Nucleophilic Reactions Of Carbonyl Groups

Nucleophilic Reactions Of Carbonyl Groups

Nucleophilic Reactions Of Carbonyl Groups: Aldehydes And Ketones

The carbon-oxygen double bond C=O is termed a carbonyl group and represents one of the most important reactive functional groups in chemistry and biochemistry. Since oxygen is more electronegative than carbon, the electrons in the double bond are not shared equally and the carbon-oxygen bond is polarized, with the oxygen atom attracting more of the electron density. This polarization may be represented via the resonance structures A and B, where A is uncharged and B has full charge separation.

However, since the contribution from B is smaller than that from A, the charge distribution is often presented as in C. The partial charges δ+ and δ− indicate the imbalance in electron density.

Nucleophilic Addition:

Electrophilic Addition:

• Ketones have two alkyl or aryl groups attached to the carbonyl group; in aldehydes, one or both of these groups is hydrogen (the simplest aldehyde is formaldehyde H₂C=O).
• The carbon atom is sp2 hybridized, so the carbonyl group and the directly bonded atoms are in one plane (see Section 2.6.2). The two pairs of nonbonding electrons, i.e. lone pairs, on the carbonyl oxygen tend to be omitted in most representations; but, as we shall see, these frequently play an important part in mechanisms involving carbonyl groups.
• For convenience, or out of sheer laziness, we may show only one of these lone pairs even when they do play a role. Because of the polarization, the carbonyl group can be involved in both nucleophilic and electrophilic reactions.
• Most addition reactions involve both steps, but the order in which these occur depends on the nature of the reagent and the reaction conditions. Under basic conditions, the nucleophile attacks the carbonyl first, and the reaction is completed by the abstraction of an electrophile, a proton, from the solvent.

• Under acidic conditions, electrophilic addition occurs first, namely the carbonyl’s protonation and the conjugate acid’s formation.
• The conjugate acid, with a full positive charge, is now a more reactive electrophile than the original uncharged carbonyl group, which only has partial charge separation due to polarization.
• As a result, addition can now occur with less-reactive nucleophiles, and typically these are uncharged and attack via their lone pair electrons.

• In most of the reactions that we shall encounter, there will be an attack of a charged nucleophile onto an uncharged carbonyl or an attack of lone pair electrons in an uncharged nucleophile onto a charged conjugate acid. An uncharged nucleophile tends to be insufficiently reactive for addition reactions to occur with an uncharged carbonyl.
• At the other extreme, the combination of charged nucleophiles and charged carbonyl is not usually favorable. Since negatively charged nucleophiles are also bases, an acidic environment will not permit their involvement.
• The most significant change in these reactions is the formation of the carbon–nucleophile bond; so, in both types of mechanism, the reaction is termed a nucleophilic addition. It should be noted that the polarization in the carbonyl group leads to nucleophilic addition, whereas the lack of polarization in the C=C double bond of an alkene leads to electrophilic addition reactions (see Chapter 8).
• Carbonyl groups in carboxylic acid derivatives undergo a similar type of reactivity to nucleophiles, but the presence of a leaving group in these compounds leads to substitution reactions rather than addition.

Aldehydes Are More Reactive Than Ketones

The reactions undergone by aldehydes and ketones are essentially the same, but aldehydes are more reactive than ketones. There are two rational reasons for this. Alkyl groups have an electron-donating inductive effect (see Section 4.3.3) and the presence of two such groups in ketones against just one in aldehydes means the magnitude of δ+ is reduced in ketones. Put another way, the carbonyl group in aldehydes is more electrophilic than that in ketones.

It should also be noted that aromatic aldehydes, such as benzaldehyde, are less reactive than alkyl aldehydes. This is because the aromatic ring allows electron delocalization via a resonance effect that also reduces the positive charge on the carbonyl carbon.

• The second feature is a steric consideration. During nucleophilic addition, the planar SP² system of the carbonyl compound (bond angle 120◦) is converted This crowding is more severe with two alkyl substituents (from ketones) than with one alkyl and the much smaller hydrogen (from aldehydes).
• A consequence of this change is that the planar aldehyde or ketone can be attacked from either side of the plane with essentially equal probability. If the substituents are all different, then this will result in the creation of a chiral center; but, since both enantiomers will be formed in equal amounts,
• The product will be an optically inactive racemate (assuming no other chiral centers are present in the R groups). into a tetrahedral sp³ system in the product (bond angle 109◦) creating more steric crowding, i.e. the groups are brought closer together.

Nucleophi Les And Leavi Ng Groups: Reversible Addition Reactions

In principle, all carbonyl addition reactions could be reversible; but, in practice, many are essentially irreversible. Let us consider mechanisms for the reverse of the nucleophilic addition reactions given above. For the base-catalyzed reaction, we would invoke the following mechanism:

• It becomes clear that, in the reverse reactions, we need the original nucleophile to behave as a good leaving group, either as Nu− in base-catalyzed reactions or as Nu–H in acid-catalyzed situations. Conversely, if the nucleophile cannot act as a leaving group, then the reverse reaction is going to be unfavorable and the addition will be essentially irreversible.
• By appreciating this fundamental concept, we shall be able to rationalize the various carbonyl addition reactions of importance described below. We shall also be able to link easily the behavior of carboxylic acid derivatives, where the presence of an alternative leaving group needs to be considered.

• Reversible reactions include the addition of water, alcohols, thiols, HCN, and amines. Irreversible reactions include the addition of hydride and organometallics. In the latter cases, hydride H^– and carbanions such as Me^– are going to be very poor leaving groups, predictable from the pKa values of H₂ (35) and MeH (48). We have seen in that good leaving groups are the conjugate bases of strong acids.
• We can also rationalize why some addition reactions simply do not occur, e.g. halide ions do not add to carbonyl groups. Although we know that a halide such as bromide can act as an effective nucleophile in SN₁ and SN₂ reactions (see Section 6.1.2),
• it is also a very good leaving group (pKa value for HBr −9). This means that the reverse reaction becomes very much more favorable than the forward reaction. In cases where both forward and reverse reactions are feasible, we can often usefully disturb the equilibrium by using an excess of one reagent (see below).

Oxygen as a Nucleophile Hemiacetals, Hemiketals, Acetals And ketals

The addition of 1 mol of alcohol to an aldehyde gives a hemiacetal, and to a ketone a hemiketal. However, most chemists do not now differentiate between hemiacetals and hemiketals; these are both termed hemiacetals. This reaction is usually catalyzed by acids, but may also be achieved in the presence of a base.

The reactions follow the general mechanisms given above. The main difference between the acid-catalyzed and base-catalyzed mechanisms is that acid increases the electrophilicity of the carbonyl group by protonation, whereas base increases the nucleophilicity of the alcohol via ionization to the conjugate base.

Base-catalyzed Formation Of Hemiacetals:

Each reaction is reversible, and by extrapolating from the forward reaction it is relatively easy to propose a mechanism for the reverse reaction. Mechanisms are shown here for both acid- and base-catalyzed decomposition of hemiacetals.

Acid-Catalyzed Decomposition Of Hemiacetals:

Base-catalyzed Decomposition Of Hemiacetals:

The position of equilibrium, i.e. whether the carbonyl compound or the addition product is favored, depends on the nature of the reagents. The equilibrium constant is often less than 1, so the product is not favored, and many simple hemiacetals and hemiketals are not sufficiently stable to be isolated.

However, stable cyclic hemiacetals and hemiketals can be formed. A cyclic product arises if the alcohol function is in the same molecule as the carbonyl, allowing an intramolecular reaction rather than an intermolecular one. When these functional groups are separated by three or four carbons, this results in the generation of stereochemically favorable five- or six-membered rings respectively.

Hydroxyaldehyde–Cyclic Hemiacetal Equilibria

However, as seen in Table 7.1, rings other than five and six-membered are either not formed or are not particularly favored.

A vast range of natural sugars exemplify these cyclic addition products. A typical sugar exists predominantly in the form of a hemiacetal or hemiketal in solution, although this is an equilibrium reaction, and the open-chain carbonyl form is always present to a small extent (<1%).

