PRECISE MAGNETOMETERS ON BASE OF PULSED NMR TECHIQUES G.Karpov, A.Medvedko, E.Shubin, Budker Institute of Nuclear Physics SB RAS, Novosibirsk, Russia. Budker INP, NovosibirskRUPAC 2006

Budker INP, NovosibirskRUPAC 2006 Precise Magnetometers on base of pulsed NMR techniques In a new generation of precise Magnetometers on base of pulsed NMR techniques has been developed and fabricated at BINP. Main features: 1) High precision: field measurement error less than 10 -5, resolution is of order (for homogeneous fields) 2) Wide field range: 0.025÷13 T 3) Large field range provided with one probe: B MAX /B MIN is up to 30 4) Acceptable field gradients: grad(B)/B up to 5×10 -4 /mm 5) Multichannel measurements 6) Small probe dimensions: minimal probe thickness is ~1 mm 7) Small probe sensitive volume – up to 1 mm 3 8) Capability of measurements at liquid helium temperature 9) VME standard

PULSED NMR TECHNIQUE USED IN BINP MAGNETOMETERS ТТ Spin echo signal FID signal Larmor frequency of nuclear precession, where γ – gyromagnetic ratio Fourier-transformation F NMR = ω L0 /(2π), где ω L0 – precession frequency in the centre of probe sensitive volume If RF pulses frequency F RF is close to precession frequency F L (F RF - F L < 1/T 90 ) free induction decay (FID) signal and spin echo signal with precession frequency are appeared. B – magnetic field value

Functional diagram of NMR Magnetometers

Digital signal processing Obtained in the magnet ЕМ3.S of VEPP-4М NMR waveform envelope and weighting function Purpose: finding of the module and sign of difference frequency F D = F NМR – F SYNТ Signal-to-noise ratio in frequency domain: where E S – signal energy, S N – noise spectral density Matched filtering: signal array before Fourier transformation is multiplied on matched weighting function.

Two modes of operation 1) Search mode: scanning with Synthesizer frequency in specifying range and finding of coarse value of NMR frequency 2) Measurement mode: precise measurement of NMR frequency; Synthesizer frequency is close to NMR frequency and follows to it

During the measurement cycle signal accumulation is performed: summing of N registered NMR signals (digit arrays). It follows to signal-to-noise ratio increasing in times. Simultaneously with signal accumulation digital signal processing is performed. During the cycle M digital processing of the signal accumulated in cycle M-1 is performed. Measurement cycle structure Time duration of the elementary cycle T C = 20÷100 ms Number of accumulated signals N = 10÷100

Structure of the Magnetometers Control System VME crate controller Magnetometers microcontroller VME bus ETHERNET Magnetometers functional parts OS: LINUX or VxWorksHigh level Low level

NMR probes wide range NMR probe Resonant NMR probes Miniature NMR probes – with thickness up to 1 mm. Probes with small sensitive volume – up to 1 mm 3. Probe Head electronics One of NMR Probe design

Parameters of working substances Working substance Working nuclei Gyromagnetic ratio (with chemical shift) γ/(2π), МHz/Т Spin-lattice relaxation time Т 1, ms Spin-spin relaxation time Т 2, ms waterprotons rubberprotons ~20 (В=0.5Т)~0.7 Lithium salts solution 7 Li nuclei and protons ~1000 heavy waterdeiterium ~1000 Metal aluminium powder 27 Al nuclei ~30 (T=4.2K)~0.04 Teflon 19 F nuclei ~40~0.09

Dead time of the Magnetometers receiver circuit. Two modifications of NMR Magnetometers: ModificationDead time T D Range of NMR signal duration 1~100 μs0.5÷100 ms 23÷5 μs10÷1000 μs The problem of dead time decreasing is connected with large difference between RF pulses amplitude (~20V) and NMR signal amplitude (a few µV).

