Verify Twice, Publish Once

The message behind an old carpenter’s adage ‘measure twice, cut once’ can be applied to the process of structure elucidation as ‘verify twice, publish once’. Committing the time to double check a candidate structure against the experimental data can save a lot of hassles and embarrassment from a mistake being published/presented/stored/reported. Without picking on a specific case, journals are littered with examples of incorrect structures.

As the holidays approach, Philosophy to Chemistry to Elucidation blogging will be taking a short break and will resume in the new year. In the meantime, the following link is an interesting read on the importance of shape within the field of Chemistry: Angew. Chem. Int. Ed. Engl. 30 (1991)1-16.

Verify Twice, Publish Once

The message behind an old carpenter’s adage ‘measure twice, cut once’ can be applied to the process of structure elucidation as ‘verify twice, publish once’. Committing the time to double check a candidate structure against the experimental data can save a lot of hassles and embarrassment from a mistake being published/presented/stored/reported. Without picking on a specific case, journals are littered with examples of incorrect structures.

As the holidays approach, Philosophy to Chemistry to Elucidation blogging will be taking a short break and will resume in the new year. In the meantime, the following link is an interesting read on the importance of shape within the field of Chemistry: Angew. Chem. Int. Ed. Engl. 30 (1991)1-16.

Missing the Big Picture?

When peak picking a 2D NMR experiment, past weblogs have advocated zooming in on correlations especially in cases dealing with ambiguity. Depending on the data collection parameters, a 1H-13C HMBC experiment can contain paired 1J coupling responses. Without careful scrutiny of the data, these extra responses can be misinterpreted as long-range correlations (2J or longer).

The 1H -13C HMBC spectrum below indicates two peaks picked, thus, correlating proton 1.24 ppm to carbon 20.47 ppm and proton 1.35 ppm to carbon 23.07 ppm. This is an easy interpretation that cannot possibly be wrong, or is it?

    

Before accepting the peak picking, it is best to take a step back—actually zoom out a little further to get the bigger picture. The correlations from the 1H -13C HMBC spectrum are 1J coupling responses and thus not long-range correlations. The blue lines indicate the paired 1J coupling responses. When assigning long-range correlations, 1J coupling responses are best not to be picked.

    

NOTE: Although it is possible for a long-range correlation to overlap with a 1J coupling response, it may be wise to examine the volumes of the responses for any significant differences.

Missing the Big Picture?

When peak picking a 2D NMR experiment, past weblogs have advocated zooming in on correlations especially in cases dealing with ambiguity. Depending on the data collection parameters, a 1H-13C HMBC experiment can contain paired 1J coupling responses. Without careful scrutiny of the data, these extra responses can be misinterpreted as long-range correlations (2J or longer).

The 1H -13C HMBC spectrum below indicates two peaks picked, thus, correlating proton 1.24 ppm to carbon 20.47 ppm and proton 1.35 ppm to carbon 23.07 ppm. This is an easy interpretation that cannot possibly be wrong, or is it?

    

Before accepting the peak picking, it is best to take a step back—actually zoom out a little further to get the bigger picture. The correlations from the 1H -13C HMBC spectrum are 1J coupling responses and thus not long-range correlations. The blue lines indicate the paired 1J coupling responses. When assigning long-range correlations, 1J coupling responses are best not to be picked.

    

NOTE: Although it is possible for a long-range correlation to overlap with a 1J coupling response, it may be wise to examine the volumes of the responses for any significant differences.

Confirmation of Synthesis: using MS to identify a protective group

Many organic chemists employ Mass spectrometry (MS) as a convenient verification tool for their product in a synthetic reaction. Derivatization such as adding a protective (or protecting) group can often be detected by MS.

The EI mass spectrum for tert-butyl 3-aminopiperidine-1-carboxylate is shown below. The 'terminal' atoms belonging to the protective group, tert-Butyloxycarbonyl (BOC or t-BOC), are coloured in blue. Although the parent ion is not clearly evident, the mass spectrum shows 3 ion clusters: m/z 57, 99 and 127 with the associated fragments listed below.

 

Confirmation of Synthesis: using MS to identify a protective group

Many organic chemists employ Mass spectrometry (MS) as a convenient verification tool for their product in a synthetic reaction. Derivatization such as adding a protective (or protecting) group can often be detected by MS.

The EI mass spectrum for tert-butyl 3-aminopiperidine-1-carboxylate is shown below. The 'terminal' atoms belonging to the protective group, tert-Butyloxycarbonyl (BOC or t-BOC), are coloured in blue. Although the parent ion is not clearly evident, the mass spectrum shows 3 ion clusters: m/z 57, 99 and 127 with the associated fragments listed below.