• The formation of a six-membered cyclic hemiacetal from glucose is achieved by attack of the C-5 hydroxyl onto the protonated carbonyl (conjugate acid). The cyclic form of glucose is termed glucopyranose since the new ring system is a reduced form of the oxygen heterocycle pyran.
• Nucleophilic attack onto the planar carbonyl may occur from either of its two faces, generating two different stereochemistries at this new chiral center, designated as α or β. This new chiral center is termed the anomeric center. Since there are other chiral centers in the molecule, the mixture of α- and β-anomeric forms is not a racemate, but a mixture of diastereoisomers (see Section 3.4.4).
• The mixture does not contain 50% of each anomer (see below). Although both forms are produced, the β form with the equatorial hydroxyl is thermodynamically favored (see Section 3.3.2).
• Further, the two forms can also equilibrate via the open-chain carbonyl form of the sugar, so that the single isomers in solution are rapidly transformed into the equilibrium mixture (see Box 7.1). Since there are two anomeric forms, these are often in equilibrium via the acyclic carbonyl compound.
• we use a new type of bond to indicate that the configuration is not specified, and could be of either stereochemistry. This is the wavy or wiggly bond; and to emphasize our indecision further, it is usually sited halfway between the two possible positions.

The Mutarotation Of Glucose

It is possible to separate the two anomeric forms of glucose by careful recrystallization from water. The two forms have different specific optical rotations (see Section 3.4.1), [ α]D + 112◦ for α-D-glucopyranose, and [ α]D + 18.7 for β-D-glucopyranose. If either of these forms is dissolved in water, the optical rotation slowly changes to yield the same final rotation, [ α]D + 52.7◦. Because this process produces a change in rotation from that of either pure substance, it is termed mutarotation.

The final product is an equilibrium mixture of the α and β anomeric forms, and their interconversion involves the open-chain aldehyde form as shown.

A similar transformation is common to all aldehyde-containing hexoses (aldohexoses) and pentoses (aldopentoses); see Section 12.2.3. From these data, it is also easy to calculate the proportions of the two forms in the equilibrium mixture.

• Since we are dealing with isomeric compounds, we can relate specific rotations to the amount of each isomer. If the fraction of the α form is x and that of the β form is 1 − x, then 112x + 18.7(1 − x) = 52.7 From this one can calculate that x = 0.36; there is thus 36% α anomer and 64% β anomer in the equilibrium mixture.
• The approximately 2: 1 preference for the β anomer is consistent with our knowledge of conformations in six-membered rings (see Section 3.3.2); this anomer has the variable hydroxyl in a more favorable equatorial position.
• Note that the difference in thermodynamic stability is not sufficient to force the equilibrium completely in favor of the β anomer, and it is perhaps unexpected that there is quite so much of the α anomer present at equilibrium. We shall return to this topic, the anomeric effect, in Section 12.4.
• Glucose is also capable of forming a five-membered hemiacetal ring by the attack of the 4-hydroxyl onto the carbonyl, though this is much less favorable than the formation of the six-membered ring just discussed (see Section 12.2.2).
• Five-membered rings are termed furanose rings by comparison with the oxygen heterocycle furan, with the most prominent example being that from the five-carbon sugar ribose.

The two anomeric forms are called α- and β- ribofuranose. Again, in solution, there exists an equilibrium between the open-chain carbonyl form and the two anomeric hemiacetal forms. Ribose also forms six-membered pyranose anomers, and an aqueous solution contains about 76% pyranose forms to 24% furanose forms. In the vast majority of cases, ribose is found in nature combined with the β-furanose form.

Some Biologically Important Ribose Derivatives

Nature has exploited ribose derivatives for several crucially significant biochemicals. Many of these contain a heterocyclic base attached to the β-anomeric position of D-ribofuranose and are termed nucleosides. Adenosine, guanosine, cytidine, and uridine are fundamental components of ribonucleic acids.

and similar derivatives of 2-deoxy- β-D-ribofuranose, the deoxynucleosides deoxyadenosine, deoxyguanosine, deoxycytidine, and deoxythymidine, are the building blocks of deoxyribonucleic acids (DNA; see Section 14.1). Nucleosides are also encountered in the structures of adenosine triphosphate (ATP) and coenzyme A (HSCoA).

ATP provides nature with its currency unit for energy. Hydrolysis of ATP to adenosine diphosphate (ADP) liberates energy, which can be coupled to energy-requiring processes in biochemistry, and synthesis of ATP from ADP can be coupled to energy-releasing processes.

Coenzyme A is used as the alcohol part of thioesters, which are more reactive than oxygen esters (see Section 7.9.3) and are thus exploited in biochemistry in a wide range of reactions, e.g. fatty acid biosynthesis and metabolism

The addition of 2 mol of an alcohol to an aldehyde gives an acetal, and to a ketone a ketal. Most chemists do not now differentiate between acetals and ketals; these are both termed acetals. These products are formed by further reaction of the hemiacetal or hemiketal with a second molecule of alcohol. In contrast to hemiacetal and hemiketal formation, this reaction is catalyzed only by acids, not by the base.

Acid-Catalysed Formation Of Acetals:

Initially, the reaction involves the protonation of one of the oxygen atoms, followed by the loss of this group as a neutral molecule and the formation of a resonance-stabilized carbocation. If the oxygen protonated were that of the alkoxy group, then the product would merely be the protonated aldehyde, and the reaction becomes a reversal of hemiacetal formation. Only when the oxygen of the hydroxyl is protonated can the reaction lead to an acetal, and this requires a nucleophilic attack of the second alcohol molecule onto the alternative resonance-stabilized carbocation.

• This is a further example of a carbonyl–electrophile complex, and equivalent to the conjugate acid so that the subsequent nucleophilic addition reaction parallels that in hemiacetal formation. Loss of the leaving group occurs first in an SN1-like process with the cation stabilized by the neighboring oxygen; an SN2-like process would be inhibited sterically.
• It is also possible to rationalize why base catalysis does not work. The base would simply remove a proton from the hydroxyl to initiate hemiacetal decomposition back to the aldehyde – what is needed is to transform the hydroxyl into a leaving group, hence the requirement for protonation.
• The reactions are reversible, including that which regenerates the aldehyde, and the equilibrium must be disturbed to achieve good conversion to an acetal. This is usually achieved by using an excess of the alcohol; if we are using a simple alcohol like methanol or ethanol, we might perhaps employ it as the solvent.
• Acetals and ketals are usually stable but are readily hydrolyzed back to aldehydes and ketones by acid hydrolysis, a reversal of the synthetic procedure. This makes acetal or ketal formation a valuable means of protecting an aldehyde or ketone carbonyl from reaction with other reagents being used during a synthetic procedure.
• For instance, protection of a ketone group may be achieved by forming a cyclic ketal with an excess of ethylene glycol; when protection is no longer required, this protecting group may be removed by acid-catalyzed hydrolysis using an excess of water.
• By having the two alcohol functions in the same molecule, the formation of a ketal from the intermediate hemiketal becomes favorable, since it requires an intramolecular reaction rather than an intermolecular one. In the example shown, it is feasible under mildly acidic conditions to carry out these reactions on a ketoester without affecting the ester function.

Of course, we can use the same strategy to protect a diol. We would convert this into an acetal or ketal using a suitable aldehyde or ketone.

Acetal and ketal linkages are widely found in natural sugars and polysaccharides. The structure of sucrose is a splendid example.

• Sucrose is a disaccharide, composed of two linked monosaccharide units, glucose in pyranose ring form and fructose in furanose ring form.
• As we have seen above, glucopyranose is a hemiacetal derived from the aldehyde-containing sugar glucose.
• In sucrose, it is present as its α anomer. On the other hand, fructose is a ketone-containing sugar, and it forms a hemiketal furanose ring by reaction of the C-5 hydroxyl with the ketone group.

• In sucrose, fructose is present as the β anomer. Now, one of these sugars has acted as an alcohol to make a bond to the other sugar. We can look at this in two ways. Either fructose acts as an alcohol to react with the hemiacetal glucose to form an acetal, or glucose is the alcohol that reacts with the hemiketal fructose to form a ketal.
• In sucrose, the pyranose ring is an acetal, whilst the furanose ring is a ketal. This all seems rather complicated at first – look carefully at the structures whilst considering the text. In an aqueous solution, both glucose (hemiacetal) and fructose (hemiketal) exist as equilibrium mixtures of cyclic and open-chain carbonyl forms.
• Sucrose, however, is a single stable substance (acetal and ketal), and conversion back to glucose and fructose requires more rigorous hydrolytic conditions, such as heating with aqueous acid.

Invert Sugar

Invert sugar is the name given to an equimolar mixture of glucose and fructose, obtained from sucrose by hydrolysis with acid or using the enzyme invertase. During the process, the optical activity changes from that of sucrose, [ α] D + 66.5◦, to that resulting from an equal mixture of glucose and fructose.

