Time Resolved Diffuse X Ray Scattering

SwissFEL X-ray Free Electron Laser 21

Poster # 18

! #

! " # " # $

%# !" !

@ # )# *

' +! ( # &"' ( , * * )# &"' ' +!- ( . )# / )# , !(

,+!-#/ ,+ ( &'( ) ) * % +%, " +9 52 (# (# )# ' A 56 % , A B5 A B5 " )# , C B5 + 4 * $ 4$ 153 ! 555 ,* C B56 @*( * 0!0 !566C $ 4 0 5652 # #

$ 26 - &/; # . ! " A 56 $ B26 / - " " 9 # D 9 # D 97 # D ,* # D $ # D - , C =66 / + 866 0$-+ 1234$ 5 E 5656 758+9 ! 5666 # &";% !# : , # $ #0 ( )# ( % & # , 153 ; $ '<= ==5>5??? ! 123 ; $ ,<2 2=82682656 22 Hard X-ray Instrumentation Workshops 2011

Poster # 19

^ĞƌŝĂů&ĞŵƚŽƐĞĐŽŶĚƌǇƐƚĂůůŽŐƌĂƉŚǇ Ăƚ^ǁŝƐƐ&>yͲZĂǇ&ƌĞĞůĞĐƚƌŽŶ>ĂƐĞƌ &ƌĂŶĐĞƐĐŽ^ƚĞůůĂƚŽϭ͕<ĂƌŽůEĂƐƐϮ͕^ĂƓĂĂũƚϯ͕ĂƌůĂůĞŵĂŶϭ͕ĂŶŝĞůĞWŽŶƚĞϭ͕ŶĚƌĞǁs͘DĂƌƚŝŶϭ͕dŚŽŵĂƐ͘tŚŝƚĞϭ͕ŶƚŽŶĂƌƚǇϭ͕,ĞŶƌǇE͘ŚĂƉŵĂŶϭ͕Ϯ

ϭĞŶƚĞƌĨŽƌ&ƌĞĞͲůĞĐƚƌŽŶ>ĂƐĞƌ^ĐŝĞŶĐĞ͕^z͕EŽƚŬĞƐƚƌĂƐƐĞϴϱ͕ϮϮϲϬϳ,ĂŵďƵƌŐ͕'ĞƌŵĂŶǇ ϮhŶŝǀĞƌƐŝƚǇŽĨ,ĂŵďƵƌŐ͕>ƵƌƵƉĞƌŚĂƵƐƐĞĞϭϰϵ͕ϮϮϳϲϭ,ĂŵďƵƌŐ͕'ĞƌŵĂŶǇ ϯWŚŽƚŽŶ^ĐŝĞŶĐĞ͕^z͕EŽƚŬĞƐƚƌĂƐƐĞϴϱ͕ϮϮϲϬϳ,ĂŵďƵƌŐ͕'ĞƌŵĂŶǇ tŚǇƐĞƌŝĂůĐƌǇƐƚĂůůŽŐƌĂƉŚǇĂƚ^ǁŝƐƐ&>͍ ǆƉĞƌŝŵĞŶƚĂůůĂǇŽƵƚ

ͻ DŝĐƌŽĐƌǇƐƚĂůƐ ĂƌĞŵŽƌĞĞĂƐŝůǇŽďƚĂŝŶĞĚƚŚĂŶŵĂĐƌŽƐĐŽƉŝĐ ^ŝŶŐůĞ ĐƌǇƐƚĂů yͲƌĂǇ ĚŝĨĨƌĂĐƚŝŽŶ ƐŶĂƉƐŚŽƚƐ ŽŶĞƐ͘ ĐĂŶďĞĐŽůůĞĐƚĞĚĨƌŽŵĂƐƚƌĞĂŵŽĨĐƌǇƐƚĂůƐ ͻ /Ŷ ǀŝǀŽ ĐƌǇƐƚĂůƐ ĂůůŽǁ ƚŽ ŚĂǀĞ ƉƌŽƚĞŝŶƐ ŝŶ ƉŚǇƐŝŽůŽŐŝĐĂů ƵƐŝŶŐ ĨĞŵƚŽƐĞĐŽŶĚ ƉƵůƐĞƐ ĨƌŽŵ Ă ŚĂƌĚ yͲ ĐŽŶĚŝƚŝŽŶƐ;ŝŶĐůƵĚŝŶŐƉŽƐƚͲƚƌĂŶƐůĂƚŝŽŶĂůŵŽĚŝĨŝĐĂƚŝŽŶƐͿϭ͘ ƌĂǇ&ƌĞĞůĞĐƚƌŽŶ>ĂƐĞƌ;&>Ϳϯ dŚĞƐĞ ĐƌǇƐƚĂůƐ ĂƌĞ ƚŽŽ ƐŵĂůů ĨŽƌ ĐŽŶǀĞŶƚŝŽŶĂů ĂŶĂůǇƐĞƐ͘ ͻ dŝŵĞͲƌĞƐŽůǀĞĚƐƚƵĚŝĞƐŽĨ ŝƌƌĞǀĞƌƐŝďůĞ ƉƌŽĐĞƐƐĞƐ ƉŽƐƐŝďůĞ͕ ŝĨĨƌĂĐƚŝŽŶ ĨƌŽŵ ƵůƚƌĂͲƐŚŽƌƚ ƉƵůƐĞƐ ĐĂŶ ďĞ ďǇƐƚŽƉͲĨůŽǁŵŝǆŝŶŐŽƌƉŚŽƚŽŝŶŝƚŝĂƚŝŽŶ͘ ĐŽůůĞĐƚĞĚďĞĨŽƌĞƐŝŐŶŝĨŝĐĂŶƚĐŚĂŶŐĞƐŽĐĐƵƌ ͻ DĞĂƐƵƌĞŵĞŶƚƐ ĂƌĞ ĐĂƌƌŝĞĚ ŽƵƚ Ăƚ ƌŽŽŵ ƚĞŵƉĞƌĂƚƵƌĞ͕ ƚŽƚŚĞƐĂŵƉůĞϰ ĂǀŽŝĚŝŶŐƚŚĞŶĞĞĚĨŽƌĐƌǇŽŐĞŶŝĐƚĞŵƉĞƌĂƚƵƌĞƐǁŚŝĐŚĚŝƐƚŽƌƚ ƐƚƌƵĐƚƵƌĞƐϮ͘ dŝŵĞͲƌĞƐŽůǀĞĚ ƐƚƌƵĐƚƵƌĂů ƐƚƵĚŝĞƐ ŽĨ ͻ ^ĞƌŝĂů ĐƌǇƐƚĂůůŽŐƌĂƉŚǇ ĂůůŽǁƐ ŚŝŐŚĞƌ ƚŚƌŽƵŐŚƉƵƚ ŽĨ ŝƌƌĞǀĞƌƐŝďůĞƌĞĂĐƚŝŽŶĚǇŶĂŵŝĐƐŝŶďŝŽůŽŐŝĐĂů ƐƚƌƵĐƚƵƌĞƐĂŶĚƉŽƚĞŶƚŝĂůůǇĂůŽǁĞƌĐŽŶƐƵŵƉƚŝŽŶ ŽĨŵĂƚĞƌŝĂů ƐǇƐƚĞŵƐ ŝƐ ƉŽƐƐŝďůĞ͗ ŽƉƚŝĐĂů ƉƵŵƉ ůĂƐĞƌƐ ƚŚĂŶĐŽŶǀĞŶƚŝŽŶĂůƚĞĐŚŶŝƋƵĞƐ͘ ƐǇŶĐŚƌŽŶŝǌĞĚƚŽ&>ƉƵůƐĞƐŚĂǀĞďĞĞŶƵƐĞĚ ƚ ϭϬϬ ,ǌ͕ Ă ĐŽŵƉůĞƚĞ ĚĂƚĂƐĞƚ ĐĂŶ ďĞ ĂĐƋƵŝƌĞĚ ŝŶ Ϯϱ ƚŽŽďƚĂŝŶyͲƌĂǇĚŝĨĨƌĂĐƚŝŽŶ ƐŶĂƉƐŚŽƚƐĨƌŽŵ ŵŝŶƵƚĞƐ;ĂƚϭϬϭϭ ηͬŵůĂŶĚϭϱйŚŝƚƌĂƚĞͿ͕ǁŝƚŚŶŽĚĞĂĚͲƚŝŵĞ ƚŚĞ ĞǆĐŝƚĞĚ ƐƚĂƚĞƐ ŽĨ ƉƌŽƚĞŝŶƐ ŝŶ ϱ ĨŽƌƐĂŵƉůĞĐŚĂŶŐĞ͘ ŵŝĐƌŽĐƌǇƐƚĂůƐ ͘ ϭZ͘<ŽŽƉŵĂŶŶĞƚĂů͘ Ͳ ^ƵďŵŝƚƚĞĚ;ϮϬϭϭͿ͖Ϯ:͘^͘&ƌĂƐĞƌĞƚĂů͘ ʹ WE^ϭϬϴ͕ϯϵ ;ϮϬϭϭͿ͖ϯ,͘E͘ŚĂƉŵĂŶĞƚĂů͘ Ͳ EĂƚƵƌĞϰϳϬ͕ϳϯ ;ϮϬϭϭͿ͖ϰ͘ĂƌƚǇĞƚĂů͘ ʹ EĂƚƵƌĞWŚŽƚŽŶŝĐƐ;ϮϬϭϭͿ͖ϱh͘tĞŝĞƌƐƚĂůůĞƚĂů͘ZĞǀ͘^Đŝ͘/ŶƐƚƌ͘^ƵďŵŝƚƚĞĚ;ϮϬϭϭͿ ^ĂŵƉůĞĚĞůŝǀĞƌǇ ĂƚĂWƌŽĐĞƐƐŝŶŐĂŶĚĂŶĂůǇƐŝƐ

