Brian C. Smith

Our final look at the infrared spectra of organic nitrogen functional groups will be of the nitro, or NO

, group. This group has two strong and characteristic infrared features, making it easy to identify. However, attaching a NO

 group to a molecule scrambles its electronic structure, which is why these compounds are explosive, and why nitro groups make some of the infrared interpretation rules we learned irrelevant.

Over the several years that I have been writing these articles, there have been large sections of material devoted to the spectroscopy of functional groups that contain specific chemical bonds. When I began discussing organics containing nitrogen, I stated that the stretching and bending of N-H bonds would be important in determining whether or not there was nitrogen present in a molecule (1). We subsequently used N-H stretching and bending peaks to analyze the spectra of amines (2,3), amides (4,5), imides (6), and urethanes (7).

The nitro group is going to be the odd man out in this discussion. This functional group consists of a nitrogen atom with no hydrogens, but with two oxygens and a carbon attached, as seen in Figure 1.

, and that the carbon atom singly bonded to the nitro nitrogen is called the 

. Depending on whether the alpha carbon is saturated, or part of an aromatic ring, nitro molecules can be divided into saturated and aromatic nitro compounds.

 group is unusual. Normally, oxygen atoms form two chemical bonds (8). However, there is also a C-N bond in the nitro group, as seen in Figure 1. Given that nitrogen normally forms three bonds (8), how do we apportion the electrons in a nitro group to keep these compounds from falling apart?

There are three bonding electrons shared between the two oxygens in the nitro group, giving what are essentially two “bonds and a half,” as seen in Figure 1. The dashed lines in Figure 1 represent the half bonds. These NO bonds are similar to the carboxylate C-O bonds and a half discussed in a previous column (9). Neither oxygen atom has its full complement of two full chemical bonds, making the nitro groups unstable. Saturated nitro compounds such as nitroalkanes, otherwise known as rocket fuel (8), are rarely analyzed by infrared spectroscopy because they are liable to detonate during analysis. We will not study them further.

However, nitro groups attached to benzene rings can be relatively stable, assuming not too many nitro groups are attached. The chemical structure of trinitrotoluene, a commonly used explosive known as 

The presence of the three nitro groups destabilizes the benzene ring, resulting in TNT’s explosive properties. Di-nitrotoluene and mono-nitrotoluene are stable. We will restrict our discussion to non-explosive aromatic nitro compounds.

The Infrared Spectroscopy of the Nitro Functional Group

The infrared spectrum of an aromatic nitro compound, meta-nitrotoluene, is seen in Figure 3. Recall that highly polar bonds have intense infrared features due to the large change in dipole moment with respect to bond length, 

, during a vibration (10). Oxygen is more electronegative than nitrogen; therefore, the N-O bonds in the nitro group are relatively polar, and as a result, their asymmetric and symmetric stretching peaks are unusually large. The details of these vibrations are shown in Figure 4.

 stretch typically falls from 1550 to 1500 cm

, and is seen in Figure 3 labeled A at 1527 cm

 units, even if not labeled as such). The symmetric stretch is seen in Figure 3 labeled B at 1350 cm

, and, in general, this peak appears from 1390 cm

. Note how peaks A and B in Figure 3 are the two most intense peaks in the spectrum, and they stick down like eye teeth in the middle of the spectrum. The combination of a pair of intense peaks in these wavenumber ranges is unique, making the presence of a nitro group in a sample easy to spot.

The nitro group also exhibits a scissors bending vibration, similar to that of the methylene group (11). This gives rise to a medium intensity peak from 890 cm

. It can be seen in Figure 3 labeled C at 881 cm

The good news about the nitro group is that it has two strong infrared bands that are easy to spot. The bad news is that whenever the nitro group is attached to a benzene ring, it makes it difficult to determine the substitution pattern on the benzene ring. Recall that the benzene ring out-of-plane C-H bend band, in combination with the presence or absence of the aromatic ring-bending band at 690 cm

, can be used to determine the substitution pattern on a benzene ring (12). The presence of a nitro group makes it difficult to apply these rules. This is caused by the unique electronic structure of the nitro group, and how it interacts electronically with the benzene ring. Suffice it to say, one may need to use an analytical technique other than infrared spectroscopy to determine the substitution pattern on nitro-substituted aromatic rings.

