged radiation (Fermilab-Conf-12-639-AD-APC December 2012, LBNE-doc-6659)

• Cooling system would be more complicated

– Aluminum gun-drilled water lines are simple, provide good thermal heat transfer – Graphite would need exterior cooling plates forced into good thermal contact

• Graphite is less consistent batch-to-batch than Aluminum

We could likely make graphite work, but a significant R&D effort would be required to qualify graphite for LBNE absorber core and geometry might be complicated.

50

Slide from: Jim Hylen 1/20/2015

Jim Hylen | Absorber Introduction & Overview

Ozonolysis

For oxidation of graphite at room temperature by radiation:

• Ozonolysis. The key steps, in what is known as the Criegee[21] mechanism, have

been confirmed by Berger et al.[22]

• [21] Criegee R. Mechanism of ozonolysis. Angew Chem 1975;87:765-71.
• [22] Geletneky C, Berger S. The mechanism of ozonolysis revisited by 17O-NMR
• spectroscopy. Eur J Org Chem 1998;1625-7

Am not a materials scientist, and have not tried scoping calculations to extrapolate this to a graphite core, since we are concentrating on the Aluminum design. Perhaps NuMI experience with graphite deterioration in targets also biases us against using graphite in the absorber core. Believe 2nd target probably deteriorated due to gas contamination, even though it was below the nominal 400 C threshold for

• xidation in non-radiation environment.

NuMI Decay pipe window oxidation at beam spot is another example of what is called radiation accelerated corrosion.

1/20/2015 Jim Hylen | Absorber Introduction & Overview 51