Receiver module

12-channel Probe Multiplexer

12-channel Magnetometer in VME standard

Error of absolute field measurements Error of relative field measurements Error due to space uncertainty of measurement point (~10 -1 ΔB/B) Error due to uncertainty of the gyromagnetic ratio and field distortions caused by diamagnetism or paramagnetism of the working substance ~(2÷10)×10 -6 Error caused by instability of the signal spectrum shape ( ~ ΔB/B) Random error caused by noise of the Receiver circuit (defines Resolution of meas.) Error caused by instability of the Synthesizer frequency ( ~ / °C)

RMS frequency measurement error, caused by noise Width of the signal spectrum: where ΔF 0 – working substance NMR linewidth T 2 – spin-spin relaxation time of working substance ΔB – field dispersion within probe sensitive volume γ – gyromagnetic ratio where – signal-to-noise ratio in frequency domain, ΔF S – width of the signal spectrum.

Dependence of σ F on the width of NMR signal spectrum Minimizing of the signal spectrum width: 1) NMR probe has to be set in the region with most homogenous field 2) In some cases – using of the compensating coils 3) Optimal probe sensitive volume ( corresponds to condition: 1/T 2 k · ΔB·γ/π ) Resolution of measurements of the field with grad(B)/B less than /mm in the range 0.1÷13 T is of order /Hz (for the probes on base of aluminium is of order /Hz ) NMR signal spectrums obtained with the same probe at the fields with different gradient Grad(B) 1 0.1·grad(B) 2

NMR Magnetometer for electron-positron storage ring VEPP-4 (BINP) Magnetometer is used for field measurements of four different bending magnets in the ring and one additional magnet (magnet H) connected in series with main bending magnets of the ring. MagnetHNEM.1SEM.1SIM.3EM3.S Field range, Т Field range of VEPP-4 magnets

FID signal in time and frequency domain obtained in magnet H on the field Т ( beam energy МeV). Relative width of signal spectrum: ΔF S /F Signal-to-noise ratio in frequency domain for measurement time 1 s: S/N 500 Resolution of measurements σ F /F NМR /Hz

Results of field monitoring of the magnet H (VEPP-4) Field of the magnet H before experiments Time of one measurement ~2 sec σ F /F NМR 5×10 -8 Field of the magnet H during experiments

Fast NMR measurements at VEPP-4 Time of one measurement ~50 ms (without signal accumulation) Field measurement results of magnet H σ F /F NМR 4×10 -7

NMR Magnetometer for electron-positron collider VEPP-2000 (BINP) NMR probes are set at all eight bending magnets of the ring. Main problem: Large field gradient within the probe volume – up to 120 G/cm at maximal field ~2.4 T (Beam energy ~1 GeV). Position of NMR probe in the gap of VEPP-2000 bending magnet: 1 – magnet coil, 2 – vacuum chamber, 3 – NMR probe. NMR probe with rubber working substance sample 0.3×1.2×3 mm Width of the signal spectrum in the field range 0.3÷2.4 T is (1÷2)×10 -4 Resolution is better than /Hz

NMR Magnetometer for electron cooling facility (Fermilab, USA) Layout of electron cooling facility NMR Magnetometer is used for field stabilization of ten bending magnets of electron transport system. ParameterUnitsValue Magnet type–45-degree90-degree Number–82 Nominal fieldG Field rangeG Layout of the electron cooling facility Bending magnets of the electron transport system

NMR signal spectrum in the field 320 G of the 45-degree magnet (F NМR = 1.36 МHz) Parameters of the signal spectrum: Relative width ~10 -4 Signal-to-noise ratio (measurement time is 1 sec) ~ 250 Resolution: σ F /F NMR /Hz. Maximal error of relative field measurements: – for year

Field stabilization on base of NMR magnetometer Results of field stabilization of 45-degree bending magnet (obtained with another NMR Magnetometer) At present all 10 bending magnets of the transport system are stabilized. Long-time field stability is ~10 -5 and limited by 16-bit Power Supplies DACs.

Position of NMR Magnetometer at MI-31 (Fermilab) VME crate with NMR

Field measurements with NMR magnetometer at liquid helium temperature One of the tasks is magnetic field measurements in the superconducting 3-pole Wave Length Shifters (WLS). Task: measurements of the WLS central pole field with precision of order Field along WLS central axis measured with Hall probes Position of NMR probes inside the WLS. Field range: 3÷7 T for BESSY-2 WLS and 5÷10 T for SpRing-8 WLS.