Note that the structure of meta-nitrotoluene does contain a methyl group. We have learned that the diagnostic pattern for a methyl group includes saturated carbon asymmetric and symmetric stretches near 2962 cm

 (13,14). Note in Figure 3 that the saturated C-H stretches fall at 2926 cm

. Under normal circumstances, we would interpret these two peaks as the asymmetric and symmetric stretches of methylene groups, because they have peaks at 2926 cm

 (14). In this case, that interpretation is wrong, because the nitro groups scramble the electronic structure of the molecule, throwing off these peak positions. If you notice the nitro peaks first, it will give you a heads up that the saturated C-H stretches may be troublesome. Normally we can depend on the methyl group umbrella mode to indicate that the saturated C-H stretches could be an issue. However, in this case, the intense nitro symmetric stretching peak at 1350 cm

 is sitting on top of it. Unfortunately, then there is nothing in this spectrum clearly showing the presence of the methyl group. As stated above, the nitro group throws off some of the interpretation rules we have learned.

The diagnostic pattern for the presence of a nitro group in a sample then is a pair of intense peaks at about 1550 cm

, along with the scissor peak around 850 cm

. Table I summarizes the group wavenumbers for the nitro group.

The nitro group consists of a nitrogen atom with two oxygens and one carbon attached. The two nitrogen-oxygen bonds are “bonds and half” that destabilize nitro compounds, making their analysis an explosive proposition. The large dipole moments for the NO bonds give two strong peaks around 1550 cm

 as a result of the asymmetric and symmetric stretching of the NO

functional group. This is an unusual pattern, and is easy to spot. There is also a scissoring peak around 850 cm

. Nitro groups tend to scramble the electronic structure of molecules, making the interpretation of their spectra problematic.

Infrared Spectral Interpretation: A Systematic Approach

Your Next Infrared Spectral Interpretation Challenge

Using everything you have learned in this and previous columns, determine the functional groups present in the Figure i spectrum and try to determine the chemical structure of this compound. Remember that inclusion of a peak position in the table does not necessarily mean that it will be useful in the structure determination.

Because this is a particularly tricky problem, answer these questions about the spectrum in this order to guide you. In all cases, be sure to justify your answer.

2. If so, is there a benzene ring present?

3. Are there any other functional groups present?

is the founder and CEO of Big Sur Scientific, a maker of portable mid-infrared cannabis analyzers. He has over 30 years experience as an industrial infrared spectroscopist, has published numerous peer-reviewed papers, and has written three books on spectroscopy. As a trainer, he has helped thousands of people around the world improve their infrared analyses. In addition to writing for 

 Dr. Smith writes a regular column for its sister publication

and sits on its editorial board. He earned his PhD in physical chemistry from Dartmouth College. He can be reached at: 

Articles

We review one final organic nitrogen functional group, representing explosive compounds. This is the nitro group, which requires a different set of interpretation rules.

Organic Nitrogen Compounds X: Nitro Groups, An Explosive Proposition

The urethane functional group is found almost exclusively in polymer form, known as 

. These ubiquitous polymers make urethanes an economically important functional group. Urethanes are made using diisocyanates. How to identify diisocyanates, and distinguishing among different urethane types using infrared spectroscopy, are explained in this installment of the column. 

		
		The urethane functional group is found almost exclusively in polymer form, known as 

. These ubiquitous polymers are used as wood finishes and make up polyurethane foam, also known as foam rubber. The structural framework of a urethane is shown in Figure 1.