• The equilibrium mixture of α and β isomers of glucose has [ α]D + 52.7◦ (see Box 7.1), and that of fructose is strongly laevorotatory with [ α]D − 92.4◦. Since glucose and fructose are structural isomers and have the same molecular weight, it can be calculated that the resultant optical activity will be [ α] D − 19.85◦.
• This derives from (−92.4) × 0.5 + (+52.7) × 0.5, the 0.5 factor being necessary because 1 g of the disaccharide sucrose gives 0.5 g of each monosaccharide.
• This change in optical activity from plus to a minus is the reason for the terminology ‘invert’. The high sweetness of fructose combined with that of glucose means inverted sugar is sweeter than sucrose, so it provides a cheaper, less calorific sweetener than sucrose.
• The relative sweetness figures for sucrose, glucose, and fructose are 1.0, 0.7, and 1.7 respectively. Honey is also composed mainly of inverted sugar.

Polysaccharides: Starch, Glycogen And Cellulose Are Polyacetals Of Glucose

Polysaccharides fulfill two main functions in living organisms, as food reserves and as structural elements. Plants accumulate starch as their main food reserve, a material that is composed entirely of glucopyranose units but in two types of molecule. Amylose is a linear polymer containing some 1000–2000 glucopyranose units linked through α1 → 4 acetal groups. This terminology means that the 1-position of the first ring is linked to the 4-position of the second ring, and the configuration at the anomeric center in the first ring is α.

Amylopectin is a much larger molecule than amylose (the number of glucose residues varies widely, but may be as high as 106), and it is a branched-chain molecule. In addition to α1 → 4 acetal linkages, amylopectin has branches at about every 20 units through α1 → 6 acetal linkages, i.e. similar acetal bonding but to the 6-hydroxyl of another glucose residue.

These branches continue with α1 → 4 linkages but then may have subsidiary branching giving a tree-like structure. The mammalian carbohydrate storage molecule is glycogen, which is analogous to amylopectin in structure, but is larger and contains more frequent branching, about every 10 residues.

Cellulose is reputedly the most abundant organic material on Earth, being the main constituent in plant cell walls. It is composed of glucopyranose units linked β1 → 4 in a linear chain, i.e. this time the configuration at the anomeric center is β. Alternate residues are found to be ‘rotated’ in the structure, allowing hydrogen bonding between adjacent molecules, and construction of the strong fibers characteristic of cellulose, for example in cotton.

Acetallinkages In Etoposide

Etoposide is an effective anticancer drug used in the treatment of small-cell lung cancer, testicular cancer, and lymphomas. It is a semi-synthetic modification of the natural lignan podophyllotoxin and contains three acetal linkages. Can you identify them?

• Initially, it is possible to see the cyclic form of glucose as a component of the structure. This is normally a hemiacetal, but here is further bound to an alcohol derived from the podophyllotoxin derivative through an acetal linkage.
• Secondly, it should be noted that two of the hydroxyl groups of glucose are also bound as a cyclic acetal to acetaldehyde; this linkage can be formed because the two hydroxyls of glucose are suitably positioned and allow a favorable six-membered ring to be constructed.
• The third acetal linkage is not so obvious, and it is in the five-membered ring fused onto the aromatic ring of the podophyllotoxin derivative. This is called a methylenedioxy group, and it is a common bidentate substituent on many natural aromatic structures. However, it can be formally regarded as an acetal of formaldehyde.
• Nature does not make a methylenedioxy group using formaldehyde. Instead, it modifies an existing ortho-hydroxy-methoxy arrangement. Enzymic hydroxylation of the methoxy methyl converts this substituent into what is identical to a hemiacetal of formaldehyde, and then acetal formation follows in a process analogous to a chemical synthesis.
• The hydroxylating enzyme involved is a cytochrome P-450 mono-oxygenase.

Water As A Nucleophile: Hydrates

Water, as the simplest alcohol, should also be able to act as a nucleophile towards aldehydes and ketones and produce a g e m-diol, sometimes termed a hydrate. The prefix gem is an abbreviation for geminal (Latin Gemini: twins); we use it to indicate two like groups on the same carbon. However, for most aldehydes and ketones, the equilibrium is unfavorable, and the reaction is not important.

The equilibrium only becomes favorable if the δ+ charge on the carbon of the carbonyl can be increased. Since alkyl groups have a positive inductive effect and decrease the δ+ charge,

• we need to have no alkyl groups, such as in formaldehyde, or a functional group with a negative inductive effect that destabilizes the carbonyl group.
• The equilibrium percentages of hydrate for formaldehyde, acetaldehyde, and acetone in water are found to be about 100, 58, and 0 respectively

Formaldehyde is normally a gas at room temperature but dissolves in water. In an aqueous solution, formaldehyde exists almost entirely as the gem-diol; a 37% solution is called formalin and is used for preserving biological tissues. However, the hydrate cannot be isolated, since the reverse reaction is rapid and the hydrate decomposes to formaldehyde.

If there is a suitable electron-withdrawing substituent, hydrate formation may be favored. Such a situation exists with trichloroacetaldehyde (chloral). Three chlorine substituents set up a powerful negative inductive effect, thereby increasing the δ+ charge on the carbonyl carbon and favoring nucleophilic attack.

• Hydrate formation is favored, to the extent that chloral hydrate is a stable solid, with a history of use as a sedative.
• These observations emphasize the fact that gemdiols are usually unstable and decompose into carbonyl compounds. However, it can be demonstrated that hydrate formation does occur by exchange labeling of simple aldehyde or ketone substrates with 18Olabelled water.
• Thus, after equilibrating acetone with labeled water, isotopic oxygen can be detected in the ketone’s carbonyl group.

Sulfur As A Nucleophile Hemithioacetals, Hemithioketals, Thioacetals And Thioketals

The reaction of thiols with aldehydes and ketones parallels that of alcohols. However, the reactions are more favorable because sulfur is a better nucleophile than oxygen. Electrons in larger atoms are more easily polarizable and it becomes easier for them to be donated to an electrophile.

As a consequence, thiols are preferred to alcohols for the protection of aldehyde and ketone groups in synthetic procedures. Thioacetals and thioketals are

Excellent protecting groups. They are more readily formed and are more stable to hydrolytic conditions than acetals and ketals.

Hydride As A Nucleophile: Reduction Of Aldehydes And Ketones, Lithium Aluminium Hydride A Nd Sodium Borohydride

The carbonyl group of aldehydes and ketones may be reduced to an alcohol group by a nucleophilic addition reaction that appears to involve hydride as the nucleophile. The reduction of the carbonyl group may be interpreted as a nucleophilic attack of hydride onto the carbonyl carbon, followed by abstraction of a proton from solvent, usually water.

• This is not strictly correct, in that hydride, from say sodium hydride, never acts as a nucleophile, but because of its small size and high charge density, it always acts as a base.
• Nevertheless, there are several complex metal hydrides such as lithium aluminum hydride (LiAlH4; LAH) and sodium borohydride (NaBH4) that deliver hydride in such a manner that it appears to act as a nucleophile.
• We have already met these reagents under nucleophilic substitution reactions (see Section 6.3.5). Hydride is also a very poor leaving group, so hydride reduction reactions are also irreversible.

Whilst the complex metal hydride is conveniently regarded as a source of hydride, it never actually produces hydride as a nucleophile, and it is the aluminum hydride anion that is responsible for the transfer of the hydride. Then, the resultant negatively charged intermediate complexes with the residual Lewis acid AlH₃.

• This complex can also transfer hydride to another molecule of the carbonyl compound in a similar manner, and the process continues until all four hydrides have been delivered. Since all four hydrogens in the complex metal hydride are capable of being used in the reduction process, 1 mol of reducing agent reduces 4 mol of aldehyde or ketone.
• Finally, the last complex is decomposed by the addition of water as a proton source. Lithium aluminum hydride reacts violently with water, liberating hydrogen, and must therefore be used in rigorously anhydrous conditions, usually in ether solution. Any solvent containing OH or NH groups would destroy the reagent by acting as a proton donor for hydride.
• In the case of LAH reductions, the addition of water as the proton source has to be carried out with considerable caution since any unreacted LAH will react violently with this water. In the laboratory, safe removal of excess LAH may be achieved by initially adding small amounts of an ester, such as ethyl acetate.

An analogous series of reactions is involved when sodium borohydride is used as the reducing agent. Sodium borohydride is considerably less reactive than LAH and may be used in alcoholic or even aqueous solutions, so there are no particular problems associated with its use.