>ŝƋƵŝĚŵŝĐƌŽͲũĞƚ ŽŶĐĞŶƚƌĂƚŝŽŶ ,ŝƚƌĂƚĞ dŝŵĞ sŽůƵŵĞ ^ĞƌŝĂůĐƌǇƐƚĂůůŽŐƌĂƉŚǇĞǆƉĞƌŝŵĞŶƚƐŐĞŶĞƌĂƚĞůĂƌŐĞǀŽůƵŵĞƐŽĨĚĂƚĂ͗ ;ϭͲϱђŵͿϲ ηͬŵů й ŚŽƵƌƐ ŵů ƚϴDďͬŝŵĂŐĞǁĞŐĞŶĞƌĂƚĞĂďŽƵƚϮ͘ϴdďͬŚŽƵƌΛϭϬϬ,ǌ ϱ ϭϬϵ Ϭ͘ϴ ϳ Ϯ ǆƉĞƌŝŵĞŶƚĂůƐĞƌŝĞƐĐĂŶŐĞŶĞƌĂƚĞŽǀĞƌϭϬdďŽĨĚĂƚĂ;ϭϬϲ ŝŵĂŐĞƐͿ͘ ^ĂŵƉůĞ͗ ϭϬϭϬ ϭ͘ϱ ϯ͘ϱ ϭ ĂƚĂŚĂǀĞƚŽďĞƐƚŽƌĞĚĂŶĚƌĂƉŝĚůǇĂŶĂůǇǌĞĚ͖ŚŝƚƐŚĂǀĞƚŽďĞƐĞůĞĐƚĞĚĂŶĚ ŵŝĐƌŽĐƌǇƐƚĂůƐ ;ϭͲϭϬђŵͿ ϱ ϭϬϭϬ ϳ Ϭ͘ϳ Ϭ͘Ϯ ďĂĐŬŐƌŽƵŶĚŚĂƐƚŽďĞƌĞŵŽǀĞĚ͘ƵƚŽŵĂƚĞĚĚĂƚĂŝŶĨƌĂƐƚƌƵĐƚƵƌĞŝƐĞƐƐĞŶƚŝĂů͘ ƐƵƐƉĞŶĚĞĚ ŝŶ ĂƋƵĞŽƵƐ ϭϬϭϭ ϭϰ Ϭ͘ϰ Ϭ͘ϭ ƐŽůƵƚŝŽŶ͘ ^ĂŵƉůĞƌĞƋƵŝƌĞŵĞŶƚƐƚŽŐĞƚĂĐŽŵƉůĞƚĞ ŚĞĞƚĂŚʹ ĂƚĂƌĞĚƵĐĞƌĂŶĚŚŝƚĨŝŶĚĞƌ ^ĂŵƉůĞ ĞǆĐŚĂŶŐĞ ĂŶĚ ĚĂƚĂƐĞƚ;ϱϬϬϬŝŶĚĞǆĞĚƉĂƚƚĞƌŶƐͿƵƐŝŶŐϭђŵ ǆͲƌĂǇďĞĂŵĚŝĂŵĞƚĞƌĂŶĚϮђŵũĞƚĚŝĂŵĞƚĞƌ &>WƌĞͲƉƌŽĐĞƐƐŽƌĞĨĨŝĐŝĞŶƚůǇĂŶĂůǇƐĞƐƐĞƌŝĂůŝŵĂŐŝŶŐĚĂƚĂĂƚƵƉƚŽϲϬ,ǌĨƌĂŵĞ ŵŝǆŝŶŐĂƌĞƉŽƐƐŝďůĞ͘ ƌĂƚĞ͕ƉƌŽǀŝĚŝŶŐƌĂƉŝĚĨĞĞĚďĂĐŬĂŶĚĂŶĞĨĨŝĐŝĞŶƚĚĂƚĂĂŶĂůǇƐŝƐƉĂƚŚ͘ ϳ ^ŝϯEϰ ƐƵďƐƚƌĂƚĞ ǁŝƚŚ ŶĂŶŽůŝƚĞƌ ĚƌŽƉůĞƚƐ ŽĨ ĐƌǇƐƚĂů ϱϬϬђŵ ƌǇƐƚ&>ʹ ĂƚĂĂŶĂůǇƐŝƐ ƐƵƐƉĞŶƐŝŽŶ &>ĚĞǀĞůŽƉĞĚƐŽĨƚǁĂƌĞƚŚĂƚŝŶĚĞǆĞƐ͕ŝŶƚĞŐƌĂƚĞƐĂŶĚŵĞƌŐĞƐϴ ƐĞƌŝĂů ϯϬŶŵƚŚŝĐŬǁŝŶĚŽǁƐƐĞƉĂƌĂƚĞĚďǇϱϬϬђŵ ĐƌǇƐƚĂůůŽŐƌĂƉŚǇĚĂƚĂ ϭϬ ϭϬĐŵϮ ĐŚŝƉǁŽƵůĚŐŝǀĞĂďŽƵƚϭϬϰ ƐŝŶŐůĞĐƌǇƐƚĂůŚŝƚƐ ϲ͘ĞWŽŶƚĞĞƚĂů͘ Ͳ :͘WŚǇƐ͘ϰϭ͕ϭϵϱϱϬϱ;ϮϬϬϴͿ ϳd͘͘tŚŝƚĞĞƚĂů͘ʹ ƐƵďŵŝƚƚĞĚ͖ϴZ͘͘<ŝƌŝĂŶĞƚĂů͘ ĐƚĂƌǇƐƚϲϳ͕ϭϯϭ;ϮϬϭϭͿ >>^ZĞƐƵůƚƐ ŽŶǀĞƌŐĞŶƚĞĂŵʹ ,ŝŐŚĂŶĚǁŝĚƚŚ ĂŵĂŐĞ ^ŝŵƵůĂƚŝŽŶƐŽŶW^/ƐŚŽǁƚŚĂƚĨĞǁĞƌƉĂƚƚĞƌŶƐĂƌĞ ƐŝŶŐůĞ ĐƌǇƐƚĂů ƌĞƋƵŝƌĞĚ ǁŚĞŶ ƵƐŝŶŐ Ă ĐŽŶǀĞƌŐĞŶƚ ďĞĂŵ Žƌ ŚŝŐŚĞƌ ƉĂƚƚĞƌŶ ĂĐƋƵŝƌĞĚ ďĂŶĚǁŝĚƚŚ ^ŝŵƵůĂƚĞĚĚĂŵĂŐĞ Ăƚ>>^ϯ ŝŶƚŚĞĐĞŶƚĞƌ ŽĨĂϳϬϬŶŵ W^/ĐƌǇƐƚĂů

^ŝŵƵůĂƚŝŽŶƐĂƌĞĚŽŶĞƵƐŝŶŐĂƉůĂƐŵĂĐŽĚĞϵ ;ĐƌĞƚŝŶͿ͕ ĂŶĚĚŝƐƉůĂĐĞŵĞŶƚƐĂƌĞĐĂůĐƵůĂƚĞĚďĂƐĞĚŽŶĚŝĨĨƵƐŝŽŶŝŶ ŶĂůǇƐŝƐŽĨ>>^ĞǆƉĞƌŝŵĞŶƚĂů ĂǁĞĂŬůǇŝŽŶŝǌĞĚƉůĂƐŵĂϰ ϯ ƌĞƐƵůƚƐ ŽŶWŚŽƚŽƐǇƐƚĞŵ/;W^/Ϳ ZĂĚŝĂƚŝŽŶĚĂŵĂŐĞǁŝůůŶŽƚďĞĂŶŝƐƐƵĞĂƚ^ǁŝƐƐ&> ƐŚŽǁƐƚŚĂƚĂďŽƵƚϭϬϰ ƉĂƚƚĞƌŶƐĂƌĞƌĞƋƵŝƌĞĚ ϵ͘ĂůĞŵĂŶĞƚĂů͘:͘DŽĚ͘KƉƚ͘ϱϴ;ϮϬϭϭͿ ĞƚĞĐƚŽƌZĞƋƵŝƌĞŵĞŶƚƐ ,ŝŐŚƌĞƐŽůƵƚŝŽŶ͗ϮϬϰϴǆϮϬϰϴƉŝǆĞůƐŽƌŵŽƌĞ͘tŽƌŬŝŶŐĚŝƐƚĂŶĐĞŽĨĂďŽƵƚϳ ĐŵĨŽƌĂ WĂƌĂŵĞƚĞƌ ZĞƋƵŝƌĞŵĞŶƚ DŽƚŝǀĂƚŝŽŶ ϭϬĐŵǁŝĚĞĚĞƚĞĐƚŽƌ;ϱϬђŵƉŝǆĞůƐͿǁŽƌŬƐǁĞůů; Ϯ ƌĞƐŽůƵƚŝŽŶͿĂƚϭϬŬĞs͘>ĂƌŐĞƌ ϰͲϭϰŬĞs ,ŝŐŚƌĞƐŽůƵƚŝŽŶ ƉŝǆĞůƐK<ŝĨǁŽƌŬŝŶŐĚŝƐƚĂŶĐĞĐĂŶďĞŝŶĐƌĞĂƐĞĚ͘'ŽŽĚYĂĐƌŽƐƐĞŶĞƌŐǇƌĂŶŐĞŽĨϰͲ ŶĞƌŐǇ ϭϰŬĞs ^ƚĂďŝůŝƚǇ Ϭ͘ϭй DƉŚĂƐŝŶŐ ,ŝŐŚĚǇŶĂŵŝĐƌĂŶŐĞ͗хϭϬϰ ĂƚϵŬĞs;ŶĞĞĚƚŽŵĞĂƐƵƌĞŝŶƚĞŶƐŝƚǇŝŶďƌŝŐŚƚƉĞĂŬƐͿ ƚŚŝŐŚĞƌďĂŶĚǁŝĚƚŚ ĐĐƵƌĂƚĞŵĞĂƐƵƌĞŵĞŶƚŽĨƐŵĂůů͕ďƌŝŐŚƚƉĞĂŬƐ͗ǁĞůůĐŽŶƚƌŽůůĞĚƐĂƚƵƌĂƚŝŽŶĂŶĚŶŽ Ϭ͘ϭͲ ϰй ĂŶĚǁŝĚƚŚ ĨĞǁĞƌƉĂƚƚĞƌŶƐĂƌĞ ƐƉŝůůŽǀĞƌŝŶŽƌĚĞƌƚŽĚŝƐƚŝŶŐƵŝƐŚďĞƚǁĞĞŶĐůŽƐĞƉĞĂŬƐ ^ƚĂďŝůŝƚǇ ϭϬй ZĞĂĚŽƵƚĂƚϭϬϬ,ǌ;ƐĂǀĞŽŶĞĨƌĂŵĞƉĞƌƉƵůƐĞͿ ŶĞĞĚĞĚ ZĂĚŝĂƚŝŽŶŚĂƌĚŶĞƐƐͬZĞƉůĂĐĞĂďůĞŵŽĚƵůĞƐ͗ƉŝǆĞůƐĐĂŶďĞĚĂŵĂŐĞĚďǇƐƚƌŽŶŐ ĞĂŵWŽƐŝƚŝŽŶ ^ƚĂďŝůŝƚǇ фϭђŵ :ĞƚͬƐĂŵƉůĞƐŝǌĞ ĚŝĨĨƌĂĐƚŝŽŶĚƵƌŝŶŐĞǆƉĞƌŝŵĞŶƚƐ ĞĂŵ^ŝǌĞ ϭђŵ ,ŝŐŚƉŽǁĞƌĚĞŶƐŝƚǇ ^ĂŵƉůĞŶǀŝƌŽŶŵĞŶƚZĞƋƵŝƌĞŵĞŶƚƐ DĂǆŝŵƵŵƉŽǁĞƌ sĂĐƵƵŵĐŚĂŵďĞƌ;хϭϬϬϬůͬƐƉƵŵƉŝŶŐƐƉĞĞĚͿ η хϭϬϭϬ DŽƚŽƌŝǌĞĚƐƚĂŐĞƐƚŽĂůŝŐŶŝŶũĞĐƚŽƌĂŶĚͬŽƌĨŝǆĞĚƚĂƌŐĞƚƐĂŵƉůĞƐ WŚŽƚŽŶƐƉĞƌƉƵůƐĞ ĚĞŶƐŝƚǇĚĞƐŝƌĞĚ ^ƉĂƌĞĨůĂŶŐĞƐĂůŝŐŶĞĚƚŽŝŶƚĞƌĂĐƚŝŽŶƌĞŐŝŽŶƚŽĐŽŶŶĞĐƚŝŶũĞĐƚŽƌƐ͕ĂĚĚƉƵŵƉůĂƐĞƌ ^ƚĂďŝůŝƚǇ ϭϬй ^ǇŶĐŚƌŽŶŝǌĞĚƉƵŵƉůĂƐĞƌĨŽƌƚŝŵĞƌĞƐŽůǀĞĚĞǆƉĞƌŝŵĞŶƚƐ ĨƐ ϭϬͲϱϬ ͚ŝĨĨƌĂĐƚŝŽŶďĞĨŽƌĞ ;dƵŶĞĂďůĞŽǀĞƌϯϬϬŶŵфʄфϳϬϬŶŵ͕ƉƵůƐĞĞŶĞƌŐǇϱͲϭϬђ:͕ƉƵůƐĞůĞŶŐƚŚϭͲϭϬϬŶƐͿ WƵůƐĞůĞŶŐƚŚ ŝĂŐŶŽƐƚŝĐZĞƋƵŝƌĞŵĞŶƚƐ ^ƚĂďŝůŝƚǇ ϭй ĚĞƐƚƌƵĐƚŝŽŶ͛ yͲƌĂǇ/ŶƚĞŶƐŝƚǇ ;WƵůƐĞĞŶĞƌŐǇŝŶŵ:͕ƉĞƌƐŝŶŐůĞƐŚŽƚͿ WƵůƐĞĂƌƌŝǀĂůƚŝŵĞ ĨƐ Ͳ Ͳ yͲƌĂǇŶĞƌŐǇͬ^ƉĞĐƚƌƵŵ ;yͲƌĂǇĞŶĞƌŐǇĂŶĚďĂŶĚǁŝĚƚŚƐƉƌĞĂĚ͕ŝŶĞs͕ƉĞƌƐŝŶŐůĞƐŚŽƚͿ WƵůƐĞĚƵƌĂƚŝŽŶ;yͲƌĂǇƉƵůƐĞĚƵƌĂƚŝŽŶŝŶĨƐ͕ƉĞƌƐŚŽƚͿ SwissFEL X-ray Free Electron Laser 23

