Diffusion Ordered Spectroscopy (DOSY):

USE THE 2D VERSION with Automation Now at UIC

Data Acquisition

• rpar h1.bbo all  (read in 1d proton)
• lockdisp (open lock display window)
• rsh shims.bbo (read starting shims and adjust manually too)
• lock (lock on solvent).
• wobb (tune the probe, Very Important now)
• re dosy_400 1 (reads in 1st expt for dosy series. please contact me if it doesn't exist)
• rundosy (a 'macro' which will run all the dosy expts)

Here is a little script to generate the dosy setup macros for you NMR managers. It is run within the /mac directory of xwinnmr or topspin and generates 2 files
'rundosy' and 'setdosy' which you run from the command line. mkdosy.pl

DOSY Processing

The original 'glove' fitting software was developed by Jonathon C. Lansing in the Wright/Dyson group at the Scripps Research Institute. The Code was never published however.

This fitting routine can actually be used to for any series of 1D bruker or varian formats. For example, T1/T2, kinetics, or any exponentially decaying model. Other functions can easily be incorporated with some modification of the source code.

Some comments regarding the processing routine. Bruker usually has a processing/analysis routine however I like the control of my own scripts. All the processing and analysis is done in the linux environment (Fedora 14) on a dual boot machine. For UIC students I would recommend trying to learn and install Linux as there is a lot of freeware for data processing. A list of needed programs would include:

• nmrDraw/nmrPipe
• XMGR or XMGRACE
• PERL
• GLOVE (will be provided)

Here is a gzip'd and tar ball of an entire directory with all scripts and data. Use gunzip and tar xvf to extract.. DOSY.tar.gz
In general I'd run a series of processing steps using the scripts contained in above:

• ./procdosy 1 16 (processes all the 1D's from expt 1 to 16). Note it contains the header for converting bruker to nmrPipe format so edit as needed.
• ./foreach1d this is a shell script which will convert all nmrDraw spectra to text format. This is very useful for using own graphing programs
• ./mkInt.pl test.tab > Int (this is a conversion script to prepare for the data fitting routine)
• note that test.tab is a nmrDraw peaklist. also the mkInt.pl script has a header line which contains all the X-input information. In this case it is the DOSY gradient strengths, Delays and pulse widths (please see within).
• cd into the 'glove_pk' directory and run the go script. look inside go to see what is happening.
• in short, the go script removes existing xmgr files, creates 'input.dat' file for data fitting, fits the data and generates xmgr files according to test.tab peak list
• ./getdosy converts the glove results into a table 'results.txt'

Viewing the Results

All the results will be viewed in XMGR. I've have a file 'xmgr.results' which serves as a template to give:

In order to read in existing data use 'soft kill' function in xmgr for both graphs and read in '1.dat' file for top and 'results.txt' for bottom. Note the top graph is in XY
format and the bottom is XYDY.

Run the script ./makexmgr.pl > stack.dat in order generate a series of 1D stack plots like below. You can edit the 'mkxmgr.pl' to change the X and Y offsetting as you like..

Working with Integrations

./mkIntegral man.xpk > Int will generate a file using an integration range instead of peaks. The integral range is defined in the file 'man.xpk' (open to see simple format)
I usually just manually edit this file while looking at and xmgr plot of say 1.dat.

cd into glove_integral and run the go script. This will fit the Int file data as before. getdosy can then be used to extract a results.txt file to read into xmgr.
xmgr.results here is a template to read in data as before. Note top graph is XY format and bottom is now XYDXDY.

The results are now displayed as integral ranges represented with error bars in XDX and errors in YDY.