 

Examining the IR pattern for C=O vibrations

Infrared (IR) spectrometry is an excellent diagnostic tool for identifying, or lack of, a carbonyl functional group(s). Carbonyl stretching vibration absorbs between 1900-1600 cm-1—a region where few other functional groups absorb. In addition, the carbonyl vibration is typically intense and thus easy to spot.

The gas phase FT-IR spectra, shown below, represent five carbonyl bands differing in the neighbouring atoms: acid halide, carboxylic acid, ester, ketone and amide. The general pattern can be used to suggest a carbonyl group based on the wavenumber.

TIP: conjugated carbonyl groups and hydrogen bonding will typically shift the C=O band to a lower frequency. Ring strain and heteroatoms will shift the C=O band to a higher frequency.

Examining the IR pattern for C=O vibrations

Infrared (IR) spectrometry is an excellent diagnostic tool for identifying, or lack of, a carbonyl functional group(s). Carbonyl stretching vibration absorbs between 1900-1600 cm-1—a region where few other functional groups absorb. In addition, the carbonyl vibration is typically intense and thus easy to spot.

The gas phase FT-IR spectra, shown below, represent five carbonyl bands differing in the neighbouring atoms: acid halide, carboxylic acid, ester, ketone and amide. The general pattern can be used to suggest a carbonyl group based on the wavenumber.

TIP: conjugated carbonyl groups and hydrogen bonding will typically shift the C=O band to a lower frequency. Ring strain and heteroatoms will shift the C=O band to a higher frequency.

Examining IR spectra for the presence of OH

Infrared (IR) spectrometry can serve as a simple method to gather information on the presence of OH. Characteristic OH absorptions occur around the ranges of 3700-3200 cm-1 (OH stretching) and 1200-1000 cm-1 (OH bending). In the case of hydrogen-bonded OH, the band in the region 3700-3200 cm-1 generally appears broad and sometimes can go unnoticed.

Below is the IR spectrum for butanoic acid. The presence of the two bands at 3582 cm-1 and 1150 cm-1 suggest that an OH group is present.

 

 

TIP: An incorrect interpretation may occur for samples that contain impurities with OH groups, for example, water.

Examining IR spectra for the presence of OH

Infrared (IR) spectrometry can serve as a simple method to gather information on the presence of OH. Characteristic OH absorptions occur around the ranges of 3700-3200 cm-1 (OH stretching) and 1200-1000 cm-1 (OH bending). In the case of hydrogen-bonded OH, the band in the region 3700-3200 cm-1 generally appears broad and sometimes can go unnoticed.

Below is the IR spectrum for butanoic acid. The presence of the two bands at 3582 cm-1 and 1150 cm-1 suggest that an OH group is present.

 

 

TIP: An incorrect interpretation may occur for samples that contain impurities with OH groups, for example, water.

Does my Unknown Compound contain Chlorine?

Like bromine, compounds that contain chlorine atoms have a distinct ion pattern on a mass spectrum. The A+2 peak for a monochlorinated compound will be at almost one-third the intensity to the 35Cl peak due to the presence of 37Cl isotope. A compound with two chlorine atoms will show distinct A+2 and A+4 peaks with an approximate ratio of 10:6:1.

The following EI mass spectra are for 3-chloropropanenitrile and (1E)-1,2-dichlorobut-1-ene with nominal masses at 89 and 110 Da, respectively. The top MS, for 3-chloropropanenitrile, shows 2 identifiable ion clusters at m/z 49/51 and 89/91 that contain chlorine. The ion peak at m/z 54 does not contain chlorine.

 

The MS for (1E)-1,2-dichlorobut-1-ene shows 2 ion clusters at m/z 75/77 and 110/112/114. Only the ion cluster at m/z 110 is dichlorinated.

TIP: Check the fragment ion peaks too for the distinct pattern especially when the molecular ion peak is not visible.

Does my Unknown Compound contain Chlorine?

Like bromine, compounds that contain chlorine atoms have a distinct ion pattern on a mass spectrum. The A+2 peak for a monochlorinated compound will be at almost one-third the intensity to the 35Cl peak due to the presence of 37Cl isotope. A compound with two chlorine atoms will show distinct A+2 and A+4 peaks with an approximate ratio of 10:6:1.

The following EI mass spectra are for 3-chloropropanenitrile and (1E)-1,2-dichlorobut-1-ene with nominal masses at 89 and 110 Da, respectively. The top MS, for 3-chloropropanenitrile, shows 2 identifiable ion clusters at m/z 49/51 and 89/91 that contain chlorine. The ion peak at m/z 54 does not contain chlorine.