Poster # 20

Probing magnetic phase transitions S.O.Mariager, P.Beaud, A.Caviezel, S.Grübel, J.Johnson, G.Ingold FEMTO group, SLS, Paul Scherrer Institut, 5232 Villigen, Switzerland. U.Staub, V.Scagnoli RESOXS team, SLS, Paul Scherrer Institut, 5232 Villigen, Switzerland. S.L.Johnson Institute for Quantum Electronics , Swiss Federal Institute of Technology (ETH), Zürich, Switzerland. L. Le Guyader, F. Nolting, C. Quitmann SLS, Paul Scherrer Institut, 5232 Villigen, Switzerland. Motivation Exploiting the coherence $SRVVLEOHPHWKRGIRUVWXG\LQJGRPDLQG\QDPLFV Structural and magnetic dynamics &RKHUHQWVFDWWHULQJ ;UD\SKRWRQFRUUHODWLRQVSHFWURVFRS\ ⋅ :HKDYHVWXGLHGWKHDOOR\)H5KDWWKH)(072VOLFLQJVRXUFH$W7a 'RQHZLWK SKVQRZ±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

Resonant magnetic x-ray diffraction Non-resonant magnetic x-ray diffraction 7KHZHDNPDJQHWLFVFDWWHULQJFDQEHVLJQLILFDQWO\HQKDQFHG a DWDEVRUSWLRQHGJHV )RUVWXG\LQJPDJQHWLFRUGHUFRPELQHGZLWKVWUXFWXUDOPRGXODWLRQV6SLQDQG +DUG[UD\VXLWDEOHIRU.HGJHVRIWUDQVLWLRQPHWDOVDQG/HGJHVRIIDQGGUDUHHDUWKHOHPHQWV RUELWDORUGHUFDQEHGHFRXSOHG 5HTXLUHVWXQDEOHHQHUJ\ORZEDQGZLGWKDQGTXLFNHQHUJ\VFDQV )DUIURPDEVRUSWLRQHGJHV,≈ _) 4δU ± L (PF ) _ 2QHDGYDQWDJHFRPSDUHGWRVRIW[UD\VLVDFFHVVWRDODUJHYROXPHLQUHFLSURFDOVSDFH δ 0 +HUH) 4δU IRU4 + DQGδU .HGJHPDJQHWLFFLUFXODUGLFKURLVP KDVEHHQUHSRUWHG δ ,QWKHFODVVLFDOH[DPSOH1L2 ) ) a⋅GRDEOHDW6ZLVV)(/ &LUFXODUSRODUL]DWLRQSRVVLEOHZLWKSKDVHUHWDUGHUV7UDQVPLVVLRQa/LPLWHGIOH[LELOLW\ 0DJQHWLF FKDUJH 5HTXLUHVSRODUL]DWLRQFRQWUROIRULQFRPLQJEHDP ZKHQFKDQJLQJIURPOHIWWRULJKWSRODUL]DWLRQ $QGSRODUL]DWLRQDQDO\VLVRIWKHVFDWWHUHGEHDP /(%HUPDQ HWDO5HY6FL,QVWU &0D]]ROL HWDO3K\V5HY% $GLDPRQGFU\VWDO[UD\SKDVHUHWDUGHU FDQ 6HWXSXVHGIRUFRPSOHWHSRODUL]DWLRQ SURGXFHFLUFXODUSRODUL]HGOLJKWIURPOLQHDULO\ DQDO\VLVDSSOLHGWRUHVRQDQWVFDWWHULQJDWWKH SRODUL]HGEHDPVLQWKHKDUG[UD\UDQJH &U.HGJH NH9 NH9 %-.LP HWDO6FLHQFH +&:DONHU HWDO1DWXUH 0RWWLQVXODWLQJVWDWHLQGXFHGE\UHODWLYLVWLF 7KHPDJQHWLFVWUXFWXUH &\FORLGDO GRPDLQV ' VSLQRUELWFRXSOLQJLQWKHOD\HUHG'WUDQVLWLRQ FRPELQHGZLWKDVWUXFWXUDOPRGXODWLRQ ( LQIOXHQFHVWKHUHFRUGHGVLJQDO PHWDOR[LGH6U,U2

Future end station Parameter Unit Requirement Motivation/remarks 6DPSOHPDQLSXODWRU IRUJUDFLQJLQFLGHQFHGLIIUDFWLRQ keV 2-12 Select edges for resonant diffraction Energy &RPSOHWHJRQLRPHWHU KH[DSRGRUHTXLYDOHQWVWDJH Stability eV <1 Monochromator needed. 6DPSOHURWDWLRQDURXQGVXUIDFHQRUPDODQGIRULQFLGHQFHDQJOH <1 Monochormator needed 6XIILFLHQWWUDQVODWLRQVWRSRVLWLRQVDPSOH Bandwidth Stability % bw <10 6DPSOHHQYLURQPHQW 7HPSHUDWXUHFRQWURO± . Beam position Stability μm <1 For laser overlap and gracing incidence. 0DJQHWLFILHOG3HUPDQHQWPDJQHWV7LQIOH[LEOH6&PDJQHWV!7ELJVHWXS μm <5 Gracing incidence diffraction requires horizontal Beam size &KDOOHQJH6HFXUHDFFHVVIRUSXPSODVHU focus. &KDOOHQJH6HFXUHFRKHUHQFHRI[UD\EHDP 108-1012 Likely limited by sample damage. Attenuation could 'HWHFWRUWZRGHJUHHVRIIUHHGRP FLUFOHV IRUGHWHFWRUPRYHPHQW Photons per pulse be needed. ,QDGGLWLRQYDULDEOHVDPSOHWRGHWHFWRUGLVWDQFHLVSUHIHUDEOH Stability Shot to shot normalization for time-resolved. $QGQRQUHVRQDQWPDJQHWLFVFDWWHULQJFDQUHTXLUHSRODUL]DWLRQDQDO\VLV fs (FWHM) 10 Control of jitter to pump laser is just as important. Pulse length ,QWHJUDWHGGLDJQRVWLFV Stability % 10 [UD\GLDJQRVWLFVRIWLPLQJSRODUL]DWLRQFRKHUHQFHHWF Beam parameter changes during experiments Range eV 100 Energy scan at resonance Energe Step eV < 1 For resolving resonant peak 3UHVHQWGHVLJQIRUDJUDFLQJLQFLGHQFH GLIIUDFWLRQHQGVWDWLRQDW)(0726/6 Scan eV/sec 1 7KHVDPSOHLVPRXQWHGLQDYDFXXP Beam size Only changed during setup FKDPEHUZLWKRSWLRQDOFRROLQJ7KH Only changed during setup GHWHFWRULVPRXQWHGRXWVLGHWKHFKDPEHU Pulse length $SDLURI.%PLUURUVIRFXVHVWKH;UD\ Beam geometry EHDP:RUNVSDFHQHHGHG P min mm ~500 UHV chamber, sample rotation/translation, Working distance magnetic field. 24 Hard X-ray Instrumentation Workshops 2011

Poster # 21

Paul Scherrer Institute, 5232, Villigen, Switzerland

EMPA, Feuerwerkerstrasse 39, 3602 Thun, Switzerland SwissFEL X-ray Free Electron Laser 25

Poster # 22

NUCLEAR ENERGY AND SAFETY

Dynamics of irradiation-induced defects in nuclear materials: a proposed energetic-ion pump - X-ray FEL probe experimental approach A. Froideval1, J. Chen1, C. Degueldre1, M. Krack1, G. Kuri1, M. Martin1, S. Portier1, M. A. Pouchon1, B. D. Patterson2

1 Nuclear Energy and Safety (NES), Paul Scherrer Institute, 5232 Villigen PSI, Switzerland 2 SwissFEL Project, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland

Context of the study Motivation of the study Nuclear reactor types with specific operating conditions [1]: ƒ Understanding of: ƒ temporal evolution of the primary stage of the irradiation-induced damage in nuclear materials (e.g. cladding materials (Zr-alloys), reactor pressure vessel (RPV) steels, ceramics, composite materials) [2]

production of primary defects diffusion and reactions MD simulation of primary damage state in UO * of defects 2 e.g. 14 nm ƒ clustering ƒ annihilation at sinks ƒ interaction with foreign atoms

0.5 ps 1.3 ps 2.0 ps

Uvac Uint < ~ 10 ps < ~ 10 nm > ~ 10 ps > ~ 10 nm demanding current future O O operating conditions: Gen. II systems Gen. IV systems vac int * from F. Devynck & M. Krack (LRS / NES / PSI)) ƒ temperature ~ 300 - 400 °C up to 700 °C ƒ impact on the materials microstructure ƒ dosecladding ~ 20 dpa up to 200 dpa ƒ fuel burnup ~ 60 MWd/kg ~ 200 MWd/kg → Irradiation can induce modifications of the materials properties – e.g. irradiation-induced hardening and embrittlement, irradiation-induced high Irradiation damage is and will be – in the future – a safety-relevant issue. temperature creep, irradiation-induced swelling – and dimensions.