• All four hydrides in LAH and NaBH4 may be exploited in the reduction of the carbonyl compounds, the intermediate complexes also being reducing agents. However, these complexes become sequentially less reactive than the original reagent, and this has led to the development of other complex metal hydride-reducing agents that are less reactive and, consequently, more selective than LAH.
• They are produced by treating LAH with various amounts of an alcohol ROH, giving compounds with the general formulae (RO)MH₃−, (RO)₂MH₂−, and (RO)₃MH− as their anionic component. These provide a range of reducing agents with different activities. LAH itself is a powerful reducing agent and will react with several other functional groups (see Sections 7.7.1 and 7.11).
• Note that LAH does not reduce carbon-carbon double bonds; these double bonds lack the charge separation that distinguishes the carbonyl group, and there is no electrophilic character to allow a nucleophilic attack. An effective way of reducing C=C is catalytic hydrogenation.

Nicotinamide Adenine Dinucleotide As Reducing Agent

The biological reduction of aldehydes and ketones is catalyzed by an appropriate enzyme, a dehydrogenase or reductase, and most of these use a pyridine nucleotide, such as the reduced form of nicotinamide adenine dinucleotide (NADH), as the cofactor. This cofactor may be considered as the reducing agent, capable of supplying hydride similarly to lithium aluminum hydride or sodium borohydride (see Section 7.5).

• NADH is a complex molecule (see Box 11.2), and only the dihydropyridine ring part of the structure is considered here. Some reactions employ the alternative phosphorylated cofactor NADPH; the phosphate does not function in the reduction step but is merely a recognition feature helping to bind the compound to the enzyme.
• The biological reduction of aldehydes and ketones is catalyzed by an appropriate enzyme, a dehydrogenase or reductase, and most of these use a pyridine nucleotide, such as the reduced form of nicotinamide adenine dinucleotide (NADH), as the cofactor.
• This cofactor may be considered as the reducing agent, capable of supplying hydride similarly to lithium aluminum hydride or sodium borohydride (see Section 7.5). NADH is a complex molecule (see Box 11.2), and only the dihydropyridine ring part of the structure is considered here.
• Some reactions employ the alternative phosphorylated cofactor NADPH; the phosphate does not function in the reduction step but is merely a recognition feature helping to bind the compound to the enzyme.

Hydride may be transferred from NADH to the carbonyl compound because of the electron-releasing properties of the ring nitrogen; this also results in the formation of a favorable aromatic ring, a pyridinium system since the nitrogen already carries a substituent.

• The cofactor becomes oxidized to NAD+. The reaction is then completed by the abstraction of a proton from water. There is a rather important difference between chemical reductions using complex metal hydrides and enzymic reductions involving NADH, and this relates to stereospecificity.
• Thus, chemical reductions of a simple aldehyde or ketone will involve hydride addition from either face of the planar carbonyl group, and if reduction creates a new chiral center, this will normally lead to a racemic alcohol product. Naturally, the aldehyde → primary alcohol conversion does not create a chiral center.

In contrast, an enzymic reduction utilizing NADH will be executed stereospecifically, with hydride attaching to one particular face of the planar carbonyl. Which face is attacked depends upon the individual enzyme involved. For example, the reduction of pyruvic acid to lactic acid in vertebrate muscle occurs via an attack of hydride from the Reface (see Section 3.4.7), and produces the single enantiomer (S)-lactic acid. Hydride addition onto the alternative Si face is a feature of some microbial dehydrogenase enzymes.

These enzymes often also catalyze the reverse reaction, the oxidation of an alcohol to an aldehyde or ketone (see Box 11.2). In such reactions, the cofactor NAD+ abstracts hydride from the alcohol, and may thus be regarded as an oxidizing agent; hence the dehydrogenase terminology for some enzymes, even when they are carrying out a reduction.

Carbonasa Nucleophile

Cyani De: Cyanohydrins

Aldehydes and ketones react with HCN to give 2- 2-hydroxynitriles, compounds that are generally termed cyanohydrins. HCN is only a weak acid (pKa 9.1), and proton availability is insufficient to initiate a typical acid-catalyzed reaction via the conjugate acid of the carbonyl compound. Instead, the partial ionization of HCN provides a source of cyanide anions, which then react as a nucleophile towards the carbonyl compound.

The reaction is terminated by the strongly basic alkoxide ion abstracting a proton from the solvent or a further molecule of HCN. To avoid the use of HCN, which is a highly toxic gas, aqueous sodium or potassium cyanides in buffered acid solution are usually employed in the reaction.

The reaction is reversible, and cyanohydrin formation is more favorable with aldehydes than with ketones, as with other addition reactions. The reverse reaction is easily affected by treating a cyanohydrin with an aqueous base since cyanide is a reasonable leaving group.

Cyanohydrin formation is a useful synthetic reaction, in that it utilizes a simple reagent, cyanide, to create a new C–C bond. The cyano (nitrile) group may easily be modified to other functions, e.g. carboxylic acids via hydrolysis (see Box 7.9) or amines by reduction.

Natural Cyanohydrins A Nd Cyanogenic Glycosides

Natural cyanohydrins feature as toxic constituents in several plants, e.g. laurel, bitter almonds, and cassava. In the plant, the cyanohydrin is bound through an acetal linkage to a sugar, usually glucose, to produce what is termed a cyanogenic glycoside.

• Cyanogenic means cyanide-producing, because, upon hydrolysis, the glycoside breaks down to the sugar, the carbonyl compound, and HCN. When a plant tissue containing a cyanogenic glycoside is crushed, hydrolytic enzymes also in the plant, but usually located in different cells, are brought into contact with the glycoside and begin to hydrolyze it.
• Alternatively, hydrolysis may be brought about by ingesting the plant material.
• Either way, it leads to the production of HCN, which is extremely toxic to humans. The glycoside itself is not especially toxic, and toxicity depends on the hydrolysis reaction.

The main cyanogenic glycoside in laurel is prunasin, the β-D-glucoside of benzaldehyde cyanohydrin. The enzymic hydrolysis of prunasin may be visualized as an acid-catalyzed process, first of all hydrolyzing the acetal linkage to produce glucose and cyanohydrin. Further hydrolysis results in the reversal of cyanohydrin formation, giving HCN and benzaldehyde.

• Bitter almonds contain amygdalin, which is the β-D-glucoside of prunasin, so it hydrolyses sequentially to the same products. Cassava, which is used in many parts of the world as a food plant, contains linamarin, which is the β-D-glucoside of acetone cyanohydrin.
• Preparation of the starchy tuberous roots of cassava for food involves prolonged hydrolysis and boiling to release and drive off the HCN before they are suitable for consumption.

Organometallic Cs: Gri Gnard Reagents And Acetylides

The use of organometallic reagents as nucleophiles towards carbonyl compounds is also synthetically important since it results in the formation of new C–C bonds, building up the size and complexity of the molecule. For carbon to act as a nucleophile, we require a negative charge on carbon, i.e. a carbanion or equivalent.

Although there are a variety of organometallic reagents available, we include here only two types of reagents, Grignard reagents and acetylides. We have met these organometallic reagents earlier (see Section 6.3.4) Reacting an alkyl or aryl halide, usually bromide, with metallic magnesium in ether solution, produces Grignard reagents (see Section 6.3.4). An exothermic reaction takes place in which the magnesium dissolves, and the product is a solution of the Grignard reagent RMgBr or ArMgBr.

The formation of this product need not concern us, but its nature is important, in that it contains the equivalent of R− or Ar−, i.e. the alkyl or aryl group has been transformed into its carbanion. The addition of an aldehyde or ketone to the solution of the Grignard reagent allows a nucleophilic addition reaction to occur.

The reaction resembles that of reduction with complex metal hydrides, in that the metal forms a complex with the oxygen from the carbonyl; and to complete the addition, this complex must be decomposed by the addition of a proton source through acidification of the mixture. Reactions are also going to be irreversible since the carbanions are very poor leaving groups.

• It should be noted that, on reaction with Grignard reagents, aldehydes will produce secondary alcohols, whereas ketones will form tertiary alcohols. Often forgotten is the possibility of synthesizing primary alcohols by using formaldehyde as the substrate.
• Acetylides are formed by treating terminal acetylenes with a strong base, sodium amide in liquid ammonia being the most commonly employed. Acetylenes with a hydrogen atom attached to the triple bond are weakly acidic (pKa about 25) due to the stability of the acetylide anion (see Section 4.3.4), and this anion can then act as a nucleophile.

It reacts with aldehydes and ketones in the manner expected, and after acidification yields an alcohol. This reaction also extends the carbon chain by two or more atoms, depending on the acetylide used, inserting a triple bond for further modification.

Synthesis O F T He Oral Contraceptive Ethinylestradiol

Several steroidal drugs are produced by procedures that include nucleophilic attack of sodium acetylide onto a ketone, particularly that at position 17 on the five-membered ring of the steroid.