	Note that this functional group contains a carbonyl bond with an oxygen atom and nitrogen atom attached to it. The latter is why urethanes are included in this series on organic nitrogen compounds. On the left-hand side of Figure 1, the urethane looks like an ester because of the O-C=O linkage (1), but on the right-hand side of Figure 1, the N-C=O linkage looks like an amide (2,3). As we will see, the infrared spectra of urethanes combine features of both esters and amides.

 if one carbon and two hydrogens are attached to the nitrogen atom, as 

 if two carbon atoms and one hydrogen atom are attached to the nitrogen, and as 

 if three carbon atoms are attached to the nitrogen. Figure 1 shows the structure of a secondary urethane. Note that it contains two C-N bonds and one N-H bond.

	Polyurethanes are made by reacting alcohols with a functional group called a 

. The structural framework of a diisocyanate is seen in Figure 2.

	Note that a diisocyanate contains two isocyanate -N=C=O linkages, hence the “di” in the name. A molecule with one -N=C=O unit would simply be called an 

. When urethanes and alcohols are polymerized to make polyurethanes, there can be isocyanate end groups that occasionally appear in the spectra
	of polyurethanes.

	The infrared spectrum of a diisocyanate is shown in Figure 3.

	Note that the spectrum is dominated by the peak labeled A at 2277 cm

 (assume all peak positions are in wavenumbers going forward, even if not labeled as such). This peak is due to the asymmetric stretch of the -N=C=O group, which normally falls from 2280 to 2240. This peak is the perfect example of a group wavenumber, because it is intense and shows up in a part of the spectrum where very few other functional groups absorb (5). Throughout the over 30 installments of this column, you may have noticed that most of the time we focus our attention from 3500 to 2800, and from 2000 to 400. That is because very few functional groups absorb between 2800 and 2000. The only other infrared peak we have seen that absorbs near diisocyanates is the C≡C stretch of disubstituted alkynes, which fall from 2260 to 2190 (6). These two functional groups are easily distinguished, because the diisocyanate -N=C=O asymmetric stretching peak is broader and stronger than a typical C≡C stretch. This is because of the high polarity of the -N=C=O group compared to that of an alkyne triple bond (7). The strength and unusual position of the -N=C=O stretching peak are why it sometimes appears in polyurethane spectra, even though it is in low concentration.

	Based on the structural framework of urethanes shown in Figure 1, we can predict to a point what urethane spectra will look like. Primary and secondary urethanes contain N-H bonds, and so have N-H stretches that are medium in width and intensity, and fall in the same wavenumber range as the N-H stretching peaks of amides and amines (2–4). The C=O bond in urethanes means there should be a strong C=O stretching peak (8) from around 1700, and the C-O bond means there should be a strong peak between 1300 and 1000 (9).

	The infrared spectra of primary urethanes behave as expected. Their asymmetric and symmetric NH

 stretches are found from 3450 to 3400 and 3240 to 3200, respectively. Primary urethanes have two NH stretching peaks, as do primary amines and amides (2–4), because, in all these cases, the primary nitrogen contains two N-H bonds, hence two N-H stretching peaks (3). The carbonyl stretch of primary urethanes falls at 1740 to 1730, and there is an NH

 scissors (in-plane bending) peak from 1630 to 1610.

	Unfortunately, the C=O stretching peak of primary urethanes falls in the same region as other carbonyl-containing functional groups, such as ketones, aldehydes, esters, and carboxylic acids (8,9). However, these functional groups do not contain nitrogen, hence there are no N-H stretching peaks, making distinguishing primary urethanes from these compounds easy. But primary amides have two NH

stretches, and a C=O stretch (3), as do primary urethanes. Fortunately, the primary urethane C=O stretch is at 1740 to 1730, whereas, for primary amides, it falls from 1680 to 1630 (10). This difference in C=O stretching peak position means primary amides and urethanes can be distinguished from each other. The diagnostic group wavenumber pattern, then, for primary urethanes is two NH stretching peaks, a C=O peak around 1730, and an NH

	The infrared spectrum of a secondary urethane, a “polyurethane/polyether foam,” better known as foam rubber, is seen in Figure 4.