Proposed experiment ƒ Key parameters of the experiment ƒ Sketch of the experiment ƒ Proposed PUMP-PROBE diffuse X-ray scattering experimental scheme [°] : Parameter Unit Requirement Motivation / Remarks diamond based Fresnel ƒ K e y p a r a m e t e r s of t h e X - r a y F E L r a d i a t i o n zone plate [3] 2D detector Energy keV 5 - 20 X-ray diffuse scattering stability microscope incoming Bandwidth % ~ 0.05 X-rays (sample visualization) scattered stability < ± 10% bw X-rays Position stability nm < 100 stable incoming beam is wished Size nm ~ 200 Photons / pulse #ph ~1011 streak camera stability % ± 10 Pulse duration fs 50 pulsed ion source temperature stability ƒ control Pulse arrival fs ~ 200 stability 6-axis nanopositioning time sample manipulator ƒ B e a m p a r a m e t e r c h a n g e s d u r i n g e x p e r i m e n t (hexapod) Energy range / step eV < 1 use of high-brightness seeded pulses: possible additional approach: stack of coherent diffraction rate pump vacuum chamber eV/min patterns recorded at an absorption edge, i.e. elemental sample specificity (e.g. Zr, Nb, Y, Am, … for Ephoton > 12 keV) (possibly radioactive) Size range / step μm --- rate --- d § 0.5 m Pulse length range/step --- rate --- dsample-detector § 1 m laboratory C ƒ B e a m g e o m e t r y μ (1 A 100 LA) (LA: licensing limit (specified in Annex 3 Column 10 of the “Strahlenschutzgesetz, March, 22nd, 1991)) Slope rad --- minimum required [°]common scientific interests with B. Larson and c o-workers (Oak Ridge National Working distance mm ~ 250 Laboratory) and with T. Cowan and co-workers (HZDR). ƒ O t h e r Diamond-based Fresnel focusing of hard X-rays down to ~200 ×200 nm2 spot zone plate yes size at the sample surface (C. David, LMN / PSI, [3]) ƒ Expected results Transverse coherence yes required for coherent X-ray diffraction Laboratory C yes sample activity (A): 1 A 100 LA, with LA: licensing limit) ƒ statistics of the formation of the primary interstitials and vacancy * defects, and their nanometer sized clusters Ion source (laser-accelerated high-energetic ion source ) Ion energy keV 10 - > 1000 ƒ evolution of their structure, distribution and mobility Particle type element e.g. Cs, I, Fe, Kr, Xe, He, H ƒ diffusion rate of the vacancies and interstitials clusters Single/multiple ion(s) yes adjustable fluence experimental validations of MD simulations Arrival time ps < 1 synchronization possible development of advanced irradiation-resistant materials *Interested partners: J. Ullrich, R. Moshammer (MPI Heidelberg, Germany)

References Acknowledgments [1] S.J. Zinkle, in: NEA Workshop Proceedings, Karlsruhe, June 4-6, 2007. The authors wish to thank F. Devynck, C. David, B. Pedrini, A. Cervellino from PSI, B. Larson from Oak [2] A. Froideval et al., J. Nucl. Mater. 416 (2011) 242-251. Ridge National Laboratory, J. Ullrich, R. Moshammer from MPI Heidelberg, and T. Cowan from HZDR for [3] C. David et al., Scientific Reports 1, 57 (2011); DOI: 10.1038/srep00057. helpful discussions and interest in the work. 26 Hard X-ray Instrumentation Workshops 2011

Poster # 23

Diffractive optics for focusing and characterization of hard X-ray free electron laser radiation Petri Karvinen1 , Simon Rutishauser111 , Sergey Gorelick , Aldo Mozzanica , Dominic Greiffenberg 1 , Jacek Krzywinski22 , David M. Fritz , Henrik Till Lemke 2 , Marco Cammarata 2 , Christian David 1 (1) Paul Scherrer Institut, 5232 Villigen PSI, Switzerland (2) SLAC National Accelerator Laboratory, Menlo Park, CA 94025, USA Motivation Diamond optics Conventional Fresnel zone plates (FZPs) and diffraction gratings, commonly used The high melting point, good thermal conductivity and low X-ray absorption make at hard X-ray synchrotron sources, are quickly damaged by the high heat loads diamond a suitable material for optics that can withstand the high power density of imposed by intense XFEL radiation. a free electron laser. No damage was observed on diamond based FZPs after 10 000 LCLS shots.

4 µm 4 µm 4 µm

Figure 1: Gold Fresnel zone plates exposed to 8 keV FEL pulses at LCLS. Figure 2: Diamond Fresnel zone plates used to focus LCLS beam. (a) overview, (a) no irradiation, (b) after 1000 pulses, (c) after 10 000 pulses. (b) central zones, (c) 100nm wide outermost zones.

Nanofocusing of hard XFEL radiation with diamond Fresnel zone plates Diamond based FZPs with diameter of 500 µm and outernmost zone width of 100 nm were used at LCLS. The focusing properties were analyzed using the imprint method (Fig. 3), observing a 320 nm focus (FWHM) limited by chromatic aberrations (Fig. 4). [1]

Figure 3: SEM images of 8 keV XFEL shot imprints on Au coated glass substrates. The beamline attenuators were used Figure 4: (a) The squared imprint diameters as a function the logarithm of the pulse energy reveals two to make imprints with different pulse energies, ranging from regions that can be approximated by straight lines corresponding to a double-Gaussian spot. (b) Beam unattenuated beam with ~5x10-5 J pulse energy and 3.5 µm profile derived from the same data set. Knowing the pulse length, total energy and zone plate efficiency 17 2 imprint diameter (a) to ~2x10-8 J pulse and 190 nm imprint (f). peak power densities of approximately 4x10 W/cm can be calculated. A non-invasive single shot spectrometer setup based on a focusing diamond grating A spectrometer setup based on a diamond grating and a charge integrating microstrip detector (Gotthard) in grazing incidence was used to measure single shot spectra of the LCLS beam. The grating diffracts only a small fraction (less than 1%) of the beam, making the setup ideal for monitoring purposes.

4 x 10 a: side view Different photon energies 3 Shot #1 X-ray diffracted in different angles Shot #2 beam Shot #3 2.5 Shot #4 1.5 eV Detector

Diamond grating 2

bb:: ffrontront c: top 1.5 I [a.u.]

dd:: ffrontront e: top Sample 0.5

0 Figure 5: Schematic diagrams of setups based on diamond spectrometer gratings 5990 6000 6010 6020 6030 6040 6050 6060 6070 6080 (a) side view, (b) and (c) front- and top views of a setup producing one focus, Photon Energy [eV] respectively (d) and (e) front- and top views of a setup producing two foci for Figure 6: Single shot spectra of four consecutive XFEL shots, measured at LCLS. spectroscopy with single-shot reference spectra.

[1] “Nanofocusing of hard X-ray free electron laser pulses using diamond based Fresnel zone plates”, C. David, S. Gorelick, S. Rutishauser, J. Krzywinski, J. Vila-Comamala, V.A. Guzenko, O. Bunk, E. Färm, M. Ritala, M. Cammarata, D.M. Fritz, R. Barrett, L. Samoylova, J. Grünert and H. Sinn, Sci. Rep. 1 (2011) , 57 SwissFEL X-ray Free Electron Laser 27

Poster # 24

CAMP: Flexible User End Station for Multidimensional Imaging Experiments at XFELs

Artem Rudenko1,2, Sascha Epp1,2, Daniel Rolles1,3, Robert Hartmann4, Lutz Foucar1,3, Benedikt Rudek1,2, Benjamin Erk1,2, Faton Krasniqi1,3, Lothar Strüder5, Ilme Schlichting1,3 and Joachim Ullrich1,2 1Max-Planck Advanced Study Group (ASG) at CFEL, Hamburg; 2Max-Planck-Institut für Kernphysik, Heidelberg; 3Max-Planck-Institut für Medizinische Forschung, Heidelberg; 4PN Sensor GmbH, München, 5Max-Planck-Halbleiterlabor, München

Introduction Key features of CAMP

X-ray free-electron lasers (FEL), delivers X-ray pulses (0.1 – 10 KeV) with unprecedentedly high intensities (~ ¾ Simultaneous photon, ion and electron detection 1013 photons/pulse) and short pulse durations (3 – 300 fs), opening the way towards numerous applications ¾ Fast readout, large area energy resolving pnCCD detectors for X-ray photons ranging from atomic, molecular and optical (AMO) physics to femtochemistry, material science and biology. In particular, it enables a variety of novel imaging schemes with Å spatial and fs temporal resolution. ¾ ‘Reaction microscope’ or ‘velocity map imaging’ (VMI) spectrometers for charged particles ¾ Atomic, (ultracold) molecular & cluster jets, injectors for nanocrystalls, single bioparticles & Here we present a concept of a dedicated multi-purpose instrument for multi-parameter imaging experiments at XFELs based on simultaneous detection of ions, electrons, scattered and fluorescence photons. Such end aerosols; fixed target samples station is successfully employed in more than 20 experiments at LCLS, has been recently commissioned at ¾ Laser incoupling for molecular alignment and pump-probe experiments SACLA XFEL in Japan, planned for FLASH, and (in modified form) for European XFEL. ¾ Dedicated software package: CASS – CFEL-ASG Software Suite (developed by L. Foucar et. al.)

CAMP: CFEL-ASG Multi Purpose instrument pnCCD detectors

Top: Side-view of the CAMP instrument. Chamber C1 houses Photograph of a 150 mm silicon wafer with two The pnCCD detector module with an circuit board on top Top: Detailed view of chamber C2 with the the removable charged-particle spectrometer (yellow), jet large area pnCCD devices. and the CCD chip and read-out ASICs on the bottom targets (not depicted), holders for fixed targets, apertures, two pnCCD detectors and the possible etc. Chamber C2 houses two pnCCD x-ray photon detectors. translations of the front detector.

Bottom: CAMP installed at the AMO beamline at LCLS Bottom: Sketch of the detector geometry. Key detector parameters

Activearea 59cm2 (76,8x76,8mm2),1024x1024pixel Pixelsize 75x75ȝm2 FramereadͲoutrate Upto200Hz

SingleͲphotonenergyresolution 50eV @800eV,150eV @5900eV

Countrate/pixel/bunch >103 photons@2keV Quantumefficiency >80%from0.3to12keV Operatingrange 0.1–25keV

Bottom: The VMI spectrometer and the front pnCCD inside the CAMP chamber pnCCD1 • mounted on three dimensional moving stages pnCCD quantum efficiency Each detector module hosts one pnCCD detector of 512×1024 pixel together with ionionion detectordetectordetector frontfrontfrront pnCCDpnpnCCDCCD • distance from the interaction point adjustable by up to 25cm along the FEL beam • vertically the modules are • eight 128-channel, low noise readout ASICs independently moveable • clock drivers • additional filters for contamination • analog signal buffers VMI protection • power rail filters • Made of UHV compatible materials pnCCD2 • mounted fix to the chamber FEL • alignment of the FEL-beam by One detector system is moving the CAMP instrument withwith aappliedpplied oopticalptical llightight ffilters:ilterrs:s composed of two modules. • distance to the interaction point can be increased by introducing a § 10-3 spacer tube. -6 targettargettarrget § 10 beambeam Optical light filters installed for pump- probe experiments

electronelectronelecttron detectordetectordetector L. Strüder et al., Nucl. Instr. Meth. Phys. Res. A. 614, 483 (2010).

Examples of imaging experiments in CAMP CAMP experiments at LCLS Femtosecond protein nanocrystallography Coherent diffractive imaging of single viruses • Correlated electron, ion, and fluorescence Chapman et. al., Nature 470 73 (2011) Seibert et. al., Nature 470 78 (2011) spectroscopy of atoms and small molecules • Ultrafast imaging of X-ray excited clusters • Coherent diffractive imaging of fixed and injected nanocrystals, viruses and aerosoles • (Time-resolved) diffractive imaging of laser-aligned molecules • Time-resolved photoelectron diffraction in laser-aligned molecules • Ultrafast probing of x-ray-induced lattice and electron Left: Low-angle diffraction patterns dynamics in graphite of the photosystem I nanocrystals recorded on the rear pnCCDs, • Coulomb explosion imaging of small organic molecules Right: Single-shot diffraction patterns from single mimi virus • Time-resolved photoelectron holography • Ultrafast dynamics in highly excited semiconductors 28 Hard X-ray Instrumentation Workshops 2011

Poster # 25

Laboratory of Biomolecular Research (BMR)

Femtosecond analysis of protein nanocrystals and supramolecular complexes Valérie Panneels, Ching-Ju Tsai, Xiao-Dan Li, Guido Capitani, Michel Steinmetz and Gebhard Schertler

Membrane proteins like GPCRs and transporters and supramolecular complexes like centrioles are of prime interest! Membrane proteins (MPs) represent key components of cell membranes, about one fourth of the human proteome and 60% of the drug targets. We are interested in learning about the different conformations a MP can adopt in response to interacting molecules.