Thus, the natural estrogen estrone can be converted into the drug ethinylestradiol by nucleophilic addition. Ethinylestradiol is some 12 times more effective than estradiol when taken orally, and is widely used in oral contraceptive preparations. In drug nomenclature, the systematic name ethynyl– for the HC≡C– group is usually presented as ethinyl–

• With a simple aldehyde or ketone substrate, there is an equal probability that the nucleophile will attack the carbonyl carbon from each face of the planar system, thus producing a racemic product, assuming that there are no other chiral carbons in the starting material.
• Estrone contains the complex steroidal ring system with four fused rings (see Box 3.19), and the product ethinylestradiol is formed as just one of the two possible epimers at C-17. By considering the three-dimensional shape of estrone.
• we can appreciate that nucleophilic attack from the upper face is hindered by the methyl group adjacent to the ketone. Therefore, the nucleophile can only approach from the lower face, and the product is formed stereospecifically.

Nitrogen As A Nucleophile: Imines And Enamines

Imines

The addition of primary amines to the carbonyl group of aldehydes and ketones is generally followed by the elimination of water (dehydration), and the product is called an imine or Schiff base. The elimination reaction is catalyzed by acid, but the initial nucleophilic attack depends upon the presence of a lone pair on the nitrogen.

• Accordingly, imine formation occurs only within a very limited pH range, typically pH 4–6. At lower pH values, the amine is extensively protonated and is therefore non-nucleophilic.
• Addition followed by elimination is not a feature encountered with the other nucleophiles considered in this chapter, but we shall see quite similar processes in aldol reactions where the nucleophile is an enolate anion.
• How are we going to explain this rather different behavior? It all depends upon leaving groups (see Section 7.1.2), and particularly the presence of two alternative leaving groups.

The intermediate amino alcohol in an acidic solution is going to be protonated, either on nitrogen or on oxygen. An equilibrium will be set up, and although we would expect nitrogen to be protonated in preference to oxygen, it is the next step that determines how the overall reaction proceeds, and how this equilibrium plays its part.

• We now have two possibilities for the loss of a leaving group. With nitrogen protonated, an amine is lost, and the protonated carbonyl reforms, so that we end up with the reverse reaction. With oxygen protonated, water is lost, and an iminium cation forms by an analogous type of electron movement from the nitrogen lone pair.
• Water (pKa of conjugate acid −1.7) is a better-leaving group than the amine (pKa of conjugate acid about 10.6), so the loss of water is favored, and the amino alcohol protonation equilibrium is disturbed accordingly. The product then is the imine.
• The elimination reaction is mechanistically an alternative version of the reverse reaction but involves the second, more favorable, leaving group. Further, and just to stress that this is not new or novel chemistry, let us go back and compare imine formation with acetal formation from hemiacetals.

The amino alcohol intermediate is analogous to the hemiacetal, and both undergo protonation and loss of water, facilitated by the heteroatom. The iminium cation can then lose a proton, but the oxonium cation has no proton to lose; instead, it is attacked by a nucleophile, namely a second molecule of alcohol. Imines are most conveniently visualized as nitrogen analogs of carbonyl groups since many of the reactions they undergo are paralleled in aldehyde and ketone chemistry.

As a simple example, we need only consider the reverse of imine formation – imines are readily hydrolyzed back to carbonyl compounds. In fact, because of this, many imines are somewhat unstable.

Protonation to the conjugate acid (iminium cation) increases the potential of the imine to act as an electrophile (compare carbonyl; see Section 7.1), and this is followed by nucleophilic attack of water.

• The protonated product is in equilibrium with the other mono-protonated species in which the nitrogen carries the charge. We shall meet this mechanistic feature from time to time, and it is usually represented in a mechanism simply by putting ‘−H+, +H+’ over the equilibrium arrows.
• Do not interpret this as an internal transfer of a proton; such transfer would not be possible, and it is necessary to have solvent to supply and remove protons. Protonation of nitrogen allows loss of the amine leaving group and formation of the conjugate acid of the carbonyl compound.
• Despite the comments made above regarding alternative leaving groups, imine formation and hydrolysis are reversible, though it will usually be necessary to disturb the equilibrium, as required, by using an excess of the appropriate reagent. In Section 10.6 we shall meet the Mannich reaction, where an imine or iminium ion acts as an electrophile for nucleophiles of the enolate anion type.

Hydrolysis Of Nitriles To Carboxylic Acids

Just as imines may be viewed as nitrogen analogs of carbonyl compounds, the C≡N group may also be viewed as carbonyl-like for interpretation of some of its reactions. For instance, nitriles are readily hydrolyzed in acid to give carboxylic acids (see Section 7.6.1). This process begins similarly to the hydrolysis of imines.

• In acid-catalyzed hydrolysis, we usually invoke protonation of the nitrogen to the conjugate acid to increase the potential of the nitrile to act as an electrophile, though the nitrile nitrogen is a very weak base.
• This is followed by a nucleophilic attack of water. Loss of a proton from the product cation generates a hydroxy-imine, which is a tautomer of a carboxylic acid amide (compare keto-enol tautomerism, Section 10.1).
• The keto-like tautomer (amide) is the more favored and is subsequently hydrolyzed under acidic conditions to a carboxylic acid (see Section 7.9.2). Hydrolysis under basic conditions is mechanistically similar, also proceeding through a hydroxy-imine. The tautomeric amide then undergoes basic hydrolysis.

The main theme to be appreciated here is that nucleophilic attack onto the nitrile triple bond can be interpreted mechanistically by extrapolation from carbonyl chemistry.

Nucleophilic Addition Of Carbon To Imines: The Strecker Synthesis Of Amino Acids

A nice example of the chemical similarity between imines and carbonyl compounds is the Strecker synthesis of amino acids. This involves the reaction of an aldehyde with ammonia and HCN (usually in the form of ammonium chloride plus KCN) to give an intermediate α-aminonitrile. Hydrolysis of the α-aminonitrile then produces the α-amino acid.

The sequence can be rationalized mechanistically as involving a nucleophilic attack of ammonia onto the aldehyde to produce an imine, which then acts as the electrophile for further nucleophilic attack, this time by the cyanide ion (see Section 7.7.1). The racemic amino acid is then formed by acid-catalyzed hydrolysis of the nitrile function, as above.

• This synthesis is fairly general and can be used for many amino acids, provided the R-side chain contains no other functional group that is sensitive to the reagents (see Section 13.1). R groups containing –NH₂, for example, would require appropriate protection measures. There is also considerable scope for making labeled amino acids via the use of 14C-labelled cyanide.
• Imine formation is an important reaction. It generates a C–N bond, and it is probably the most common way of forming heterocyclic rings containing nitrogen (see Section 11.10). Thus, cyclization of 5-aminopentanal to 1 -1-piperidine is merely intramolecular imine formation.
• A further property of imines that is shared with carbonyl groups is their susceptibility to reduction via complex metal hydrides (see Section 7.5). This allows imines to be reduced to amines, such as piperidine.

A combined reductive amination sequence has been developed as a useful way of synthesizing amines, with sodium cyanoborohydride as the reducing agent of choice. This complex metal hydride is a less reactive version of sodium borohydride (see Section 7.5) since the electron-withdrawing cyano group lowers the ability to transfer hydride. Consequently, sodium cyanoborohydride is rather selective, in that it will reduce iminium systems but does not reduce carbonyl compounds.

• The combined reaction thus involves the initial formation of the iminium ion from the carbonyl compound and amine at pH 6, and this intermediate is then reduced by the complex metal hydride to give the amine. This can also be a way of making methyl-substituted amines via intermediate imines with formaldehyde.
• We shall see later that reductive amination of a keto acid is the way nature synthesizes amino acids, using the biological analog of a complex metal hydride, namely NADPH.

Pyridoxal And Pyridoxamine: Vitamins That Participate Viaimine Formation

The terminology vitamin B6 covers several structurally related compounds, including pyridoxal and pyridoxamine and their 5-5-phosphates. Pyridoxal 5
-5-phosphate (PLP), in particular, acts as a coenzyme for a large number of important enzymic reactions, especially those involved in amino acid metabolism.