	Because this polymer is a secondary urethane, as seen in the figure, it has a single N-H bond, and, hence, a secondary nitrogen. The spectral signature of a secondary nitrogen, as we saw for secondary amides and amines (2–4), is a single N-H stretching peak. In Figure 4, this peak is labeled A at 3326. In general, for secondary urethanes, this peak falls from 3340 to 3250. Note that the NH stretching peak in Figure 4 is of medium intensity and height, consistent with many of the other NH stretches we have seen (11).

	The carbonyl stretch in Figure 4 labeled B is split, with peaks at 1720 and 1705. In the past, a double carbonyl stretch like this indicated the presence of two separate carbonyl bonds in a functional group, such as for acid anhydrides (12) and imides (13). However, this is clearly not the case here; urethanes do not contain two carbonyl bonds. The cause of this split carbonyl peak is that some urethane C=O groups engage in hydrogen bonding, whereas others don’t. The 1720 peak is from non-hydrogen bonded, or “unassociated,” carbonyls, whereas the 1705 peak is from the hydrogen bonded, or “associated,” carbonyl groups (10). We have already studied the effects of hydrogen bonding on peaks, such as for alcohols (9). The tendency for peaks upon hydrogen bonding is to broaden and move to lower wavenumbers, as we have seen here.

	The peak labeled C in Figure 4 at 1531 is the in-plane N-H bend of a secondary urethane. In general, this peak falls from 1540 to 1520. Secondary amides also have a peak from the same vibration in the same place (3). However, secondary amide C=O stretches fall from 1680 to 1630, whereas, for secondary urethanes, this peak falls from 1740 to 1705, higher than in secondary amides, allowing us to distinguish between secondary urethanes and amides.

	The diagnostic group wavenumber pattern then for secondary urethanes is a single N-H stretch, a carbonyl stretch from 1740 to 1705, and an N-H in-plane bend from 1540 to 1520. Primary and secondary urethanes can be distinguished, because the former has two NH stretching peaks, whereas the latter has one.

	A further examination of Figure 1 shows that urethanes have a C-O bond, and we have learned in general that C-O stretching peaks are intense and fall from 1300 to 1000 (9). The peak labeled D at 1222 in Figure 4 is a urethane C-O stretch.  The C-O stretch for all urethanes falls from 1250 to 1200 (10). There is a large peak at ~1050 in Figure 4 that is not labeled. It is from the C-O stretch of the ether groups (14)in this particular polyurethane/polyether foam.

	Tertiary urethanes contain no NH bonds, hence they have no NH stretching or bending peaks. Their only useful group wavenumber is their C=O stretch from 1690 to 1680 (10), which also overlaps with amides. Thus, tertiary urethanes, like tertiary amines and amides (2–4) are a functional group that is hard to distinguish by infrared
	spectroscopy.

	A summary of the group wavenumbers discussed in this column is seen in Table I.

	Urethanes are often made into polyurethanes by reaction with diisocyanates. Diisocyanates have a strong and uniquely positioned -N=C=O asymmetric stretching peak from 2280 to 2240. Like amides and amines, urethanes can be classified as primary, secondary, and tertiary, based on the number of C-N and N-H bonds. Primary urethanes exhibit two NH stretches, a C=O stretch, and an NH

 scissoring vibration. Secondary urethanes have one NH stretch, a C=O stretch, and an NH in plane bend. Tertiary urethanes are difficult to discern by infrared spectroscopy.

Infrared and Raman Characteristic Group Frequencies 

(Wiley, New York, New York, 3rd Ed., 2001).

Solution to the Last Infrared Spectral Interpretation Challenge

	The spectrum from the last interpretation challenge is seen in Figure i.

	As we work our way from left to right across this spectrum, the first peak we encounter is at 3043. We have learned that peaks between 3100 and 3000 are typically the C-H stretch of unsaturated carbons (15). However, this peak by itself cannot help us distinguish between the two most common unsaturated carbon types, alkenes or benzene rings. Alkenes have a C=C stretching peak from 1680 to 1630 (15), but, given that there are no peaks in this region in this spectrum, we can rule out an alkene. Benzene rings have ring mode peaks that are sharp, and appear from 1630 to 1400 (16). The sharp peak at 1500 can be assigned as a ring mode, indicating we have a benzene ring present.

