Beamline-to-MICE Matching Ulisse Bravar University of Oxford 2 August 2004 MICE performance with ideal Gaussian beam JUNE04 beam from ISIS beamline (Kevin.

Beamline-to-MICE Matching

Ulisse Bravar University of Oxford

2 August 2004 • MICE performance with ideal Gaussian beam• JUNE04 beam from ISIS beamline (Kevin Tilley)• Software, beam momentum and coil currents

manipulation • Results, cooling of the actual ISIS beam

Cooling a Gaussian beam

• Beam in the middle of the upstream spectrometer: = 6 mm rad<p> = 200 MeV/c = 33 cm

• Gaussian distributions: x = 3.3 cm

px = 20 MeV/c

• Coils from Mice Note 49:about 15% cooling

LH

spectrometersz (m)

(

m r

ad)

ISIS beam• Simulations of the MICE channel performed with ICOOL• Beamline designed by Kevin Tilley with TURTLE

• Latest beam design: JUNE04 • Pb diffuser is 8 mm thick

• Turtle-Icool interface planes: a) Middle of upstream spectrometerb) After Pb diffuser, prior to solenoidc) After Q9, prior to Pb and solenoid

• Work in progress:a) Interface Turtle-Icool midway between Q6-Q7 b) Simulate the entire beamline with Icool

Beamline layout

• MICE beamline (NOT TO SCALE !!!)

Q7 Q8 Q9

Quadrupole triplet

Drift

Pb diffuser 0.8 cm

Solenoid & Spectrometer

LHFocus coil

TURTLE – ICOOL INTERFACES

Beam design concept (1)MICE wants……

x/y

x’/y’

2. Particular input emittance RQD

qB

p

x

xxx

20

'max

max

1. Matched beam

…. at matching point (4T Spec Solenoid)

Beam design concept (2)Scheme to provide simultaneously:-

This is the driving Design Concept in this design work:

To use ‘Beamsize’ & ‘Scatterer thickness’ to provide both beam matching, & required emittance generation.

200 2

1 RQD

0

00

0

00

1. Focus Beam with

x/y

x’/y’

MICE ACCEPTANCE A

0, 00 beamR

2. &

Matched after passing thru’ required scatterer

RQD 0, match

2

[Above figure illustrates case match region immediately follows scatterer]

JUNE04 beam

• Beam is not cylindrically symmetric at Q9, far from Gaussian…

• Designed to achieve = 33 cm and = 0 in spectrometer solenoid

• Central momentum of ‘useful’ beam p = 236 MeV/c

• Design emittance y ONLY = 6 mm rad • This means:

(<pz> / mc) y y’ sqrt(1-r2) = 6 mm rad

Beam at Q9 (1)

ptot (GeV/c)py (GeV/c)px (GeV/c)

x (m) y (m)

Beam at Q9 (2)

• Beam is everything but Gaussian

• Top: x – pz correlation

• Bottom:a) <x> is not 0

b) px is asymmetric

x (m)

x (m) px (GeV/c)

pz (GeV/c)

x – pz correlation

• Correlation after Q9• Disappears after Pb

scatterer and inside spectrometer

265 MeV/c

x (m)

p z (

GeV

/c)

Cooling (1)

• Ideal Gaussian beam• JUNE04 beam

Designed assuming that the optics of the cooling channel were optimised for a central momentum of:<p> = 200 MeV/c

• Interface Turtle – Icool: centre of upstream solenoid

• Transmission = 80% • COOLING = ???

z (m)

(

m r

ad)

A few steps

• MICE channel designed for pcentral = 200 MeV/c

• Beam at Q9 designed to provide <p> = 236 MeV/c, due to A2 – p correlation

Steps:• Scale ALL coil currents in the

MICE channel by 236/207• Include Pb diffuser in Icool• Propagate Turtle beam from

Q9 into centre of spectrometer • Inside upstream spectrometer,

select ONLY events with ptot within 5% of 236 MeV/c

ptot

CUT

ptot (GeV/c)

B-field in Icool

• Fringe field from Q9 not included in Icool simulation

• Bz from solenoid ‘almost’ zero at Q9

• Pb scatterer inside fringe field from solenoid

• Bz in solenoid >> 4 T

Q9

Pb

z (m)

Bz

(T)

Icool vs. Turtle

• ptot in the centre of the upstream spectrometer:

a) Icool

b) Turtle

Note: Turtle uses thin lens approximation for solenoid fringe field

ptot (GeV/c)

Longitudinal momentum

• JUNE04 beam, all events

• JUNE04 beam with ptot cut

z (m)

<p z

> (

GeV

/c)

Cooling (2)

• JUNE04 beam, all events

• JUNE04 beam with ptot cut

z (m)

(

m r

ad)

Cooling (3)• Ideal Gaussian beam = 15.8%

Transmission = 98.6%

• JUNE04 beam = 11.1%

Transmission = 72.8%

• 7,075 after Q9• 5,412 total in upstream

spectrometer• Of these, 1,543 pass the cut

on ptot• 1,124 left in downstream

spectrometer z (m)

(

m r

ad)

Beta functions

• Gaussian beam, starting in upstream solenoid

• JUNE04 beam, all events

• JUNE04 beam with ptot cut

z (m)

(

m)

Conclusions• MICE coils and JUNE04 beam need some more work

• We have an actual beam + cooling channel design that works !!!

• Beamline: a) Reduce central momentum to

<p> = 207 MeV/cb) Have central momentum in the centre

of the ptot distribution, not in the tailc) More interface planes Turtle-Icool

• Keep an eye on event rate: 1,124 + / 7,075 + = 15.9% !!!