• We shall meet some of these in more detail later, e.g. transamination (see Section 15.6) and amino acid decarboxylation (see Section 15.7), but it is worth noting at this point that the biological role of PLP is dependent upon imine formation and hydrolysis. Vitamin B6 deficiency may lead to anemia, weakness, eye, mouth, and nose lesions, and neurological changes.
• Pyridoxal 5-5-phosphate is an aldehyde, and this grouping can react with the amino group of α-amino acids to form an imine; since an aldehyde is involved, biochemists often refer to this product as an aldimine. This imine undergoes changes in which the heterocyclic ring plays an important role (see Section 15.6),
• Changes that lead to the double bond of the imine ending up on the other side of the nitrogen atom. This is in many ways similar to rearrangement in an allylic system (see Section 8.2), and at its simplest can be represented as shown. Conjugation with the heterocyclic ring system facilitates the loss of a proton to start the electron redistribution.

Hydrolysis of the new imine then allows the formation of a ketone as part of an α-keto acid, and an amine which is the previously mentioned pyridoxamine 5-5-phosphate. Since this imine is the product of an amine and a ketone, it is termed a ketimine. These reactions are reversible, allowing amino acids to be converted into keto acids, and keto acids to be converted into amino acids.

Enamines

Secondary amines react with aldehydes and ketones via addition reactions, but instead of forming imines, produce compounds known as enamines. Initially, there is the same type of nucleophilic attack of the amine onto the carbonyl system, followed by acid-catalyzed dehydration; but, since the amine is secondary, the product of dehydration is an iminium ion rather than an imine.

This needs to lose a proton to become neutral, and since there is no available proton on nitrogen, one is lost from the nearest carbon atom, which is β to the nitrogen atom. This produces an enamine (ene-amine). We shall see later (see Section 10.5) that enamines are valuable synthetic intermediates, and are essentially nitrogen analogs of enols.

Nucleophilic Substitution On Carbonyl Groups: Carboxylic Acid Derivatives

We have seen that most of the addition reactions involving aldehydes and ketones are reversible (see Section 7.1.2). If the anionic intermediate does not react with an electrophile, then the carbonyl group is reformed and the original nucleophile is lost, i.e. it becomes a leaving group.

However, if there is another leaving group in this molecule, then this may be lost instead, so that overall the reaction becomes a nucleophilic substitution, though it should be regarded as an addition–elimination sequence since the carbonyl group is essential for this reactivity. This is readily appreciated if one compares the reactivity towards water of acetyl chloride and ethyl chloride.

Acetyl chloride must always be stored under anhydrous conditions because it readily reacts with moisture and becomes hydrolyzed to acetic acid. On the other hand, if one wanted to convert ethyl chloride into ethanol, this nucleophilic substitution reaction would require hydroxide, with its negative charge a better nucleophile than water, and an elevated temperature.

• It is clear, therefore, that the carbonyl group is responsible for the increased reactivity, and we must implicate this in our mechanisms. Although it is easier to draw a mechanism as an SN2-style substitution reaction, this is lazy and wrong, and the two-stage addition–elimination should always be shown.
• There is ample experimental evidence to show that a tetrahedral addition intermediate does participate in these reactions. The SN1 style mechanism is also incorrect, in that few reactions, and certainly none that we shall consider, actually follow this pathway

• A shorthand addition–elimination mechanism sometimes encountered is also shown. This employs a double-headed curly arrow to indicate the flow of electrons to and from the carbonyl oxygen; we prefer and shall use the longer two-step mechanism to emphasize the addition intermediate.
• The reaction may also be considered as acylation of the nucleophile since an acyl group RCO– is effectively added to the nucleophile; this description, however, conceals the fact that the electron-rich nucleophile is the attacking species in the reaction.
• Since this reaction, an overall substitution depends upon the presence of a suitable leaving group in the substrate, it is not surprising to find that the level of reactivity depends very much upon the nature of the leaving group.
• We have already seen that weak bases, the conjugate bases of strong acids, make good leaving groups (see Section 6.1.4). Conversely, strong bases, the conjugate bases of weak acids, are poor leaving groups.
• We can now see why aldehydes and ketones react with nucleophiles to give additional products. This is because the tetrahedral anionic intermediate has no satisfactory leaving group apart from the original nucleophile.
• The alternative possibilities, hydride in the case of aldehydes or an alkyl carbanion in the case of ketones, are both very poor leaving groups; both are the conjugate bases of very weak acids, namely molecular hydrogen (pKa 35) or an alkane (pKa 50) respectively. Therefore, in the forward reaction, the alkoxide intermediate instead reacts with an electrophile, usually by abstraction of a proton from solvent, and the overall reaction is addition.

Much better-leaving groups are encountered in carboxylic acid derivatives. Acyl halides possess a good leaving group in chloride, the conjugate base of HCl (pKa − 7), so react very readily with nucleophiles in overall substitution reactions.

• Where the leaving group is less satisfactory, the reactivity can be improved by carrying out the reaction under acidic conditions. Thus, the reaction of carboxylic acids with nucleophiles would require the loss of hydroxide as the leaving group, and this is the conjugate base of the weak acid water (pKa 15.7).
• This is not particularly favorable, but reactivity can be increased by protonation, leading to the expulsion of the neutral molecule water (pKa conjugate acid −1.7) as a good leaving group (see Section 6.1.4).
• Purists will dislike the intermediate shown above that has both negatively charged and positively charged oxygens as an unlikely species under acidic conditions. As we shall see, the intermediate formed under acidic conditions carries only a positive charge on the potential leaving group.

• Accordingly, the reactivity of compounds in this type of reaction can now be predicted by our appreciation of leaving-group tendencies (Table 7.2). Reactive substrates are those with a good leaving group, such as halide (in acyl halides), hydrosulfide (in thioacids), alkyl thiolate or alkyl mercaptide (in thioesters), and carboxylate (in anhydrides).
• Acids, esters, and amides are only moderately reactive, in that their leaving groups cannot be classified as good until they become protonated to the conjugate acid. Under acidic conditions, the leaving group then becomes a stable neutral molecule.
• As we have already seen, aldehydes and ketones have no satisfactory leaving group and undergo addition reactions rather than substitution reactions.

Leaving Groups And Reactivity In Carboxylic Acid Derivatives

Synthesis Of Anhydrides And A Cylhalides

As one of the most reactive groups of carboxylic acid derivatives, acyl halides are very useful substrates for the preparation of the other classes of derivatives. For example, anhydrides may be synthesized by the reaction of carboxylic acid salts with an acyl halide. In this reaction, the carboxylate anion acts as the nucleophile, eventually displacing the halide-leaving group.

Pyridine is often used as a solvent in such reactions since it also functions as a weak base. As an aromatic base, pyridine will promote ionization of the carboxylic acid to carboxylate, and also react with the other product of the reaction, namely HCl. Without the removal of HCl, the anhydride formed might be hydrolyzed under the acid conditions generated.

Pyridine has another useful attribute, in that it behaves as a nucleophilic catalyst, forming an intermediate acylpyridinium ion, which then reacts with the nucleophile. Pyridine is more nucleophilic than the carboxylate anion, and the acyl pyridinium ion has an excellent leaving group (pKa pyridinium 5.2). The reaction thus becomes a double nucleophilic substitution.

Acyl chlorides themselves may be synthesized by a similar type of reaction, in which we invoke nucleophilic attack of an acid onto thionyl chloride as shown. As we shall see later (see Section 7.13.1), the S=O group behaves as an electrophile in the same way as a C=O group

The leaving group is chloride, and after proton loss, we generate what may be considered a mixed anhydride having both C=O and S=O functionalities. The C=O group in this mixed anhydride is then attacked by chloride, and the good leaving group (–SO2Cl) this time dissociates into sulfur dioxide and chloride, as shown.

Oxygen And Sulfur As Nucleophiles: Esters An Carboxylic Acids

Alcohols: Ester Formation

A well-known reaction of carboxylic acids is that they react with alcohols under acidic conditions to yield esters.

• The equilibrium constant is not particularly favorable – may have to remove water or use an excess of one reagent, e.g. use the alcohol as a solvent ester.
• This reaction is, in fact, an equilibrium that often does not favor the product. Thus, to make it a useful procedure for the synthesis of esters, one has to disturb the equilibrium by either removing the water as it is formed or by using an excess of one reagent, typically alcohol.
• It is an easy way to make simple methyl or ethyl esters, where one can employ an excess of methanol or ethanol to act both as a solvent and to disturb the equilibrium.
• The reaction may be rationalized mechanistically as below, beginning with the protonation of the carbonyl oxygen using the acid catalyst. We are using an uncharged nucleophile, i.e. the lone pair of the alcohol oxygen atom acts as a nucleophile, so it is advantageous to increase the electrophilicity of the carbonyl.
• This is achieved by protonation, which introduces a positive charge. The product from the nucleophilic attack is a protonated tetrahedral addition species.