	The most notable feature in Figure i is the pair of peaks at 1777 and 1726. Given their position and intensity, they are C=O stretching vibrations (8). The fact that there are two carbonyl stretching peaks indicates there may be a functional group present with two C=O bonds in it. Acid anhydrides are a possibility here, because we have seen that their spectra possess a double carbonyl stretching peak (12). However, the acid anhydride functional group also contains a C-O bond, and, hence, should have a strong C-O stretching peak between 1060 and 1035. There are no strong peaks in this region, meaning this is not an acid anhydride.

	We saw last time (13) that cyclic imides have a double carbonyl stretch with peaks at 1790 to 1735 and 1750 to 1680. The spectrum in Figure i has peaks at 1777 and 1726, consistent with there being a cyclic imide present.

	Now, some imides have an N-H bond, and, hence, an NH stretching peak at 3200±50 (13). The absence of an NH stretch here means there is not an NH bond, but, more than likely, a C-N bond in the imide functional group. Note that the caption in Figure i says the sample is a fiber. This means the sample is a polymer, and we probably have a polyimide where the repeat units connect through the nitrogen in the imide functional group, explaining the lack of an NH stretch.

	I am afraid this is as far as we can push this analysis. What we know is that we have a cyclic aromatic polyimide. The most common material fitting this description has the trade name Kapton, or the chemical name poly(4,4’-oxydiphenylene-pyromellitimide). Given this name, you can see why we will simply call this material Kapton. The chemical structure of Kapton is seen in Figure ii.

	Note there is also an aromatic ether in this structure, hence the C-O stretch in Figure i at 1241 (14).

is the founder and CEO of Big Sur Scientific, a maker of portable mid-infrared cannabis analyzers. He has over 30 years experience as an industrial infrared spectroscopist, has published numerous peer reviewed papers, and has written three books on spectroscopy. As a trainer, he has helped thousands of people around the world improve their infrared analyses. In addition to writing for 

and sits on its editorial board. He earned his PhD in physical chemistry from Dartmouth College. He can be reached at: SpectroscopyEdit@MMHGroup.com â

Here we delve into the interpretation of another organic nitrogen compound, known as polyurethane, a ubiquitous polymer used for wood finishes and foam rubber.

Organic Nitrogen Compounds IX: Urethanes and Diisocyanates

We continue our examination of the infrared spectra of organic nitrogen compounds with imides. Imides are a common chemical intermediate, and the engineering polymer Kapton is a polyimide. These molecules have an unusual structure consisting of two carbonyl groups connected by a nitrogen atom. Their spectra are unique as well since some imides exhibit two C=O stretching peaks. Infrared spectroscopy can be used not only to identify imides, but distinguish between straight chain and cyclic imides as well.

	
	We are on installment eight of our series on the infrared spectra of organic nitrogen compounds. Who would have thought there were so many of them? The current installment looks at the spectra of imides. These molecules are not to be confused with amides, which we studied in the last two installments (1,2), even though their names and structures are similar. Amides have many pronunciations (2), my preferred one being 

 where the first syllable rhymes with the “im” in “immobile,” and the last syllable rhymes with the word “id,” which Freud told us is a part of the human psyche (3).

	The structural framework of an imide is seen in Figure 1.

	Recall that amides (1,2) have a carbonyl group attached to a nitrogen. An imide, however, is like an amide on steroids, as it has two carbonyl groups connected by a central nitrogen atom, as seen in Figure 1. As we will see, the spectra of amides and imides have similarities. Imides also have some similarity to acid anhydrides. Recall (4) that an acid anhydride is two carbonyl groups connected by a central oxygen atom. The double carbonyl group in acid anhydrides gives rise to a double carbonyl stretch. Some imides exhibit this type of spectral feature as well.

