Class Note for CHEM 777 at UMass(3)
Class Note for CHEM 777 at UMass(3)
Popular in Course
Popular in Department
This 5 page Class Notes was uploaded by an elite notetaker on Friday February 6, 2015. The Class Notes belongs to a course at University of Massachusetts taught by a professor in Fall. Since its upload, it has received 17 views.
Reviews for Class Note for CHEM 777 at UMass(3)
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/06/15
Ch 777 Spring 03 page 28 14 N MR spectrometers Magnet high field electromagnets up to 24 T m 100 MHZ 1H Larmor frequency supercons for higher fields 7 500 600 MHZ common some 800 900 MHZ homogeneity r needed for narrow lines shim system stability 7 needed for signal averaging lock Superconducting magnet r Bovey p 8 Coil of superconducting wire Niobiumrtin r superconducting below TC 7 ie at 4K7 immerse in liquid He Energize with appropriate amount of current to get desired field remove current source and current continues indefinitely because no resistance to dissipate it Only need to maintain low T refill He inner dewar 5 months N2 outer dewar mweekly Avoid quotquenchingquot magnet 7 TT gt finite resistance gt TT gt He boil off gt TT etc accelerating process rapid heat production boils off all cryogens Shim system 7 series of coils of various shapes and orientations which generate small fields to correct for nonuniformities in large magnetic field X shim coil in X direction generate small Bx etc Two types 1 Superconducting shims cryoshimsr also made of superconducting wire charged up at installation and kept at 4K also 7 always same unless bring down magnet 2 Room temperature shims r nonsuperconducting shim coils current input to ech coil is adjusted by user to finertune field homogeneity Goal of shimming most uniform field over sample volume will give longest T2 decay of FlD narrowest line Can shim on lock FlD or signal level 7 higher for narrower line or on sample FlD r aim for longest FlD decay or spectrum 7 info from type of distortion Want narrowest lines to get optimal resolution and sensitivity Lock system 7To prevent field drift which would make signal raveraging impossible Probe has additional coil around sample tuned to 2H which is pulsed and detected continuously lnclude 2H in sample then feedback circuit looks at 2H signal and maintains it at same frequency by controlling current input to coil parallell to B0 gt field stays constant Block diagram 7 Bovey p 40 Probe 7 Holds sample in quotsweet spotquot 7 most homogeneous part of magnet Maintains sample conditions 7 ie spinning just for averaging xy field gradients not needed if wellrshimmed temperature etc Electronic circuit transmit B1 amp receive signal 7 same coil around sample perpendicular to B0 Circuit is tuned to frequency of interest 7 resonates at or absorbs that frequency Simple LC circuit resonates at n LC 12 Tune by varying capacitance input range of frequencies see dip at resonant frequency change tuning and matching capacitors to bring dip to right frequency and maximum depth Can also pulse at resonant frequency and minimize relected power Ch 777 Spring 03 page 29 Broadband 7 can tune over range of frequencies with variable capacitors Multiply tuned Can transmit and receive multiple frequencies either with multiple coils solution or with one coil in a circuit tuned to multiple frequencies solidrstate singlercoil doublertuned circuit Multiple coils r inner coil gets best sensitivity so use it for observe nucleus 7 ie if HN exp observing H use inner H coil outer N coil tunable Type of coil Helmholtz r B1 perpendicular to coil quotaxisquot 7 so sample can be put in from above so typical in solution probes Solenoid 7 solid state NMR Surface coil 7 imaging Block diagram Pulse programmerr sends out digital signals coding for pulse sequence to rf transmitter r oscillator tuned to vL pulses sent into amplifiers then to probe coil Pulse excitation with specific phase 90 phase shift by delaying 7M4 Pulse sequence diagram time frequency phase each pulse important Signal induced in coil weak goes to preamplifier then to receiver Receiver mix signal v Av with carrier v to get Av which is in the kHz instead of MHz range easier to handle phasersensitive detection quadraturerdetection 7 get two signals 90 out of phase with each other filter out noise audio bandpass filter set according to sweeprwidth FW gt SW so if SW r 25 kHz perhaps FW r 35 kHz analogrtordigital conversion ArtorD converter Computer 7 stores digital signals for processing sum FT etc Single channel observe shown Additional channels lock 7 xmit observe 2nd3rd for decouplemultinuclear exps r pprog xmit amplifiers 15 Signal acquisition amp processing parameters Signal acquisition parameters Start with standard sample and standard spectra to be sure performance is normal normal SN indicates shimming probe tuning etc are OK Digitization rate 1 Nyquist frequency the highest frequency you can detect for a given sampling rate must record two data points per period of oscillation or will be indistinguishable from lower frequency Bovey fig p 64 Sanders p 23 Suppose sampling rate is 2v 3 v is highest frequency you can detect So need fast sampling rate for large SW sampling rate lsampling time interval lDW sampling time interval dwell time time per point DW l2sampling rate highest detectable frequency SW lZDW example SW 2 kHz means must sample at 4 kHz rate every 250 ps Ch 777 Spring 03 page 30 One reason why signals are mixed down to kHz so don39t have to sample at MHZ rates every 250 ns So changing DW changes SW and vice versa according to SW 12DW Folding Aliasing occurs when set SW too narrow Signals outside ie with frequency gt v are quotfoldedquot back into spectrum Signal at v Av will be indistinguishable from v 7 Av So it will appear in the spectrum at the lower frequency So frequencies higher than Nyquist frequency