 

The MS for (1E)-1,2-dichlorobut-1-ene shows 2 ion clusters at m/z 75/77 and 110/112/114. Only the ion cluster at m/z 110 is dichlorinated.

TIP: Check the fragment ion peaks too for the distinct pattern especially when the molecular ion peak is not visible.

Comparing Spectra of the Starting Material to the Unknown Product

For synthetic reactions where rearrangement, derivatization, or cyclization has occurred, a common task is to compare the NMR spectra between the starting material and the product. The similar peaks indicate a structural region where change has not occurred whereas the unique peaks indicate a region where change has occurred. Arguably, this becomes a peak mapping task as opposed to structure elucidation which is geared towards the unique pieces.

A cyclization reaction for the starting material, N-(2,6-dimethylphenyl)-N2,N2-dimethylalaninamide is shown below. The dotted circle represents the unknown region of the product. The CH3 groups circled in red indicate the similar regions between the starting material and the product.

 

The unknown product, N,N-dimethyl-1-(8-methyl-1,4-dihydro-2H-3,1-benzoxazin-2-yl)ethanamine, is shown below. The blue circle indicates the point of the cyclization and thus the area where an elucidator must piece together the experimental data to complete the structure.

 

Comparing Spectra of the Starting Material to the Unknown Product

For synthetic reactions where rearrangement, derivatization, or cyclization has occurred, a common task is to compare the NMR spectra between the starting material and the product. The similar peaks indicate a structural region where change has not occurred whereas the unique peaks indicate a region where change has occurred. Arguably, this becomes a peak mapping task as opposed to structure elucidation which is geared towards the unique pieces.

A cyclization reaction for the starting material, N-(2,6-dimethylphenyl)-N2,N2-dimethylalaninamide is shown below. The dotted circle represents the unknown region of the product. The CH3 groups circled in red indicate the similar regions between the starting material and the product.

 

The unknown product, N,N-dimethyl-1-(8-methyl-1,4-dihydro-2H-3,1-benzoxazin-2-yl)ethanamine, is shown below. The blue circle indicates the point of the cyclization and thus the area where an elucidator must piece together the experimental data to complete the structure.

 

Disclaimer for P2C2E

The following weblog represents my experiences in and out of the lab. As such, the presented material serves as a personal resource and should in no way be considered as a comprehensive approach to lab work or otherwise. I do not warrant or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed.

 

Reader discretion is advised.

 

Readers cannot assume that the external sites will abide by the same Policy to which P2C2E adheres. It is the responsibility of the reader to examine the copyright and licensing restrictions of the linked pages and to secure all necessary permissions.

 

The material listed from this weblog site may not be freely distributed and copied without the written permission from the author. For contact details, please visit the About Me section. 

 

To cite the weblog P2C2E, please follow the sample guideline below:

 

Moser, A. "How to reference 1D and 2D NMR data? … Part 1" Philosophy to Chemistry to Elucidation. 2008. http://acdlabs.typepad.com/elucidation/2008/08/how-to-referenc.html.

 

Revised: Oct 30, 2008

Disclaimer for P2C2E

The following weblog represents my experiences in and out of the lab. As such, the presented material serves as a personal resource and should in no way be considered as a comprehensive approach to lab work or otherwise. I do not warrant or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed.

 

Reader discretion is advised.

 

Readers cannot assume that the external sites will abide by the same Policy to which P2C2E adheres. It is the responsibility of the reader to examine the copyright and licensing restrictions of the linked pages and to secure all necessary permissions.

 

The material listed from this weblog site may not be freely distributed and copied without the written permission from the author. For contact details, please visit the About Me section. 

 

To cite the weblog P2C2E, please follow the sample guideline below:

 

Moser, A. "How to reference 1D and 2D NMR data? … Part 1" Philosophy to Chemistry to Elucidation. 2008. http://acdlabs.typepad.com/elucidation/2008/08/how-to-referenc.html.

 

Revised: Oct 30, 2008

Identifying Peak Overlap on an HMBC Spectrum

Carbon peaks that overlap on an 1H -13C HMBC experiment can be tricky to deal with especially when additional experiments do not help to clarify the situation. A good approach is to keep note of any high correlation counts for a carbon resonance, and subsequently, treat the carbon resonance as possibly multiple carbons with coincidental chemical shifts.

The 1H -13C HMBC spectrum and attached 1D 1H NMR spectrum below illustrate a carbon resonance at 76 ppm with 7 proton correlations. Treating this as a single carbon will result in a structural dead end to the elucidation process. The next step is to treat the carbon as 2 carbons resonances.