	Note that the central nitrogen atom in the imide functional group is called the 

. This nitrogen atom may have a hydrogen atom bonded to it, as shown in Figure 1, or have a carbon bearing substituent such as an alkyl chain or phenyl group attached. In either case, the molecule is still classified as an imide. Imides come in two varieties; straight chain and cyclic. Below, we will discuss how to use infrared spectroscopy to distinguish these two imide types from each other.

	The infrared spectrum of a cyclic imide, phthalimide, is seen in Figure 2. Note that this imide has a hydrogen attached to the imide nitrogen. Imides like this will exhibit an imide N-H stretch, which falls at 3200±50 cm

 (going forward, all peak positions will be in cm

, even if not noted). This peak is labeled A in Figure 2. The N-H stretching peak position is the same for both cyclic and straight chain imides.

	Note that, like the N-H stretches of amines (5) and amides (1,2), this peak is of medium width, medium intensity, and shows up in the same region as O-H stretches. However, it is easy to distinguish OH from NH stretches since the former are much more intense and broader than the latter. Of course, if there is a carbon atom attached to the imide nitrogen, there will be no N-H bond, and hence no N-H stretching peak.  Thus, unfortunately not all imides have an N-H stretching peak.

	One of the spectral signatures of a cyclic imide is a double carbonyl stretch. These peaks are labeled B in Figure 2, and fall at 1774 and 1745. In general, these two peaks are found from 1790 to 1735, and 1750 to 1680 for cyclic imides. Acid anhydrides also have a double carbonyl stretch that falls in this range (4). However, acid anhydrides do not contain nitrogen, and hence will not exhibit an N-H stretch. Additionally, acid anhydrides have a strong C-O stretching peak that is missing from imides, because they do not contain this kind of bond. Thus, the diagnostic pattern for a cyclic imide is a single N-H stretch (when present) in combination with a double carbonyl stretch.

	Note from the structure of phthalimide that all its carbons are unsaturated. Recall (6) that such molecules only have C-H stretches above 3000. This is the case in Figure 2, where the aromatic C-H stretching peak is at 3064, and there are no C-H stretching peaks below 3000. This C-H stretching peak appears as a shoulder on the broader N-H stretching envelope.

	Unlike cyclic imides, straight chain imides have only one C=O stretch that falls from 1740-1670. This means cyclic and straight chain imides can be distinguished from each other, since cyclic imides have two carbonyl stretches, whereas straight chain imides have one C=O stretching peak. However, the N-H and C=O stretching peaks of straight chain imides fall in the same range as secondary amides (1). This means it may be difficult to distinguish between a straight imide and secondary amide based on their infrared spectra by themselves. In situations such as this, other measurements, such as physical properties, mass spectra, NMR spectra, or Raman spectra may be needed to get the job done. This is nothing to be alarmed at. As I have been mentioning throughout this column series (7), determining the complete structure of a molecule solely from its infrared spectrum is not always possible. Other data are sometimes required, and that is why chemists have developed multiple instrumental techniques to determine molecular structure. In my experience, most of the industrially important imides are cyclic, and you will run across this type of imide more often than straight chain imides. A summary of the imide group wavenumbers is seen in Table I.

	The imide functional group consists of a central nitrogen atom, the imide nitrogen, attached to two carbonyl groups. Imides come in two varieties; cyclic and straight chain. The two can be distinguished from each other because cyclic imides have two carbonyl stretches, whereas straight chain imides have one. The diagnostic peak pattern for cyclic imides is a double carbonyl stretch combined with an N-H stretch (when present). Unfortunately, the infrared spectra of straight chain imides are similar to those of secondary amides, making it sometimes difficult to distinguish between the two based on their infrared spectra alone. Other information may be needed to differentiate between them.

https://en.wikipedia.org/wiki/Id,_ego_and_super-ego#Id

is the founder and CEO of Big Sur Scientific, a maker of portable mid-infrared cannabis analyzers. He has over 30 years of experience as an industrial infrared spectroscopist, has published numerous peer reviewed papers, and has written three books on spectroscopy. As a trainer, he has helped thousands of people around the world improve their infrared analyses. In addition to writing for 

 and sits on its editorial board. He earned his PhD in physical chemistry from Dartmouth College. He can be reached at: 

Here, we continue our examination of the infrared (IR) spectra of organic nitrogen compounds with imides, which are a common chemical intermediate. IR can be used not only to identify imides, but also to distinguish between straight chain and cyclic imides. We explain how.