Membrane proteins (MPs) are still underrepresented in the protein structure database pdb

Kitagawa et al. Cell 2011

Standfuss et al. Nature 2011

Stephen White’s Database, Membrane Proteins of Known Structure Warne et al. Nature 2011 Serial crystallography of protein nanocrystals Why do we want nanocrystals? ADV. - they could be easier to grow than large single crystals - they are likely better ordered due for example to less long range disorder Electron density map of Photosystem I at 8.5 Å resolution, obtained by fs-crystallography

Nanocrystals in the

Chapman et al. Nature 2011

Fromme et al. Curr.Op.Struct.Biol. 2011 Chapman et al. Nature 2011

Serial single shots of thousands of nanocrystals on the principle of “diffraction-before-destruction” ADV. - no time for radiation damage - diffraction on fully hydrated crystals Nanocrystals Delivery to the Beam SPENCE GUN: Aerojet droplet source -triggered breakup (increases crystal hit rate), repetition rate of 120Hz -1 Requirements on the X-ray beam parameters 10 m s Parameter Unit Requirement Motivation / Remarks Beam parameters keV > 8 Diffraction at biomolecules Energy nanosecond image stability Liquid coneUntriggered breakup Triggered breakup % Bandwidth from John Spence, Shapiro et al. J.Synchrotron Radiation 2008 stability Beam position stability

Sketch of Experimental Setup for Nanocrystals Analysis Beam size μm 5 -10

#ph 11 Maximal flux wished Photons per pulse stability fs < 50 Diffract and destroy regime required Pulse length stability 20% Pulse arrival time stability fs --- Not an issue Beam parameter changes during experiment Not an issue

Other Special Requirements - special sample delivery to the beam (Spence gun or grid scanning)

- sample to detector distance ~ 1m - strong computer capacity for direct data processing

Strüder et al. Nuclear Instr.Meth.Phys.Res.A 2010 SwissFEL X-ray Free Electron Laser 29

Poster # 26

Table-top Soft X-ray Laser in the X-ray Free-Electron Laser Era D. Bleiner*, F. Staub, J.E. Balmer, Th. Feurer, Institute for Applied Physics, University of Bern, Sidlerstrasse 5, CH-3012 Berne, Switzerland. *E-mail: [email protected]

Presented at Workshop on Hard X-ray Instrumentation at the SwissFEL, Bern, 21 November 2011

Introduction Research Program Lab-scale, high-brightness, soft X-ray sources are needed to BeAGLE is the instrumental platform for two research groups: enable high throughput research and industrial activity. The - The X-ray Laser Physics group focuses on: proof-of-concept research performed at large-scale user-facilities > down-scaling the wavelength towards the water-window; (e.g. synchrotrons, xFEL) can be brought closer to society's >enhancing the repetition rate to 10-Hz; demands by means of compact performing installations. The - The EUV Plasma Radiation group focuses on enabling: Bern Advanced Glass Laser for Experiment ("BeAGLE") is one of > Nano-imaging with extreme ultraviolet ("actinic mode"); these. The main characteristics of lab-scale X-ray lasers are here > Coherent diffraction ("lensless imaging"); summarized. > High-resolution surface spectrometry;

Fundamental Aspects of Soft X-ray Lasing ComparisonAll-Transmissive with Contending Design Technologies The X-ray laser is induced by amplified Brightness Fig 1.b The X-ray laser (XRL) has significantly matured since the spontaneous emission (ASE) across a plasma early experiments in the 1960-70s. Nowadays XRLs 24+ as 0.001% Fig. 2.b column, produced by line-focus on a solid target Se 29+ 22+ Saturation Ge 12+ La 18+ 14+ offer high brightness (Fig. 2.a) and linewidths below 38+ 20+ 14+ 24+ (Fig 1.a). Lasing from various targets spans 28+ Pd Fe Ti Dy Cd 19+ Mo Se 34+ Ba 23+ 22+ ) Sb Ag Ge Sm 22+

gL 0.01%, with pulse duration in the ps range (Fig. 2.b). across the XUV (Fig 1.b). Travelling-wave (TW) Sn 24+ fs 0.01% Se t ( t 29+ excitation enhances the conversion efficiency c 37+ Y

u These set of parameters are suitable for a number Ag 24+ d Se 24+ 28+ Se 22+ (Fig 1.c). The BeAGLE is matching world-wide Sr Ge 45+ of cutting-edge research applications (Fig. 2.c). Pro Ta 5+ 32+ 0.1% Mo C 23+ state-of-art (Fig. 1.d). th 46+ As 22+ ps 37+ g W 24+ Ge 12+ Ag Se Ti 32+ 22+ en 22+ Ge 37+ Mo Ge

-L 19+ 5+ 22+ Fig 1.a 10+ 35+ Ag xFEL 32+ 20+ 22+ Ti Eu C 5+ Ge Na Mo 10+ Zn Ge 20+ 52+ 7+ 24+C 22+ m ain 8+ Stimulated O 10+Al Ge Zn 19+ 20+ m Au 42+ 35+F 10+ 11+ Al Se 19+ 22+ Cu G Si Ge Zn XRL 1% Y Eu Cu XRL ns 3d94d Emission 25 Al 5+ = C L 18+ rget Cu 3d94p Ta sma Synchr. Collisional SE Fig. 2.a HHG Pumping ASA Fast radiative Drive Beam relaxation Pulse duration 0.1mm Linewidth 3d10 r Fig 1.d Ni-like ion ase Brightness L UV X as 0.001% Fig. 2.c

Fig 1.b Driver Puls = c fs 0.01% TW v

) = 1.4c r e o W v T ps 0.1% Metrology Magnetic Wavefront 5° 1% 22. Plasma ns Semicond. (tilted at compress Nanostruct. Atto- θ = 45° vTW Catalysis Therm. science Target Biochem. Pulse duration Linewidth The Laboratory-Scale Light Source "BeAGLE" ResearchAll-Transmissive Highlight: Soft X-ray Design Microscopy Two-Concave-Mirror Setup Single-shot Imaging Contrast Mirrors Focal Length 120.6 mm Image NA = 0.008, M = 12x } Processing

CPA Nd:Glass Laser Wavelength 1054 nm Max Energy 20 J Pulse width 1.2 ps X-ray Laser Tilt Angle Wavelen. 12 nm EUV Source Comparison Brightness: Prepulses Generation The "Siemens Star" 11 Target Chamber EUV Laser Mo/Y Multilayer 10 EUV ph. Pressure <10-4 mbar 2 prepulses, 1 main pulse Reference Object in 1 ps, 3 mrad EUV Measured Incoherent Calculated

Conclusions Acknowledgements The X-ray laser is a bright lab-scale short-wavelength source The Swiss NSF funded most of the work along with matching suitable for a number of application. The repetition rate is the funds from industrial partners. Collaboration with the PSI is next challenge, to overcome limits of single-shot operation. acknowledged for the zone plate optics and nano-structured "Siemens star" samples. Future Work Industry collaboration is focused on the engineering of a EUV References microscope with nano-scale resolution. A 10-Hz EUV laser is D. Bleiner et al. Opt. Comm. 284, 4577 (2011) being integrated based on the OPCPA architecture. J. Nilsen et al. Opt. Lett. 28, 2249 (2003) 30 Hard X-ray Instrumentation Workshops 2011

Poster # 27 Inferring Networks from Distances: the “Landscape” of Glycosidase Protein Structures Sandhya Prabhakaran, Volker Roth Department of Mathematics and Computer Science, University of Basel, Switzerland

Graphical Abstract Markov Chain Monte Carlo Sampling

Objects: protein structures Contact maps Goal: infer sparse inverse covariance Ψ zeros determine graph structure. Network inference Method: Metropolis-within-Gibbs sampler on discretized set of inverse covariance matrices Ψ. Proposal Distribution: symmetric random walk in Ψ-space, Metropolis-Hastings acceptance rules. Compression−based distances Distance matrix Network inference Efﬁcient implementation: updated matrix O(n3) X Coder C(X) factorizations complexity per sweep.

Coder C(XY) NCD Compression Distance Y Coder C(Y) Normalized Information Distance (Li et al.,2004): metric which minimizes any admissible metric, proportional to the length of the shortest program that computes x|y as well as computing y|x. Bayesian Inference After proper normalization: Idea: construct probabilistic generative model, condition on {K (x|y), K (y|x)} K (xy) − {K (x), K (y)} observed data and infer distribution of parameters. NID(x, y)=max = min max{K (x), K (y)} max{K (x), K (y)} Assumption: Gaussian process with arbitrary column means S and S have same D but different coordinate origins! 0 Approximation: Normalized Compression Distance: we observe Euclidean distances D. n−dimensional Gaussian N(0,Σ ) C(xy) − min{C(x), C(y)} original mean−shifted NCD(x, y)= , 0.15 {C(x), C(y)} draw (i=1,...,d) +b i max 0.10 Distribution?

0.05 2 1 where C(xy) represents the size of the ﬁle obtained by −2 0 (u) −1 x x −1 i i D Sparse inverse cov. mat. 0 x y n 1 d d compressing the concatenation of and . −2 2

n n n In our application: x, y are vectorized contact maps computed

−1 t t Σ Xo Soo = (1/d) Xo X X S = (1/d) X X from all PDB files available from the group of Glycosidase enzymes in Escherichia coli. Distribution of D is singular Wishart, likelihood in rank-deficient Ψ˜ inverse covariance : ˜ ˜ d d ˜ ˜ t − t L(Ψ) ∝ (Ψ)2 tr(ΨD) , Ψ=Ψ− ( Ψ ) 1Ψ Ψ. det exp 4 1 1 11 Link to Gaussian graphical models: Ψ contains partial correlations between the nodes. Ψij = 0 (i, j) conditionally independent no edge. Advantage: large “zoo” of Mercer kernels (i.e. p.s.d. S) available. But: no information about origin only D is relevant! Use of specific S introduces significant bias!

Can use all kernel matrices to learn GGMs without geometric bias. We used the the ProCKSI-Server to compute NCD(x, y).