Acid-catalyzed Esterification

In an acidic medium, there will be an equilibrium set up such that any one of the three oxygen atoms may be protonated; they all have the same or similar basicities. The equilibrium will involve the loss of proton to the solvent, followed by deprotonation of another oxygen from the solvent. This equilibrium will then be disturbed as one of the protonated species is removed by further reaction.

We shall meet this mechanistic feature from time to time, and it is shown in more detail below. This type of process is usually represented in a mechanism simply by putting ‘−H+, +H+’ over the equilibrium arrows; we also met this under imines. Do not interpret this as an internal transfer of a proton; such transfer would not be possible, and it is necessary to have solvent to supply and remove protons.

• For ester formation to occur, one of the two hydroxyls needs to be protonated, so that it can be lost as a water-leaving group; protonation of the ethoxy would lead to loss of ethanol and reversal of the reaction. Both water and ethanol are good leaving groups, and the reaction is freely reversible.
• Loss of the leaving group is facilitated by a resonance effect from the other hydroxyl and leads to regeneration of the carbonyl group in a protonated form.
• The formation of the uncharged carbonyl regenerates the acid catalyst. Note that a base-catalyzed process for ester formation from acid and alcohol is not feasible, since the base would immediately ionize the carboxylic acid substrate and a nucleophile would not be able to attack the negatively charged carboxylate anion.
• The equilibrium limits the practical applicability of this reaction, and other methods would normally be employed if one were working with uncommon or expensive reagents that could not be used in excess. Esters are more conveniently prepared using the more reactive acyl halides or anhydrides, i.e. derivatives with better-leaving groups.

Note the use of a weak base to scavenge the HCl formed as a by-product in the acyl chloride reaction. The aromatic base pyridine is often used for this purpose, though it has other useful attributes. It functions as a good solvent for the reaction, but it also behaves as a nucleophilic catalyst, forming an intermediate acylpyridinium ion, which then reacts with the alcohol (see Box 7.12). Pyridine helps to catalyze reactions with anhydrides in a similar manner.

An interesting extension of the acid-catalyzed equilibrium reaction is the process termed transesterification. If an ester is treated with an excess of alcohol and an acid catalyst, then the ester OR group becomes replaced by the alcohol OR group. This reaction proceeds through a tetrahedral intermediate containing both types of OR group, and the product thus depends upon disturbing the equilibrium by using an excess of one or other of the two alcohols. A base-catalyzed process may also be used in transesterification.

Transesterification: A Spirin As An Acetylating A Gent

The mode of action of the analgesic aspirin is now known to involve a transesterification process. Aspirin (acetylsalicylic acid) exerts its action by acetylating the enzyme cyclooxygenase (COX) that is involved in the biosynthesis of prostaglandins (see Box 9.3). Prostaglandins are modified C20 fatty acids found in small quantities in animal tissues and they affect a wide variety of physiological processes, such as blood pressure, gastric secretion, smooth muscle contraction, and platelet aggregation.

Inflammation is a condition that occurs as a direct result of increased prostaglandin synthesis, and many of the non-steroidal anti-inflammatory drugs, such as aspirin and ibuprofen, exert their beneficial effects by reducing prostaglandin formation. Aspirin can do this by specifically acetylating the hydroxyl of a serine residue in COX, thus inactivating the enzyme and stopping the biosynthetic pathway to prostaglandins:

We shall meet another important example of transesterification in the action of the enzyme acetylcholinesterase (see Box 13.4).

• Cyclic esters (lactones) are formed when the carboxyl and hydroxyl groups are in the same molecule and are most favored when this results in the generation of strain-free five- or six-membered rings. Thus, 4-hydroxybutyric acid may form a five-membered lactone, which is termed a γ-lactone, its name coming from the alternative nomenclature γ-hydroxybutyric acid for the acyclic compound. Similarly, six-membered lactones are termed δ- lactones.
• It is generally easier to use the fully systematic -oxa- nomenclature (see Section 1.4) for the oxygen heterocycle in more complex lactones. This approach considers nomenclature as if we had a carbocyclic ring, and uses an -extra-syllable to indicate the replacement of a carbon with the oxygen heteroatom.
• Lactonization, like esterification, is an equilibrium process. γ-Lactones and δ-lactones are so readily formed that the carboxylic acid itself can provide the required acidic catalyst, and substantial amounts of the lactone are typically present in solutions of 4- or 5-hydroxy acids respectively (Table 7.3).
• Interestingly, the proportion of lactone is usually higher for five-membered rings than for six-membered rings.

Though other substituents, if they are present, also affect the equilibrium proportions. Larger lactones do not exist to any appreciable extent in equilibrium with the free hydroxy acids, but they may be prepared under appropriate conditions. These may include the removal of water to disturb the equilibrium, and carrying out the reaction at quite high dilutions to minimize intermolecular esterification.

Hydroxyacid–Lactone Equilibria

Large ring lactones: erythromycin and amphotericin

Very large ring lactones are called macrolides and are found in natural macrolide antibiotics. Typically, these may have 12-, 14-, or 16-membered lactone rings, though other sizes are encountered. Erythromycin is a good example. This antibiotic is prescribed for patients who are allergic to penicillins and is the antibiotic of choice for infections of Legionella pneumophila, the cause of Legionnaire’s disease.

Erythromycin is a mixture of at least three structurally similar compounds, the major component of which is erythromycin A. This has a 14-membered lactone ring, with a range of additional substituents. The ring system in erythromycin adopts a conformation that approximates the periphery of four fused chair-like rings. Note that erythromycin contains two uncommon sugar rings, cladinose, and desosamine, the latter being an aminosugar.

• Both of these sugars are bound to the lactone-containing ring through acetal linkages (see Section 7.2). Polyene macrolides have even larger lactone rings, typically from 26–38 atoms, which also accommodate a conjugated polyene of up to seven E double bonds.
• Amphotericin from cultures of Streptomyces nodosus provides a typical example and is used clinically as an antifungal agent. It is administered intravenously for treating potentially life-threatening fungal infections. Amphotericin is a mixture of compounds, the main and most active component being amphotericin B. The ring size in amphotericin B is 36 atoms, but is contracted from a potential 38 by cross-linking through a hemiketal function (see Section 7.2). An unusual amino sugar, D-glucosamine, is bound to the system through an acetal linkage.

Water: Hydrolysis S Of Carboxylic Acid Derivatives

All carboxylic acid derivatives are hydrolyzed to carboxylic acids by the action of water as a nucleophile. Acyl halides and anhydrides of low molecular weight are hydrolyzed quite vigorously.

• Esters and amides react much more slowly, and hydrolysis normally requires acid or base catalysis. This is nicely exemplified by the need to store and use compounds such as acetyl chloride and acetic anhydride under anhydrous conditions.
• On the other hand, the ester ethyl acetate is routinely used for solvent extractions of organic products from aqueous solutions. Hydrolysis of an ester can be achieved by either base- or acid-catalyzed reactions, and the nucleophilic substitution mechanisms follow processes that should now be becoming familiar to us. However, there are significant differences between the two types of process, as we shall see. In the acid-catalyzed hydrolysis of esters, the process is analogous to the acid-catalyzed formation of esters: it is merely the reverse reaction.

The reaction begins with the protonation of the carbonyl oxygen to give the conjugate acid, which increases the electrophilicity of the carbonyl. This is necessary because we are using a neutral nucleophile. A nucleophilic attack follows, giving a protonated tetrahedral intermediate.

• In an acidic medium, an equilibrium will again be set up such that any one of the three oxygen atoms may be protonated. This equilibrium also involves the loss of protons to the solvent, followed by deprotonation of another oxygen using the solvent. As in esterification (see Section 7.9.1), the process is not an internal transfer of a proton but requires solvent molecules.
• For hydrolysis to occur, the methoxyl needs to be protonated, so that it can be lost as a methanol leaving group. Loss of the leaving group is again facilitated by the resonance effect from a hydroxyl, leading to regeneration of the carbonyl group in protonated form.
• The formation of the uncharged carboxylic acid regenerates the acid catalyst. As with acid-catalyzed ester formation, the reaction is an equilibrium, and this equilibrium needs to be disturbed for complete hydrolysis, typically by using an excess of water, i.e. aqueous acid.

Autolysis Of Aspirin

The analgesic aspirin, acetylsalicylic acid, is an ester. In this compound, the alcohol part is a phenol, salicylic acid. Aspirin is synthesized from salicylic acid by treatment with acetic anhydride.

Aspirin is an ester, but it still contains a carboxylic acid function (pKa 3.5). In an aqueous solution, there will thus be significant ionization. However, this ionization now provides an acid catalyst for ester hydrolysis and initiates autolysis (autohydrolysis).