Organic Nitrogen Compounds VIII: Imides

After introducing amides in the last column, in this installment we look at their infrared spectra. We will see how N-H stretching and bending peaks are used to distinguish primary, secondary, and tertiary amides from each other. We also briefly discuss how infrared spectroscopy can be used to ascertain protein structure.

	
	In our last installment, I introduced you to the amide functional group (1). Here, I tell the rest of their story-specifically, how to interpret their infrared spectra.

	The generic molecular structure of amides is seen in Figure 1. As you may recall, amides are organized into three types: primary, secondary, and tertiary, as seen in Figure 2. The difference between them is the number of C-N bonds, with the three amide types having 1, 2, or 3 of these bonds, respectively. Additionally, the number of N-H bonds goes down as the number of C-N bonds goes up, with primary amides having an NH

 group, secondary amides having an N-H group, and tertiary amides having no N-H bonds.

	Similar to carbonyl groups attached to a benzene ring (2), amides undergo conjugation. As a result, their carbonyl stretch is lower than that of many other similar functional groups, and this peak falls for all amides from 1680 to 1630. Note, however, that we do not speak of “saturated” or “aromatic” amides, as conjugation takes place in all amides, so they all have the same carbonyl stretching peak range. Amides also engage in hydrogen bonding. However, like amines, the strength of the hydrogen bonding is less than that for O-H bonds, giving medium width and intensity N-H stretching peaks compared to O-H stretches.

	The infrared spectrum of a primary amide, benzamide, is seen in Figure 3.

 group, it has two N-H stretching vibrations, whose peaks are seen at 3366 and 3710 (going forward, assume all peaks positions stated are in cm

 units, even if not stated). Note that these peaks fall in the same range as O-H stretches, but you should never confuse the two, because N-H stretching peaks are narrower, and less intense, than O-H stretching peaks. Note that these two peaks come to a nice sharp point, but their bases are broadened due to hydrogen bonding. At their base, these two peaks are about 400 cm

 wide, whereas O-H stretching envelopes are often more than 1000 cm

 wide. For primary amides in general, this pair of peaks falls between 3370 and 3170 cm

 Again, both peaks must be present for a molecule to be a primary amide. The C=O stretch in Figure 3 is seen at 1656, well within the range expected for amides.

	Primary amides have a structural similarity to primary amines in that they both contain the NH

 group. Recall that primary amines have scissoring and wagging bending vibrations (3), and so do primary amides. In Figure 3, the primary amide scissoring peak is seen at 1622, and, in general, this peak falls from 1650 to 1620. The NH

 wag is a broadened envelope, due to hydrogen bonding, around 700. This envelope normally tops out between 750 and 600. The pair of sharp peaks on top of this envelope are the C-H wag and ring bend of a mono-substituted benzene ring (4), which is present in benzamide.

	In a previous column (3), I pointed out that C-N stretches are not useful group wavenumbers, since they are usually medium-to-weak in intensity, and fall from 1400 to 1000, right in the middle of the very busy fingerprint region, where it is easy for them to get lost. The C-N stretch of benzamide is seen at 1398. This particular molecule has a relatively simple spectrum, so this peak is visible, but you can see how it might get lost if there were other peaks of similar size around it. A summary of the group wavenumbers for primary amides is seen in Table I.

	The secondary amide linkage is important, and is found all around and within us. A number of polymeric materials, including the nylon family, contain secondary amide linkages. Also, the amino acid units in proteins are held together by secondary amide linkages. The infrared spectrum of a secondary amide, Nylon-6,6 is seen in Figure 4.

	In nylon nomenclature, the two numbers refer to the number of carbons between the nitrogens. For Nylon-6,6, there are six carbons between the nitrogens, for Nylon-4,4, there would be four, and so on.