The “Landscape” of Glycosidase Protein Structures

3qxqA 3qxqD

3qxqC 2f2hA 2f2hD 2f2hF 3qxqB 1xskC2f2hB 1xsiD 2f2hC 2f2hE 1xskA 1xsiF 1xsjC 3c68B 1xskE 1xsjF 3c67B 1xsiA 1xskD 1xsjD 3c67A 3c69B 1xskB 1xskF 1xsiE 3qxfA 1xsiC 3c69A 3c68A 1xsjE 1xsjB 3qxfD 3qxfB 1xsjA 3d3iB

2wynD 2jjbD 1xsiB 2wynB 2wynC 3qxfC 2jjbC 2jg0A 2jjbA 2wynA 3d3iA

1q8fC 2jf4A 2jjbB 1q8fA 1q8fB 1q8fD 3mkmD 3mknD 1jz7A 3mv02 1jz7B 3mknA 3i3eB 3mknB 3mv01 1jz2A 3mv04 1jz7C 1jz3A 3mkmB 1jz7D 1px3A 1px3C 1yoeA 1px4A 3i3bA 3mkmA 3mknC 1jz2B 1jz5A 3g5iA 3i3bD 3czjD 3mv03 3i3bB 3mkmC 1px3B 1dp0C 1jz6B 3i3bC 3mv11 1px3D 1jz4B 3dymD 3i3eA 3g5iD 3g5iC 3i3eD 3i3eC3i3dB 3i3dA 1jz6A 3muz1 1jz2C 3mv13 1jz8D 1jz6C 1jz4D 3dyoC 3dymB 1jz5C 1jynC 1jz8A 3g5iB 1jz5B 1jyvA 1jz3B1jyxC 3muz2 3czjA 1dp0D 1jz4A 1jynA 3mv14 3i3dC 1jyvD 1jywA3dymC 3czjB 3muz3 1px4B3dypC 1jz2D 1jywD1dp0B 1jyxA 3dypD 1jynD 3dypA1jz8C 1jyvB 3iapA3czjC 1jz5D 1jz8B 3i3dD1jz4C3iapB 1jyvC3muz4 3dymA 1dp0A 1px4D 1jz3D 3dyoB 1jynB 3mv12 1muyA 3dypB 1jyxB 3dyoD 1jywC 1px4C 3iapC 1wefA1munA 1jyxD 3dyoA 3iapD 1kqjA 1mudA 1weiA 1jz3C 1jz6D 1jywB 1gpqC 1q3bA 1kg5A 1mugA 2wskA 1wegA 1k3wA 1mpgA 2ea0A 1kg6A 1mpgB 1kg7A 2opfA 1k3xA 1gpqD 1kg4A

1kg3A 1gpqA 1kg2A 3lpfA 1k82D 1gpqB 3lpfB 1k82B

1k82A 1k82C

4eugA

3eugA 5eugA

1eugA 2eugA

http://informatik.unibas.ch @unibas.ch SwissFEL X-ray Free Electron Laser 31

Poster # 28

Surface Diffraction Materials Science Beamline X04SA

Bragg CDI at an XFEL

Steven J. Leake

1Swiss Light Source, Paul Scherrer Institut, CH – 5232 Villigen PSI, Switzerland

What is Bragg CDI? Required Beam Parameters

Illuminating a finite crystal with a completely coherent beam of x-rays leads to a continuous Parameter Unit Requirement Motivation / Remarks interference pattern in the far field. The diffraction local to a Bragg peak describes the shape of the crystal. For an ideal (perfect) 3D crystal the distribution of intensity around each Bragg point Beam parameters keV 6.0-12.4 Energy will be the same, i.e symmetric. When atoms are moved away from their ideal positions, due to Stability % 0.1 % 0.1 defects, manipulation or processing for example, the intensity distribution becomes asymmetric; Bandwidth it is comprised of a symmetric component from the average electron density and an asymmetric Stability %bw <10.0 component from the displacement of atoms from their ideal positions projected onto the Q-vector Beam position Stability um <0.1 of the chosen Bragg point. From multiple displacement fields one can deduce the full 9 Beam size um 0.5-3 Knowledge of the illumination ideal #ph >107 component strain tensor[1]. Photons per pulse Stability % <10.0 I0 measure in principle irrelevant fs <200.0 Sample survival paramount

Pulse Length Pulse length should be approximately constant, Stability - pulse structure well known/monitored

Pulse arrival time Stability fs - N/A unless pump probe is considered Beam parameter changes during experiment range/step eV - Energy rate eV/min - Beam size (microfocus range/step um - only) rate um/min - Figure 1: A) Sample covered in microcrystals B) BCDI experimental geometry C) Phase retrieval method range/step um - Pulse length applied to 3D dataset D) measured diffraction pattern and reconstructed electron density rate um/min - Beam geometry Beam slope max. tolerable urad - N/A Why Bragg CDI? Working distance min. required mm 150.0 Sufficient to allow sample rotation/manipulation Other Required for coherent illumination of diffracting Transverse coherence FULL Materials with nanocrystalline morphology demonstrate vastly different properties compared to volume the bulk due to the strong influence of their surfaces, interfaces and defects. Understanding the strain present in these materials is required in order to tailor devices to exhibit desired properties. Bragg CDI is non-destructive, capable of isolating individual crystals, highly penetrating and high resolution. The development of XFELs offers the opportunity to extend Experiments BCDI to sub 100nm grain sizes with improved spatial resolution. The 100Hz operating frequency allows one to contemplate probing reversible processes on the ultrafast timescale, in pump-probe Synchrotrons mode, and irreversible processes on the timescale of several minutes. 2D/3D nanocrystal displacement fields Synchrotron sources to date

• Bragg CDI typically achieves a spatial resolution of the order 1/15th of each dimension of the crystal. The sensitivity to displacements is of sub-Angstrom order, ca. 1/40th of a unit cell[2].

• Flux density and scattering volumes limit the crystals to approx. 100nm, the available coherence volume and experimental setup defines the upper limit (several microns).

• Sufficient SNR (signal to noise ratio) is obtained between 0.5-2 hours of total exposure Periodic (Ga,Mn)As/GaAs wires [3] Zeolite Catalysts [4] ZnO microrod full strain tensor [1] (detector readout can increase this dramatically).

Pump-probe CDI for reversible processes, for example a laser pump x-ray probe of shockwaves • Small rotation is required for a full dataset, ~1°, for a 3D displacement field. induced in a crystal [5]. In-situ studies are emerging as a function of annealing [2], time [4,5] and other binary processes. Experimental Geometry XFELs

Enhanced resolution and extension to smaller crystals is expected.

SEM from X. Huang, APS Au nanocubes [6] Au Bipyramids and Rhombic Dodecahedra [7] Au Nanoantenna [8] Operating in a non-diffract and destroy mode. A 3D Bragg CDI dataset requires 50-100 steps in the rocking curve, i.e 1Hz is possible. On the timescale of minutes the sample state is approximately constant thus allowing grain growth, equilibrium crystal shapes [9], defect propagation [10] and charge density waves [11] to be studied, in-situ, amongst others. Ultrafast Bragg CDI requires reversible processes and a pump probe method to be adopted to extend beyond 2D projections.

Figure 2: Schematic of experimental set up for Bragg CDI at an XFEL Why not… Equilibrium crystal shapes [9] Crack propagation [10] Charge Density Waves in Cr [11]

3D Bragg ptychography: this could extend Bragg CDI to non-finite samples, however this has General Description/Criterion many associated technical problems. Monochromator – YES Energy Scanning – Maybe, alternative rocking curve Focussing – YES, 0.5-3.0 microns References Sample Chamber – YES, dof – annealing/cooling, evaporation, in-situ optical microscope [1] M. C. Newton et al, Nature Materials, 9 120-124 (2010) Sample Holder – kinematic sample mounts [2] S. J. Leake et al New J. Phys. 13 113009 (2011) [3] E. Dufresne et al, NIM A (2011) Detectors – min 1024x1024 40micron pixels – the smaller the pixels the better, adjustable [4] A. A. Minkevich et al., Phys. Rev. B, 84 054113 (2011) (automated) sample to detector distance for alignment, trajectory scanning capable movement. [5] W. Cha et al. New. J. Phys. 12 035022 (2010) Synch. Source of pump radiation – YES [6] J. Zhang et al., J. Am. Chem. Soc., 132 14012–14014 (2010) [7] M. L. Personick et al. J. Am. Chem. Soc., 133 6170–6173 (2011) Other bulky equipment – NO [8] M. W. Knight, et al. Science 332, 702 (2011) Beam diagnostics – Intensity vs time. Reproducible pulse, i.e seeding ideal. [9] J.C. Heyraud and J.J. Metois, Acta Metallurgica, 28 12 1789-1797 (1980) [10] R. Meyer et al., Acta Metallurgica, 25 10 1187-1190 (1977) Optics - Illumination function at sample [11] O.G. Shpyrko et al., Nature, 447 (2007)

*contact address: [email protected] 32 Hard X-ray Instrumentation Workshops 2011

Poster # 29

Serial Femtosecond Crystallography on tiny 3D crystals

J. Steinbrener1,2, L. Lomb1,2, T.R.M. Barends1,2, S. Kassemeyer1,2, A. Jafarpour1,2, R.L. Shoeman1,2, S. Bari3,2, S.W. Epp3,2, L. Foucar1,2, A. Rudenko3,2, D. Rolles1,2, F. Krasniqi1,2, R. Hartmann4, L. Strüder5,2, J. Ullrich3,2, I. Schlichting1,2

1 - MPI für medizinische Forschung, Heidelberg, Germany; 2 - Max Planck Advanced Study Group at CFEL, Hamburg, Germany; 3 - MPI für Kernphysik, Heidelberg, Germany; 4 - PNSensor GmbH, München, Germany; 5 – MPI Halbleiterlabor, München, Germany

Introduction The elucidation of structures of macromolecules is an important step in the quest of understanding the chemical mechanisms underlying biological function. X-ray crystallography is a mature method that is only limited by the quality of the crystals investigated and by radiation damage. Intense, femtosecond X-ray pulses provided by X-ray free-electron lasers promise to break the nexus between radiation damage and crystal size, thereby allowing structure determination using nano- and microcrystals. Recent serial femtosecond crystallography (SFX) experiments at the LCLS have shown the feasibility of this approach1. A continuous liquid microjet was used to inject randomly oriented crystals into the FEL beam2,3. The diffraction patterns were collected in vacuum at the repetition rate of the FEL in the CAMP4 or CXI instruments5 using pnCCD or CSPAD detectors, respectively. Due to the mismatch between continuous sample flow and stroboscopic data collection, sample consumption is huge. FEL-triggered drop-on-demand approaches have been proposed and are being explored3. For very precious samples, other possibilities need to be explored which include preparation on fixed targets and cryo-stages, which are ideally integrated into dedicated endstations with appropriate detectors. Since the crystals intersect the FEL beam very fleetingly, only thin slices through the rocking curve are recorded, requiring many measurements of the reflections to allow a Monte-Carlo like integration of the beam profiles6,7. A pink or Laue beam has not only more flux than a monochromatic beam but is also more efficient in sampling reciprocal space. A shot-to-shot analysis of the spectrum would be highly desirable, for example to allow accurate profile fitting including coherent diffraction features as would be the availability of a divergent beam that can be matched to the sample size.

Typical Experimental Setup Reducing Sample Consumption Fixed cryo targets Data collection pnCCD/CSPAD detector -Crystals on wafer in random orientations -Synchronized fly scan (translation ~ 10 x crystal/beam size between pulses) Detectors: -Cooled stages and cold shield8,9: pnCCD4: four 512 x 1024 panels Copper anti-contaminator 75 μm pixel size CSPAD: 1520 x 1520 pixels 110 μm pixel size Fiberglass support Stainless steal support

Aluminum support Interaction volume: liquid jet single nano- Ystage crystal Xstage ddbeamuu jetmin dd beam , jet Typically 1 x 1 x 5 m3 ȝ Huang, X. et al., Nucl. Instr. Meth. A 638, 171-175 (2011).

