Mr. Hossain,

The database goes with teh instrumental setup software. Ask the person, which has conducted for you the NMR measurements.

The following comment cannot be a guideline for structure elucidation. This you may get from a University course or some good books. Learning the art of  structure elucidation requires experience and typically takes some time. As I understand, you are interested in quick results, without (much) work and using some miracle software or database. Unfortunately such setup do not exist.

first some semantics. Structure

there is the molecular formula as most basic structural information.

You may determine the molecular formula given a mass spectrum and some restrictions from the NMR-spectra (mainly 13C). The molecular formula is the lowest level of structure information which can be searched in databases of chemical compounds like SciFinder or more specialized databases on natural products. (typically those databases and their use are commerzialized)

For the next  step it would be good to determine few structural features from NMR, like the number of CH, CH2 and CH3 groups (DEPT e.g.) and from the corresponding 1H spectra the nearest neighbour of methyl-groups (CH gives a doublett splitting in 1H etc)

With this additional Information and the answer set by the search of the molecular formula you may decide, whether the natural product might be already known and described in literature, then look up the original paper and compare the NMR given there for identification.

Identification may also rely on the 13C fingerprint (the chemical shifts and multiplicities = C, CH, CH2, CH3 alone) and an appropriate 13C NMR database like ACD,  CSEARCH, nmpredict or Bruker (again commercial)

The non-commercial databases like NIST, SDBS, NMRShiftdb work the other way around. They predict the 13C NMR spectrum, provided you have the chemical structure. (which you do not have as a start...) but these predictions are helpful and highly recommended to test your structure hypothesis (from mass spec and NMR correlation data). For common compounds you find the NMR data, e.g. NIST, or databases you find on the web, used for "metabonomics", "protein-nmr" etc.

A guided tour through the process of structure elucidation (if the structure cannot be easily identified) with the help of software tools comes with the NMR-spectrometer (Bruker  commercial software package) or using the elucidation tools from Mestrec or ACD (also commercial) and typically requires (despite some claims otherwise) some own knowledge and experience.

regards 

Herbert

Dear Prof.

Thanks for your brief discussion. I am not so experienced but I have some idea. I can predict the structure. However,  calculate the coupling constant is very difficult. Because if I use 900MHz machine is differed from 600MHz machine. I do not know why it is happening. 

Sincerely yours

Mohammed Amzad Hossain

Dear Mohammed,

typically the minimum set of spectra for a structure elucidations is 1H, HSQC and HMBC. Additionally the 1D 13C and in case of compounds with nitrogen a 15N HSQC and HMBC. Having the molcecular formula will drastically reduce the number of possibilities. Knowing hte coupling constants is not always necessary. With a COSY you can get correlation without having to match up coupling constants. The coupling constants and a NOESY or ROESY will come in handy when you try to determine the 3 dimensional structure.

Clemens

Don't know about natural products, but Bruker has a database of biofluid reference compounds often used for metabolomics, which can be used with their AMIX software. However, you need a license for that, so ask your NMR facility people. There's also the Human Metabolome database, but that will involve individually looking up compounds.

Also, you J-coupling should not be different at different fields. That seems odd. Maybe you're doing something wrong.

There are software's for Structure generation using 1D,2D and High resollution mass(or molecular formula). Commercial one's available are ACD Structure elucidator and CMC-se. Also there are open source programes like SENECA and Pylsd

Thanks for all for participation. I do not understand yet how I will elucidate the structure.

I can count the total carbon and how may CH, CH2 and CH3. Then how I will proceed.

If you do not have a high resolution MS then you can start by counting all carbons. Then integrate the proton spectrum to give you an approximate count of the number of hydrogen atoms. The other atoms can only be guessed or deduced from the chemicla shit of protons and carbons. What is the total number of carbons. How many carbons with a chemical shift of 165 ppm or more? How many CH3 groups at 30 - 40 ppm. How many at 25 - 30 ppm. Do you have methyl groups that are doublets or triplets?

This allows you to assemble some basic info about the molecule. Carbonyl carbons, O-Methyls and N-Methyls and C-Methyls that are attached to C, CH or CH2.

Next are the interpretation of HSQC and HMBC data.

But that is how far one ca get in a RG post. Now it is time to get a book to learn about structure elucidation.