	Because secondary amides contain one N-H bond, there is only one N-H stretching peak, seen at 3301 in Figure 4. The easiest way then to distinguish a primary amide from a secondary amide is to remember that, because a primary amide has a NH

 group, it has two N-H stretching peaks, whereas a secondary amide has one N-H bond, and hence one N-H stretching peak. Note the peak at 3301 in Figure 4 is of medium intensity and width, like all the other N-H stretching peaks we have seen. For secondary amides, this peak generally falls from 3370 to 3170.

	The C=O stretch of our secondary amide is seen in Figure 4 at 1641, in the 1680 to 1630 range typical of all amides. Note, however, it has a companion peak at 1542. This is the in-plane N-H bend of the secondary amide group, and is normally found from 1570 to 1515. This peak’s position is unusual in that there are very few group wavenumber peaks that fall around 1550, and it is unusually intense for an in-plane bend. We have said very little about in-plane bends throughout this column series, because they are generally weak and typically fall in the middle of the busy fingerprint region where they are easily lost. This in-plane bend is higher in wavenumber and more intense than others because of the conjugation that takes place in amides (5). The C=O and N-H in plane bend peaks form a diagnostic pair of sharp, intense peaks in the middle of the spectrum. Combine these with the single N-H stretch of secondary amides and you have a trio of peaks that clearly indicate when a secondary amide is present in a sample. For the nylon family of polymers particularly, this pair peaks falls at ~1640 and ~1540, and are easily spotted. The N-H wag of secondary amides forms a broadened envelope from 750 to 680 and is seen in Figure 4 at 691. The summary of secondary amide group wavenumbers is seen in Table II.

	Proteins are polymers that consist of amino acid repeat units. The repeat units are linked together by secondary amide linkages. The spectrum of a protein, keratin, is seen in Figure 5.

	Mammal hair, skin, feathers, and reptile scales are all made from keratin, so it is a commonly found material. The spectrum in Figure 5 is of sheep hair (wool). Note that the spectrum is dominated by secondary amide peaks, with N-H stretch, C=O stretch, and N-H in-plane bend peaks clearly visible.

	I am often asked by my students in the infrared spectral interpretation training courses I teach whether infrared spectroscopy can be used to distinguish between natural and synthetic materials. I am afraid that, in general, the answer is no. Remember that infrared spectroscopy is sensitive to the functional groups present in a sample. Functional groups generally have the same structure, whether they are made in a test tube or a living cell. Figure 5 shows this; if I didn’t know this spectrum were of a protein, I might have thought it was a synthetic secondary amide, such as a nylon.

	What about tertiary amides? As seen in Figure 1, this functional group has no N-H bonds, hence no N-H stretching and bending peaks, and hence no useful group wavenumbers. Tertiary amides will exhibit a C=O stretch from 1680 to 1630 like all other amides, but there are many conjugated C=O functional groups whose carbonyl peaks also fall in this range, meaning there is nothing unique about the spectra of tertiary amides. It is an example of a functional group that are not readily detected by infrared spectroscopy.

	Amides come in primary, secondary, and tertiary forms. The diagnostic pattern of peaks for primary amides is a pair of N-H stretching peaks and a C=O stretch. The pattern for secondary amides is one N-H stretch, a C=O stretch, and an intense N-H in-plane bend next to the C=O stretch. Proteins are an important class of molecules whose spectra are dominated by secondary amide peaks. Unfortunately, there is not a good way of using infrared spectroscopy to distinguish natural from synthetic secondary amides. Tertiary amides are difficult to detect via infrared spectroscopy.

Your Next  Infrared  Spectral  Interpretation Challenge

	As promised, the Infrared Spectral Interpretation Challenge is back from its holiday break. Using everything you have learned from this and other columns in this series, do your best to identify the functional groups present, and then identify the complete chemical structure of this molecule.

, Dr. Smith writes a regular column for its sister publication 

The N-H stretching and bending peaks can be used to distinguish primary, secondary, and tertiary amides and to ascertain protein structure. Here’s how.

Author | Brian C. Smith

Articles