    

The corresponding structure is shown below to illustrate the point.

Identifying Peak Overlap on an HMBC Spectrum

Carbon peaks that overlap on an 1H -13C HMBC experiment can be tricky to deal with especially when additional experiments do not help to clarify the situation. A good approach is to keep note of any high correlation counts for a carbon resonance, and subsequently, treat the carbon resonance as possibly multiple carbons with coincidental chemical shifts.

The 1H -13C HMBC spectrum and attached 1D 1H NMR spectrum below illustrate a carbon resonance at 76 ppm with 7 proton correlations. Treating this as a single carbon will result in a structural dead end to the elucidation process. The next step is to treat the carbon as 2 carbons resonances.

    

The corresponding structure is shown below to illustrate the point.

Unassigned Protons can serve as Warning Flag

When the incorrect number of directly-bonded protons are assigned to carbons, the elucidator is left with extra protons. (This can happen in situations with a highly-crowded region on a 1H NMR spectrum.) Where possible, tallying the expected number of exchangeable protons can serve as a warning flag that something is amiss.

The following carbons, shown below, were assigned as C, CH, CH2 or CH3. As such, three protons remain unassigned. Assuming the compound contains no ammonia, the carbon assignments were incorrectly done and thus should be reinvestigated.

 

The structure methamphetamine, shown below, illustrates that a methyl was incorrectly set to a methine.

Unassigned Protons can serve as Warning Flag

When the incorrect number of directly-bonded protons are assigned to carbons, the elucidator is left with extra protons. (This can happen in situations with a highly-crowded region on a 1H NMR spectrum.) Where possible, tallying the expected number of exchangeable protons can serve as a warning flag that something is amiss.

The following carbons, shown below, were assigned as C, CH, CH2 or CH3. As such, three protons remain unassigned. Assuming the compound contains no ammonia, the carbon assignments were incorrectly done and thus should be reinvestigated.

 

The structure methamphetamine, shown below, illustrates that a methyl was incorrectly set to a methine.

Interpreting an HSQC or HMQC or HETCOR experiment

The 1H-13C HMQC, HSQC, DEPT-HSQC, HSQC-TOCSY and HETCOR experiments offer the elucidator information on the proton-carbon connectivity. The interpretation process comes down to 3 basic assignments: the correlation belongs to a methyl, methylene or methine carbon. A methyl or methine carbon exhibits at most a single correlation between the 1H and 13C axes. A methylene group can exhibit 1 or 2 correlations depending on whether the protons are anisochronous.

Three regions of a 1H -13C HSQC spectrum for quinine are shown below.

A single 1H-13C correlation is observed for a CH and CH3.

For anisochronous methylene protons, 2 correlations are observed.

Interpreting an HSQC or HMQC or HETCOR experiment

The 1H-13C HMQC, HSQC, DEPT-HSQC, HSQC-TOCSY and HETCOR experiments offer the elucidator information on the proton-carbon connectivity. The interpretation process comes down to 3 basic assignments: the correlation belongs to a methyl, methylene or methine carbon. A methyl or methine carbon exhibits at most a single correlation between the 1H and 13C axes. A methylene group can exhibit 1 or 2 correlations depending on whether the protons are anisochronous.

Three regions of a 1H -13C HSQC spectrum for quinine are shown below.

A single 1H-13C correlation is observed for a CH and CH3.

For anisochronous methylene protons, 2 correlations are observed.

Anisochronous Protons on a 1H NMR Spectrum

In NMR, nuclei can be classified as isochronous or anisochronous. “Where diastereotopic protons show the same chemical shift, they are said to be accidentally equivalent or isochronous, and where they have different chemical shifts the protons are described as anisochronous.” Stereochemistry by David G. Morris, Royal Society of Chemistry (Great Britain) Published by Royal Society of Chemistry, 2001.

The 1H NMR spectrum is shown for geminal protons, coloured in red, within a ring system. The spectrum illustrates protons from a methylene group (CH2) with different chemical shifts that are coupled to each other. The protons are termed as diastereotopic and anisochronous.

Other examples of this property are seen at the following links: DEPT-HSQC, HSQC/HMBC and 1H/DEPT-HSQC.

Anisochronous Protons on a 1H NMR Spectrum

In NMR, nuclei can be classified as isochronous or anisochronous. “Where diastereotopic protons show the same chemical shift, they are said to be accidentally equivalent or isochronous, and where they have different chemical shifts the protons are described as anisochronous.” Stereochemistry by David G. Morris, Royal Society of Chemistry (Great Britain) Published by Royal Society of Chemistry, 2001.