are detected as lower frequencies 7 quotfoldedquot into the spectrum or quotaliasedquot SW set wide enough for entire spectrum so signals outside aren39t folded in at lower frequencies set no wider than needed so not collecting excess noise amp not cutting down on resolution Hzpt Filters FW are set according to SW gradual cutoff to reduce noise being folded back into spectrum Digital resolution DR 3 Number of points collected Slze 2N for FT algorithm 1024 2048 4096 8192 referred to as 1K2K4K8K16K32K Resolution DR Hzpt SWSl2 2SWSl collecting X amp y at same time 7gt only 12 of SI helps DR note factors of 2 vary with software convention does Sl include both channels or just 1 etc ie 2K Hz8K real points 8192 actually 16K Sl total 025 Hzpt Can39t measure to better than i 025 Hz lf lines are narrower than DR they will appear broadened Sanders p 27 Actually same data 7 transform 1st 512 through all 32K points Set acquisition time Sl X DW to acquire to acquire full FlD r for full resolution 7 avoid FlD truncation artifacts No point acquiring beyond full decay of signal just noise 7 points out relationship between extent of FlD T2 decay and linewidth Useful relations DW12SW Nyquist Sl x DW AQ Aquisition time DR 2SWSl1DW x1Sl1AQ Dynamic range 4 Ratio of strongest to weakest peak 7 ie H20 to dilute protein ratio must not be so large that computer can39t store both extremes even signal averaging won39t pull weak signals out of the noise if thy are beyond the dynamic range limit can39t just overload digitizer clip FlD or will introduce distortions each point contains info about all frequencies 7 SN 1 extra pks r distort39ns around signal Sanders p35 Digitizer resolution 7 no of bits ea 0 or 1 8 bit can store 00000000 0 through 11111111 255 28 r 1 smallest signal is 1255 04 of largest 12 bit 14096 02 16 bit 165536 002 Keep TMS low sample Ch 777 Spring 03 page 31 Receiver gain 7 RC 7 turn up enough to fill digitizer without clipping get smallest signals above detection threshold though in noise on single scan avoid FlD clipping artifacts Repetition delay want max signal OR want full relaxation 5 Tl39s for full relaxation 99 90 7 max signal but wait longer to recycle best if measuring intensities smaller tip angle 7 smaller signal but recycle faster Phase cycling 7 later 7 cancel out unwanted constant signals correct for pulse imperfections essential to some exps Signal averaging r SN NNn so have to go 4x longer for 2x SN Signal processing parameters 5 r Bovey p 68 Baseline correct 7 correct for DC offsets in each receiver channels otherwise see zero frequency glitch compute average voltage and subtract from all Zero fill 7 add zeros to end of FlD r ZF to 2S1 is good routine practice resulting FT is done on more points so you get interpolated points between data and can better measure linewidths Digital ltering apodization 7 improve SN or resolution at expense of other Weight different regions of FlD more or less ie weight early FlD more 7 has highest Sn bf S decays r enhance SN are diminishing signals at long t39s so decreasing resolution Exponential multiplication EM 7 multiply FlD by e LBt LB line broadening Sanders p 30 it 1 Natural decayS e T2 Av T LOTtDIItZIan 111195 7395 2 rt 19LB For optimum SN enhancement use LB m linewidth S e 71 doubles linewidth quotmatched filterquot but lose resolution Use LB DR on a truncated FlD to force it to decay to zero amp remove distortions Truncated FlD FlD x step FTstep sinxx sinc x Derome p 23 get wiggles around peaks LB gt 0 enhances SN at the expense of resolution LB lt 0 resolution enhancement 7 sharpen lines at expense of SN r pbs Better for resolution enhancement it to enhance middle part of FlD Sinerbell function max at center of AT distorts lineshapes Phasershifted sinerbell move max weighting earlier in FlD reduces SN amp distortion penalties Gaussian multiplication best for resolution enhancement wo SN loss LB controls exponential decay 72 to SDR GB controls position of maximum 7 where FlD reaching noise can convert lineshapes to Gaussian normally Lorentzian narrower bases rgt enhances resolution 5x narrower 1 height 7 Need good SN to apply resolution enhancement Caution these alter intensity ratios weighting broad or narrow lines more Ch 777 Spring 03 page 32 Phasing FT fourier transform f t and Fn are Fourier inverses ft 1 wFmei dw 2n Fw jwftequot39 quotdr 6 cosx isin x Fw Jfoftcosntdt ifoftsinmtdt real imaginary foo foo Signals additive FT ft gt Fn C03 lf FlD starts at max or min get real absorptive signal other channel with 90 offset FlD starting max pure absorption signal 7 even functions I2I from 0 to infinity FlD starting zero pure dispersion signal 7 odd function I0 more in Nutsamp Bolts p67 FlD won39t start at maximum in one channel 7 but can combine appropriately with other channel39s signal to reconstruct real signal picture in rotating frame 7 reference signal is not exactly on y axis so all signals are phasershifted by an angle with respect to reference lag reference by x Zeroeorder constant phase correction 7 applies same across spectrum 7correct above Firstrorder linear phase correction 7 apply linearly varying correction across spectrum 7 Why needed Small delay between B1 pulse and acquisition 107100 ps 7 pulse ringrdown want it gone before try to detect much tinier signals 7 during delay offr resonant components precess proportional to how far they are off resonance 7 so don39t capture max of FlD to varying extents proportional to AV 7 appl linear correction to compensate Higher v39s are more phasershifted Shifting time domain phase factor in frequency domain Phase shift due to delay before acquisition 180 point 180 DW Same phase corrections should apply to series of spectra under same conditions same day because same cable lengths pulse ringing delays same pulse sequence
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'