NMR Probes for WLS Design of NMR probes assembling. Assembling of the pair of NMR probes. Probe working substance – aluminium powder, sensitive volume ~2 mm 3. Assembling of pair NMR probes is set in the central pole axis in the hole of 6mm diameter.

NMR signal at the WLS central pole field ~6 Т (F NМR 58 МГц ) Signal-to-noise ratio ~ 600 Relative width of the signal spectrum Resolution of measurements ~(1÷2)×10 -6 /Hz Nuclear magnetization M N is proportional to 1/kT, where T – is temperature. At liquid helium temperature (T4.1K) M N is increased in 70 times compare with room temperature. Resolution of measurements of WLS field

In ten NMR Magnetometers on base of pulsed techniques have been fabricated and applied at charge particles storage rings and field measurements stands at: BINP (Russia) Fermilab (USA) Duke University (USA) BESSY-2 (Germany) SpRing-8 (Japan) Conclusion At present new NMR Magnetometer in VME standard is developing: -one-channel magnetometer will occupy one VME module -all digital signal processing will be performed by this module

The end

ЯМР магнитометры, разработанные в России и за рубежом Изготовитель Параметр Диап. полей, Т Миним. толщ. датчика мм 3 Темпе- ратура Ср.кв. погр. (σ F ) 1 /F Неста- бильн. част. за сутки Макс. град. G/B ИЯФ СО РАН мм401Комн., 4 К / С / см Metrolab Instrument, Швейцария ~3 мм310Комн Т / С / см Drusch, Германия ~5 мм520Комн Т10 -6 / год – Virginia Scientific Instrument, США ~2 мм Комн / год – ОИЯИ, Дубна – 2 – Комн. –– / см Лаборатория PTB, Германия ––– Комн ––– Университет Гейдельберга, Германия ~5 мм –– Комн / год10 -8 – Швейцарский Федеральный Институт Технологии ~1 мм101Комн –––

Максимальная погрешность относительных измерений поля (за сутки) для двух случаев где Т И – полное время одного измерения. С учетом нестабильности частоты Синтезатора: δ ОТН , (ΔТ О = 10°C). Без учета нестабильности частоты Синтезатора: δ ОТН ( 1 2) ) Измерения поля сверхпроводящего соленоида при температуре жидкого гелия датчиком с объемом рабочего вещества (порошка алюминия) около 2 мм 3. В диапазоне полей 5 13Т : δ ОТН ( 1 2) (ΔТ О = 10гр.). 1) Измерения поля калибровочного магнита стенда магнитных измерений ИЯФ нерезонансным датчиком с объемом рабочего вещества (воды) около 4 мм 3. В диапазоне полей Т : δШδШ δ СП δ CИНТ

Максимальная погрешность относительных измерений поля калибровочного магнита где Т И – полное время одного измерения, ΔТ О – изменение температуры окружающей среды. При использовании нерезонансного датчика с объемом рабочего вещества (воды) около 4 мм 3 в диапазоне полей Т величина δ ОТН (за сутки) : δШδШ δ СП δ CИНТ При времени одного измерения Т И = 2 сек: Без учета нестабильности частоты Синтезатора: δ ОТН (1÷2) С учетом нестабильности частоты Синтезатора: δ ОТН × ΔТ О – за сутки; δ ОТН × ΔТ О – за год

Результаты измерения естественного спада поля центрального полюса вигглера в режиме Persistent Current Mode. Всего выполнено около 1000 измерений за 12 часов Постоянная времени спада поля – около одного года Погрешность измерения поля,вызванная шумами, для каждого из выполненных измерений. σ F /F

Two methods of the Resolution value measurement A few thousands field measurements are performed. Resolution is calculated on base of differences of two Magnetometers results. 1) Indirect method – on base of measured values of the NMR signal and noise 2) Direct method – by using of two NMR Magnetometers Magnetometer 1 Magnetometer 2 Start S N Probe 1 Probe 2 Magnet