The 1H NMR spectrum is shown for geminal protons, coloured in red, within a ring system. The spectrum illustrates protons from a methylene group (CH2) with different chemical shifts that are coupled to each other. The protons are termed as diastereotopic and anisochronous.

Other examples of this property are seen at the following links: DEPT-HSQC, HSQC/HMBC and 1H/DEPT-HSQC.

Assembling a set of Fragments to Complete a Candidate Structure

The progression of a structure elucidation process is to examine the experimental data, compare the results to literature if possible, build a set of fragments based on the available data and finally assemble the fragments until a candidate structure(s) is reached. The assembly part is very much like working on a jigsaw puzzle. When all the pieces are present, it is as easy as matching up the right pieces by examining the overlap and/or trying out different combinations.

The following fragments #1-5 represent all the available information extracted from a set of experimental data. The letter A represents an open point of attachment.

The goal of the elucidator is to sort out what pieces belong together and which do not. Although more than one answer is possible, a candidate structure, 6-{[5-(trifluoromethyl)cyclopenta-1,4-dien-1-yl]methyl}-2,3-dihydropyridin-4(1H)-one, is colour-coded to illustrate how the fragments fit together to complete the structure.

Assembling a set of Fragments to Complete a Candidate Structure

The progression of a structure elucidation process is to examine the experimental data, compare the results to literature if possible, build a set of fragments based on the available data and finally assemble the fragments until a candidate structure(s) is reached. The assembly part is very much like working on a jigsaw puzzle. When all the pieces are present, it is as easy as matching up the right pieces by examining the overlap and/or trying out different combinations.

The following fragments #1-5 represent all the available information extracted from a set of experimental data. The letter A represents an open point of attachment.

The goal of the elucidator is to sort out what pieces belong together and which do not. Although more than one answer is possible, a candidate structure, 6-{[5-(trifluoromethyl)cyclopenta-1,4-dien-1-yl]methyl}-2,3-dihydropyridin-4(1H)-one, is colour-coded to illustrate how the fragments fit together to complete the structure.

Differentiating M+ and M+H+ ions in an ESI+ MS experiment

In electrospray ionization MS (ESI-MS), ions are produced by the addition of a proton ([M+H]+). However, in cases where the analyte molecule is already charged, e.g. quaternary amine salts, the resulting ion may be an M+ ion.

Two ESI+ mass spectra are shown below. To test whether the molecular ion is M+ or M+H+, deuterium is added to the sample and the data is recollected. If the ion peak does not change in position relative to the original MS, the ion peak is considered an M+ and not M+H+.

Note: the deuterium exchange test may also occur with OH/NH/SH groups too. It is best to have a fair idea of the structure prior to performing this test.

Differentiating M+ and M+H+ ions in an ESI+ MS experiment

In electrospray ionization MS (ESI-MS), ions are produced by the addition of a proton ([M+H]+). However, in cases where the analyte molecule is already charged, e.g. quaternary amine salts, the resulting ion may be an M+ ion.

Two ESI+ mass spectra are shown below. To test whether the molecular ion is M+ or M+H+, deuterium is added to the sample and the data is recollected. If the ion peak does not change in position relative to the original MS, the ion peak is considered an M+ and not M+H+.

Note: the deuterium exchange test may also occur with OH/NH/SH groups too. It is best to have a fair idea of the structure prior to performing this test.

Elucidating for Isomers

The underlying essence of a structure elucidation process is to structurally distinguish an unknown from a set of possible isomers. This is evident by the number of possible isomers for a given molecular formula.

The chart below divides isomers into two groups: Structural/Constitutional/Regio and Stereo/Spatial isomers. Wikipedia links are included for further reading into the different isomer classifications.

 
  
Structural/Constitutional/Regio

Skeletal/Chain

Position

Tautomer

Isotopomers

Functional Groups

Stereo/Spatial

Diastereomers

Enantiomers

Cis/Trans

E/Z

Confomers

Rotamers

D/L

(+)/(-)

R/S

Epimers

Anomers

α/β

Elucidating for Isomers

The underlying essence of a structure elucidation process is to structurally distinguish an unknown from a set of possible isomers. This is evident by the number of possible isomers for a given molecular formula.

The chart below divides isomers into two groups: Structural/Constitutional/Regio and Stereo/Spatial isomers. Wikipedia links are included for further reading into the different isomer classifications.

 
  
Structural/Constitutional/Regio

Skeletal/Chain

Position

Tautomer

Isotopomers

Functional Groups

Stereo/Spatial

Diastereomers

Enantiomers

Cis/Trans

E/Z

Confomers

Rotamers

D/L

(+)/(-)

R/S

Epimers

Anomers

α/β