LCLS pulse parameters: Pink-Laue SFX Ȝ =2-1.3 Å, 6-9.4 keV Gas focused liquid jet2: 5-40 fs pulse duration (e-beam) ~ 5 ȝm diam., flow rate ~10 ȝl/min, ~10 m/s 120 Hz Hit rate = concentration x interaction volume ~ 3 mJ/pulse Effective hit rate smaller (multiple hits, hit -1-3% bandwidth 1011 –1012 photons/pulse finding issues, indexing efficiency) -Better data collection efficiency -Better SNR Îreduced sampling requirements Liquid jet sample consumption ÎNeed to know spectrum for each shot patterns neededu flow rate sample consumption ÎTypically > 1ml of 109 ml-1 hit rateeff u rep rate Split beam interaction volumeu rep rate sample efficiency ÎTypically 10-6 -Cover 0.1º to 1º flow rate -Better data collection efficiency ÎAlternative sample delivery schemes needed! -Better SNR Îreduced sampling requirements ÎNeed to know intensity for each beam Data collection and analysis Problems: -Different sizes of crystals -Random orientations -Thin slices through Ewald sphere Suggested Beam Parameters -Shape-transform effects

ÎMonte-Carlo-type integration is possible6: Required snapshots for high resolution > 104 Upper font detector

Lower font detector

010010

100

20 Å

Chapman, H. N. et al., Nature 470, 73-77 (2011).

References 1. Chapman, H. N. et al., Nature 470, 73-77 (2011). 2. DePonte, D. P. et al., J. Phys. D: Appl. Phys. 41, 195505 (2008). 3. Weierstall, U., Doak, R.B., Spence, J.C.H., arXiv:1105.2104v1 [physics.ins-det] (2011) 4. Strüder, L. et al., Nuclear Instruments and Methods in Physics Research A 614, 483-496 (2010). 5. Boutet, S. and Williams G. J., New Journal of Physics 12, 035024 (2010) 6. Kirian, R. A. et al., Optics Express 18, 5713-5723 (2010) 7. Kirian, R. A. et al., Acta Crystallogr. A 67, 131-140 (2011) 8. Huang, X. et al., Nucl. Instr. Meth. A 638, 171-175 (2011) 9. Schneider, G. et. al., IPAP Conf. Series 7, 349-352 (2005) SwissFEL X-ray Free Electron Laser 33

Poster # 30 34 Hard X-ray Instrumentation Workshops 2011

Poster # 31

NUCLEAR ENERGY AND SAFETY

Dynamics of ion diffusion in clays: a proposed probe - probe XPCS experiment S. V. Churakov1, R. Dähn1, A. Froideval1, B. Pedrini2

1 Nuclear Energy and Safety (NES), Paul Scherrer Institute, 5232 Villigen PSI, Switzerland 2 SwissFEL Project, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland

Context of the study Motivation of the study

ƒ Deep geological repositories: Understanding of: safe long-term containment of the radioactive waste ƒ the ion diffusion (Na+, Cs+) mechanisms – timescale: ~ 10 - 100 ps – in the interlayer of clay minerals in order to increase our confidence in numerical models. ƒƒƒ Safety barriers in a repository for high-level waste: Is the ionic transport in the interlayer governed by non-Brownian diffusion 1. glass matrix containing radioactive material engineered mechanism(s)? barriers 2. steel container 3. backfill with bentonite MD simulation* = clay minerals geological 4. host rock barrier Cs

Na Main characteristics of the clay barrier:

ƒ Dimensions: Ow in thein interlayer (a.u.)

• bentonite: 70 cm thick (taken from [1]) Ion andwater distribution = montmorillonite (75 %) + quartz / calcite ~ 2 nm -4 -2 0 2 4 Material: montmorillonite z [Å] • host rock: 100 m thick * from S. Churakov (LES / NES / PSI))

ƒ Low hydraulic conductivity: 2×10-14 m/s Types of mobile ions / molecules:

+ + mass transport by diffusion Cs Na H2O

Proposed experiment ƒ Sketch of the experiment ƒ Key parameters of the experiment

ƒ Proposed PROBE-PROBE X-ray photon correlation spectroscopy (XPCS) Parameter Unit Requirement Motivation / Remarks experiment: ƒ K e y p a r a m e t e r s of t h e X - r a y F E L r a d i a t i o n Energy keV > 10.0 required by X-ray split-and-delay unit streak microscope detector camera (sample visualization) stability % < 0.05 required by X-ray split-and-delay unit scattered Bandwidth % 0.1 / 0.01 1 nm / 1 Å resolution splitted beam stability % < 10 stable incoming beam is wished XFEL pulses XFEL pulse Position stability --- ¨t μm ~ 1 microfocus beam Size Photons / pulse μ 2 stop #ph ~1011 taking maximal fluence of 0.5 mJ/ m and assuming I 1% transmission of the split-and-delay unit 0 temperature data to be normalized with I monitor stability % < 10 0 X-ray split-and-delay unit[2] control Pulse length fs < ~ 200 mechanisms of ion diffusion on (few) tens ps timescale stability % 10 Pulse arrival time --- sample ƒ B e a m p a r a m e t e r c h a n g e s d u r i n g e x p e r i m e n t sample 6-axis nanopositioning (possibly radioactive) Energy range / step eV variable photon energy desirable (for resonance holder studies) e.g. at Na K-edge and at Cs L -edge; in the sample manipulator rate % --- III future: additional studies foreseen at LIII-edges of (hexapod) actinides (up to 20 keV) d § 0.3 m Size range / step μm--- rate μm/min --- dsample-detector § 1 m laboratory C Pulse length range/step 100 - 500 fs test at values for consistency check

nd (1 A 100 LA) (LA: licensing limit (specified in Annex 3 Column 10 of the “Strahlenschutzgesetz, March, 22 , 1991)) rate --- no scanning ƒ B e a m g e o m e t r y ƒ Expected results Slope max. μrad --- Working min. mm ~ 300 space for sample holder ƒ in situ measurements of the ionic mobility in the interlayer of clay distance minerals; ƒ O t h e r ƒ ion transport mechanism(s) in the interlayer space, namely Brownian Delay line yes split-pulse XPCS (delay: up to 3 ns with hard X-rays) versus non-Brownian diffusion mechanisms. Transverse coherence yes required for coherent X-ray diffraction for similar future studies on radionuclides ƒ experimental validations of MD simulations Laboratory C yes (1 A 100 LA, with LA: licensing limit) ƒ scientific basis for performance assessments of the clay barrier in the deep geological disposal of radioactive waste.

References Acknowledgments [1] from http://www.nagra.ch/, NAGRA: National Cooperative for the Disposal of The authors wish to thank B. Patterson, E. Wieland from PSI, and G. Grübel from DESY for helpful Radioactive Waste. discussions and interest in the work. [2] G. Grübel et al., Nucl. Instr. and Meth. in Phys. Res. B, 262 (2007) 357-367. SwissFEL X-ray Free Electron Laser 35

Poster # 32

Selected Applications on Data-Intense Research

by David Müller and Christof Bühler

Abstract/Rational ChallengesatSwissFEL SCSCompanyProfile(www.scs.ch) • Modernscience,engineering,businessesareconfronted • Datarates/volumes:10Ͳ 100TeraByte/day/detector • foundedin1993byProf.Dr.AntonGunzinger withmassivegrowthofdata(1015 –1018 Bytesperday). • Throughput(bandwidth)andlatencybottlenecks: • providesdevelopmentservicesforindustrialandacademic • Imposedchallengesencompassprocessingspeed,storage readoutÖ preͲprocessingÖ storage Ö postͲprocessing customersinSwitzerland,Europe,andUS capacity,power,cooling,andespeciallyhumanfactors: • Diversedetectors:BPM,PixelDetector,CMOS/CCD cameras • |75employees: electronicandsoftwareengineers, Whilethepowerofinformationsystemsincreases–theamountof physicists,andmathematicians • Numberofdetectors(channels) informationahumancandirectlydigestdoesnot. • Applicationfields:HPC,embeddedcomputing,intelligent • Diverse data formats • Keytodiscoveriesisdatareductionthatmustaimat sensors,lifescience,enterpriseapplications • RAM = Reliability, Availability, Maintainability providinggoodcluesfortherightquestions. • KnowHow:HW,SW,algorithms,systemdesign

COSMODynamicalCore SubjectofResearch COSMOisalocalscaleatmosphericmodelforweatherforecastandforclimateresearch.Its dynamicalcorecomputesthenonͲhydrostaticcompressibledynamicalequations;itisoneof themainworkloadsforsupercomputersinSwitzerland.TheCOSMOdynamicalcoreisvery dataͲintenseandrequiresoptimizedusageofthememorybandwidthavailableinthesystem. Experimentor SCSisreͲwritingthedynamicalcore SimulationCode • Achievehighestperformance • Achieveperformanceportabilityonx86andGPUarchitectures • DSEL(DomainSpecificEmbeddedLanguage)basedonC++

ThisongoingworkisexecutedincollaborationwithC2SM,MeteoSwiss andCSCS. PrecipitationforecastcalculatedbytheCOSMOmodel Anothertaskfocusesondatareductionandenhancedarchivingstrategies. onagridresolutionof7kmx7km DataGeneration FPGAͲacceleratedImageProcessingforSwissFEL DiagramofcamBoxPrototypeSystemͲ DesignandlogicArchitecture FortheSwissFELFreeElectronLaseratPSI(Villigen)annovelFPGAͲ acceleratedcamerasystemisdevelopedthat Readout • recordsandanalyzesbeamprofilesataframerepetitionrateч100Hz • providesagenericcamerainterface(initially:2PSIͲselectedGigEcameras) • isfullycontrolledandoperatedbyEPICS • issynchronizedbytheSwissFELEvent&Timinglogic • correctsforsalt&peppernoise,cosmics,offset&darklevels,gamma DataPreͲProcessing function,camerarotations(d 5°,+/Ͳ 90°,180°) • calculatesROIͲbasedintensityprofilesalongcamera’sxͲ andyͲaxis • deriveskeybeamparameters:centerposition,width(FW%M)viadifferent methods(e.g.CoM,curvefitting) camBoxPrototypeDevelopmentPlatformusingCOTScomponents

• isimplementedasprototypeoncommercialcard(Virtex5card,AlphaData) PMC card: Virtex5 ADC-VXS: FPGA VME-carrier board Filesystem Write andthenmigratedtotheproductionsystem. (AlphaData) XRM I/O card: JointprojectofPSIandSCSstartedin2011. GigE interface

FastTrackTrigger(FTT) ScientificTarget:HighlyEnrichedEventSelectionforD*Decays TheFTTreducestheeventrateinaparticlephysicsexperimentbeforethedataisread Storage outofthefrontendelectronics.TheFTTreconstructparticletracksandinvariantmasses. • UsedattheH1detector@HERA2000 • LinkingofparticletracksegmentsusingContentAddressable Memory(CAM)atfullinputrate(10MHz)inFPGA Filesystem Read • ReconstructionofinvariantmassesinDSPsoftware JointprojectofETHZ,Universität DortmundandSCSin2001. TriggerRack

NETEZZADataMiningAcceleration Netezza – Highestperformancedataselectionengine©byNetezza DataPostͲ TheIBMNetezza DataWarehouseAppliancefocusesonmassiveparallel Processing/ processingofhugedatasets.SearchͲoperations(project/‘SELECT’andrestrict/ Visualization ‘WHERE’)areimplementedinFPGAhardware.Theappliancethereforecan processPetabyteͲscaletablesasfastastheystreamfromtheunderlyingparallel disksystem. SCSisadevelopmentpartnerofNetezza.

MPIͲ &GPUͲacceleratedImageProcessingforQuantiative Biology PSpecDPͲ SupportedWorkflow NewapproachesindrugdiscoveryfocusoncellularimagingoftendenotedashighͲ PSpecDPmainwindowandresultpanels Forecast contentscreening(HCS).CurrentHCSplatformsprocess50,000Ͳ 200,000images/day, generatehundredsofGB/day,andrequireextendedprocessingtimes(hours).The DataAnalysis developedprogram‘MicroͲSpectroscopyDataProcessor’(PSpecDP)providesfast

(HEP) andfullyautomatedanalysisofcellularstructuresandmolecularsignalsfrom

Weather thousandsofmicroͲspectroscopicimages.

& ThesoftwarePSpecDPfullysupports Diagnostics Analysis • dataprocessingworkflow:import,analysissetup,verification,processing,export Physics

• interactiveandautomatedanalysisofcellularstructures(nuclei,membranes, speckles,…)andmolecularsignalsfrommicroͲspectroscopicimages(vectordata) 1) MPIͲbaseddataprocessingonmultiͲ 2) GPUͲacceleratedimage Beam Experiments coreand/or networked clusters processing(filters,segm.) Science Analysis • intuitive visual tools ensuring seamless data processing

Network • two stage parallelization techniques Energy Ͳ

Science 1)MPI(MPJexpress) Ö multicoreand/orclustermode(SPMD) Results 2)NVIDIA'sCUDA Ö coreimagealgorithms(SIMD) Climate High Market SwissFEL SwissFEL Social

Life JointprojectofNovartisandSCSstartedin2010.

Acknowledgments:H.Braun,R.Ischebeck,T.Korhonen,D.Anicic,B.Patterson,R.Abela(PaulScherrerInstitut,Villigen,Switzerland) 36 Hard X-ray Instrumentation Workshops 2011

Poster # 33

S. L. Johnson

Possibilities for nonlinear x-ray scattering with SwissFEL

Poster follows SwissFEL X-ray Free Electron Laser 37

Poster # 34

2D Membrane Protein Crystal Diffraction

1Ching-Ju Tsai, 2Bill Pedrini, 3Cameron M. Kewish, 2Bruce Patterson , 2Rafael Abela, 1Gebhard Schertler, 1Xiao-Dan Li SwissFEL 1Laboratory of Biomolecular Research and 2SwissFEL Project, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland 3Synchrotron Soleil, L’ Orme des Merisiers, BP 48 Saint Aubin, 91192, Gif-sur-Yvette cedex, France

2D X-ray crystallography X-ray structure determination of macromolecules 1. 2. and ptychography

qy Synchrotron X-FEL χ = 0 χ q 0 “radiation damage “diffract and y μ limit” > 0.5–10 μm 2D “Bragg” destroy” peaks

> 10 μm

2D membrane protein C.K. qx qz 3D crystallography crystallography sample orientation known sample orientation unknown 3D single-molecule Reciprocal Ewald H.C. nanocrystallography diffraction spheres space

0.2–1 μm “Traditional” 2D crystallography

> 1–80 nm J. C. Falkner et al., Chem. Mat. 17, 2679 (2005)

… decreasing scattering power

4 3. 2D X-ray crystallography experimental setup

200 - 500 mm ~ 200 mm 400−1000 mm λ = 1 Å 2 200 nm focus 50 μm detector pixel

1 3 θ max ~ 30º Pixel detector 2 Å res. @ λ = 1 Å (low-flux) Focusing SAMPLE optics 2D crystal SAMPLE qx ptychography qx Pixel detector X sam (high-flux) χ Y sam Single shot 100 nm spectrometer φ (eventually) Quadrant intensity / qx qx beam position monitor Z X,Y Vacuum chamber

Microfocus Diffract and destroy acquisition sample positioning coherent pixels of diffraction images χ (ptychography) SAMPLE Y Effectively Ysam sam 2 Focusing 2D unit overlapping beam optics cell 4 illumination areas

SAMPLE Xsam Classification of illumination areas + ultra small focus longitudinal 3 2D ptychographic (ptychography) coherence (ptychography) phase retrieval algorithm

Parameter Unit Requirement Motivation / Remarks 1 Beam parameters Energy keV > 8 Diffraction at biomolecules stability 0.05 % If not possible: single shot spectrometer Bandwidth % 0.05 0.2 nm resolution @ λ = 1 Å Ptychographic reconstruction of aquaporin (simulated data) C.K. stability < ±10 % bw Beam position stability < 0.5 μm 2D acquaporin crystal original reconstruction 0.2 / 1−2 high-photon Beam size μm Few unit cells illuminated (microfocus) number Photons per pulse #ph > 1011 Maximal flux wished stability < ±10 % Beam area Pulse length fs < 50 fs Diffract and destroy regime required stability --- Pulse length should remain < 50 fs Pulse arrival time stability fs --- Not an issue Beam parameter changes during experiment Energy range / step eV --- Energy adjusted only during setup rate eV / min --- Beam size range / step μm --- Focus size adjusted only during setup (microfocus only) rate μm / min --- range / step --- Pulse length adjusted only during setup Pulse length rate --- Beam geometry Unit cell Beam slope maximum tolerable mrad --- Not an issue

Working distance minimum required mm 100 Allow sample rotations and translations Other References Transverse coherence full For coherent diffraction at extended object transversal coherence H.C. H. N. Chapman et al., Nature 470, 73-78 (2011) (ptychography) C.K. C. M. Kewish et al., New J. of Phys. 12, 035005 (2010) 38 Hard X-ray Instrumentation Workshops 2011

Poster # 35

Energetics and Solvation Dynamics

of the excited Ru(bpy)3 complex in water

Jaroslaw J. Szymczaka, Franziska Hofmanna and Markus Meuwlya

aInstitute of Physical Chemistry, University of Basel, Klingelbergstrasse 80, 4056 Basel, Switzerland. e-mail: [email protected] Efficient MD simulations of inorganic chromophores such as pyridinated ruthenium complexes were carried out to study the behavior of water solvation shells. After electronic excitation of an pseudo- octahedral ruthenium complex in explicit solvent, a reorganization process of the corresponding solvent can be obtained, which is mainly related to the solvent electronic interactions.[1] The geometry of the complex is flexible during the MD process in order to study the internal motions in the time-dependent solvation process. The structure of the solvation shell and the reorganization time can be obtained by MD simulations and TD-IR spectroscopy derived from correlation function of the dipole moments.[2] Using the VALBOND TRANS[3] force field with different water models it was verifies experimental data gained by spectroscopy with modern laser techniques.[4]

The progress in ultrafast spec- Goal The reorganization process of solvent (water) Motivation troscopic techniques brought after excitation is studied to capture the time- new insights about structural changes on the sub- dependent solute-solvent interaction and the solvent re- picosecond scale. Experimental findings alone, sponse which includes the energy transfer from the metal however, are not able to explain the complex and complex to surrounding solvent. The time scales of the heterogenous dynamics.[3] Therefore MD simula- energy transfer process as well as signal decay time are es- tions are the method of choice helping better under- timated from non-equilibrium MD simulations. The meas- standing of related processes. Ruthenium(II)-tris- ured quantities involve radial distribution function, kinetic (2,2’-bipyridine) is known to be photostable, also in energy of water and changes in the theoretically obtained [5] its excited state, therefore Ru(bpy)3 plays an im- IR spectrum recorded during the relaxation of the photo- portant role in solar-energy research.[6] These fea- excited complex. The obtained information is expected to

tures of Ru(bpy)3 made it to the complex of our be helpful in answering the question if there is an energy choice. transfer to the surrounding solvent or if energy is rather re- Energy transfer to the solvent expressed in terms of ki- distributed to the low frequency modes as suggested by Theoretical setup netic energy of a single water molecule Chergui and co-workers[9] VALBOND TRANS (VBT) force field was used for MD simulations with TIP3P[7] and flexible KKY[8] water models. For TIP3P setup a 1fs time step was used and SHAKE for constraining the hydrogen atoms, whereby for the KKY flexible water model a timestep of 0.2 fs was used. Verlet dynamics (NVE) were carried out at 300K in explicit water (1694 water molecules). The excitation has been modeled by instantaneous change of force field parameters for the ruthenium complex from Ru(II) to Ru(III). Both force fields were first reparameterized to the data obtained from ab inito calculations, with TRANS parameters adjusted according to the previous work in our group.[3] Results VALBOND TRANS (VBT) force field is used the first time for applications in spectroscopy, were an instantaneous excitation of ruthenium was investigated and the magnitude of energy transfer which occurs to the surrounding water shells. Since changes in the signal recorded for all the water is week we concentrated on the water closest to Ru complex that would stay within the shell of 11A from the centre of the complex. Considering the RDF, usage of an 11 A radius was reasonable to make sure that the first three water-shells around the transition metal were taken into account. After excitation, an increase in kinetic energy of 0.04 kcal/mol per water molecule is determined, having approximately 120 water molecules within this radius makes an overall energy increase of almost 5 kcal/mol. This intermolecular charge transfer occurs in 40-50 fs, followed by a decay of this signal within 400 fs. III 3+ Classical IR spectra of [Ru (bpy)3] averaged over 100 trajectories evaluated at different time after excitation. Radial distribution function of solute-solvent Spectrum is obtained from the Fourier transform of the distances recorded for before and after Kinetic energy of a single water after ex- dipole-dipole correlation function. Black: total spectra excitation. citation and in the equilibrated state, val- after excitation, red: spectrum after 0.5 ps and green: Energy transfer and RDF were also deter- ues averaged over 100 trajectories. spectrum after 1.0 ps. The inset is a amplification of the mined for the rigid TIP3P water model, but to generate an IR spectra, it was necessary peak at ~3600 cm-1. to use flexible water model. The classical IR spectra is calculated from the time- dependence of the dipole moment function. According to experimental data the O-H symmetrical and asymmetrical stretches lay around 3600 cm-1 and the bending mode around 1600 cm-1 and the out of plane asymmetrical stretch below 700cm-1. By comparing the full spectra and “late spectra” (determined 0.5, 1.0, 1.5 ps) one observes no shifts in positions of the peak, but there is a visible change in amplitudes of the peaks. To verify our findings we performed simulations with artificially large charge changes, to show the same effect, but more intensive. The most pronounced change in quantities, that we can directly obtain from MD simulation, is the energy transfer to the solvent manifesting in raise of kinetic energy of water. At the current state it is a open question how this change would be seen by the experiment. Conclusion References: ValBond Trans force field has been proven to perform well giving qualitatively good decay signal times [1] M.-E. Moret et al. Chem. Eur. J. 2010, 16, 5889 [2] M. Meuwly et al. Phys. Chem. Chem. Phys. 2003, 5, 2663 comparable with predicted decay values available from QM/MM calculations (300ps).[1] On the con- [9] [3] I. Brohman et al. J. Chem. Theory. Comput. 2009, 5, 530 trary to suggestion in previous work , we observed a significant energy transfer from the excited tran- [4] F. Ingrosso et al. J. Phys. Chem. B. 2005, 109, 3553 sition metal complex to the surrounding water within 40-50 fs, decaying within 400 fs. The IR spectra [5] S. H. Wadman et al. Organometallics. 2010, 29, 1569 of the full trajectory compared to the “late spectra” registered within 0.5 ps delay periods after excita- [6] W. Gawelda et al. J. Am. Chem. Soc. 2007, 129, 8199 tion, shows differences in magnitude of the intensities of the peak; however, no shift in frequencies has [7] W. L. Joergensen et al. J. Chem. Phys. 1983, 79, 926 been observed. [8] N. Kumagai et.al. Molecular simulations. 1994, 12, 177 [9] A. Cannizzo et al. Angew. Chem. Int. Ed. 2006, 45, 3174

Paul Scherrer Institut, 5232 Villigen PSI, Switzerland Tel. +41 (0)56 310 21 11, Fax +41 (0)56 310 21 99 www.psi.ch