• The hydrolysis product salicylic acid (pKa 3.0) is also acidic; both aspirin and salicylic acid are aromatic acids and are rather stronger acids than aliphatic compounds such as acetic acid (pKa 4.8) (see Section 4.3.5). An aqueous solution of aspirin has a half-life of about 40 days at room temperature.
• In other words, after about 40 days, half of the material has been hydrolyzed, and the biological activity will have deteriorated similarly. Even aspirin tablets that have been stored under less-than-ideal conditions and, therefore, have absorbed some water from the atmosphere, are likely to have suffered partial hydrolysis. The characteristic odor of acetic acid (vinegar) from a bottle of aspirin tablets will be an indicator that some hydrolysis has occurred.
• In the base-catalyzed hydrolysis of esters, the nucleophile is hydroxide, a charged species that can attack the uncharged carbonyl. The carbonyl group is restored by the loss of alkoxide as the leaving group. However, alkoxide is a strong base, a poor leaving group, and the reaction seems unlikely to be favorable (see Section 7.8).
• It does occur, however, and this is because the strong base leaving group can abstract a proton from the carboxylic acid product, generating an alcohol and the carboxylate anion. Although the early steps of the reaction are reversible, this last step, ionization of the carboxylic acid, is essentially irreversible and so disturbs the equilibrium reaction. The ionization is not reversible: the carboxylate anion is far too weak a base to ionize an alcohol.

We can now distinguish differences between acid-catalyzed and base-catalyzed hydrolysis of esters. The acid-catalyzed reaction is an equilibrium, and the equilibrium needs to be disturbed by the use of an excess reagent (water). The acid used is a true catalyst: it is regenerated during the reaction.

• On the other hand, the base-catalyzed reaction goes to completion, because the basic leaving group ionizes the product and, in so doing, disturbs the equilibrium. This means that the base catalyst is not regenerated, but is consumed during the reaction.
• The description ‘base-catalyzed hydrolysis’ is generally used, but it is strictly incorrect since the base is a reagent rather than a catalyst; a better terminology is ‘base hydrolysis’. Basic hydrolysis of esters is usually the method of choice because the reaction goes to completion. Acidic hydrolysis would be selected where the molecules contain other functional groups that might be case-sensitive.

Ester Hydrolysis: Saponification Of Fats Andoils

Fats and oils are esters of the trihydric alcohol glycerol with long-chain fatty acids. The descriptor fat or oil is applied according to whether the material is a solid or liquid at room temperature; it has no chemical meaning. All three fatty acids in the ester may be the same, or they may be different. Common saturated fatty acids encountered are stearic acid (C18) and palmitic acid (C16), especially in animal fats, and the unsaturated acids oleic acid (C18) and linoleic acid (C18) in plant oils.

• Base hydrolysis of fats with sodium or potassium hydroxide liberates glycerol and the salt of the carboxylic acid(s). This reaction was the basis of soap making; the salt, or mixture of salts, is a soap with characteristic detergent properties. The relationship of ester hydrolysis to soap-making remains, in that base hydrolysis of esters is still commonly referred to as saponification.
• Amides may be hydrolyzed to carboxylic acids by either acids or bases, though hydrolysis is considerably slower than with esters. Although amines are bases and become protonated on nitrogen via the lone pair electrons, we know that amides are not basic (see Section 4.5.4).
• This is because the lone pair on the nitrogen in amides can overlap into the carbonyl π system, thus creating resonance stabilization in the neutral amide. This effect also diminishes the reactivity of the carbonyl towards nucleophilic attack, since the resonance contribution means less carbonyl character and more carbon-nitrogen double bond character.

Note that we can write a similar resonance picture for esters, and we shall need to invoke this when we discuss enolate anions (see Section 10.7). However, electron donation from oxygen is not as effective as from less electronegative nitrogen. We shall also see that this resonance effect in amides has other consequences, such as increased acidity of the amide hydrogens (see Section 10.7) and stereochemical aspects of peptides and proteins (see Section 13.3).

In addition, the amide derivatives have poorer leaving groups than the corresponding esters, and this also contributes to the lower reactivity of amides.

Although protonation does not occur on nitrogen in an amide, protonation can occur on the carbonyl oxygen, because this still allows the same type of resonance stabilization. Accordingly, acid hydrolysis of amides proceeds through a nucleophilic attack of water onto the protonated carbonyl, giving a tetrahedral protonated intermediate.

• The loss of a proton from this allows deprotonation on nitrogen; the nitrogen atom is no longer attached to a carbonyl, so it is basic, more basic than the oxygen atoms. The amine molecule is now a satisfactory leaving group, and this allows regeneration of the carbonyl. Of course, under acid conditions, the amine will be rapidly protonated and become non-nucleophilic, so this will help to disturb the equilibrium and discourage the reverse reaction.
• It also means that acid is used up in the hydrolysis, and we do not have true acid catalysis. Primary, secondary, and tertiary amides all undergo similar hydrolytic reactions, though hydrolysis does require heating with quite concentrated acid.
• Base hydrolysis of amides also requires quite vigorous conditions, but mechanistically it is exactly equivalent to base hydrolysis of esters. After the nucleophilic attack of hydroxide onto the carbonyl, the tetrahedral anionic intermediate can lose either an amide anion (care with nomenclature here, the amide anion is quite different from the amide molecule) or hydroxide. Although loss of hydroxide is preferred, since the amide anion is a stronger base than hydroxide, this would merely reverse the reaction.

The reaction progresses because the amide anion, once a small amount is released, abstracts a proton from the carboxylic acid product. Again, we have an analogy with the last step in the base hydrolysis of esters, and the ionization becomes an essentially irreversible step.

Furthermore, hydroxide is again consumed as a reagent. Base hydrolysis of secondary and tertiary amides is less readily achieved than with primary amides and may require stronger basic conditions.

Amide Hydrolysis: Peptides And Proteins

Proteins are fundamentally polymers of α-amino acids linked by amide linkages (see Section 13.1). It is a pity that biochemists refer to these amide linkages as peptide bonds; remember, a peptide is a small protein (less than about 40 amino acid residues), whereas a peptide bond is an amide. Therefore, peptides and proteins may be hydrolyzed to their constituent amino acids by either acid or base hydrolysis. The amide bond is quite resistant to hydrolytic conditions (see above), an important feature for natural proteins.

Neither acid nor base hydrolysis is ideal since some of the constituent amino acids are found to be sensitive to the reagents because of the nature of their R-side chains. Acid hydrolysis is the preferred method because it causes less degradation. Nevertheless, the indole system of tryptophan is known to be largely degraded in acid, and the sulfur-containing amino acid cysteine is also unstable. Those amino acids containing amide side-chains, e.g. asparagine and glutamine, will be hydrolyzed further, giving the corresponding structures with acidic side-chains, namely aspartic acid and glutamic acid.

Hiols: Thioacids And Thioesters

Thiols undergo the same types of nucleophilic reaction with carboxylic acid derivatives as alcohols. However, reactivity tends to be increased for two reasons. First, sulfur, because of its larger size, is a better nucleophile than oxygen (see Section 6.1.2); second, RS− is a better-leaving group than RO− (see Section 6.1.4), again because of its size and the less localized electrons. Simple nucleophilic reactions with H₂S parallel those with H₂O and those with RSH parallel those with ROH. This gives rise to carboxylic acid derivatives containing sulfur, such as thioacids and thioesters.

Thioesters : Coenzyme A

Thioesters are more reactive towards nucleophilic substitution than oxygen esters and are widely employed in natural biochemical processes because of this property. Coenzyme A is a structurally complex thiol, and functions in the transfer of acetyl groups via its thioester acetyl coenzyme A (acetyl-CoA; CH₃CO–SCoA).

We can understand the function of coenzyme A merely by appreciating that it is a thiol; the remaining part of the complex structure (it is a nucleotide derivative; see Section 14.3) aids its enzymic recognition and binding but does not significantly influence its reactivity. Thus, the thioester acetyl-CoA is a good acylating agent in biochemistry and can transfer the acetyl group to a suitable nucleophile.

This familiar reaction is effective under physiological conditions because it is enzyme-mediated and employs a good leaving group in the form of a thiolate anion CoAS−. It is rather interesting to note that nature uses this reaction to make acetate esters from ROH nucleophiles. Nature thus uses the more reactive combination of thioester plus alcohol, rather than the acid plus alcohol combination we might initially consider to make an ester. We shall see some other important biological reactions of thioesters in Box 10.8.

Nitrogen As A Nucleophile: Amides

Ammonia, primary amines, and secondary amines all react with carboxylic acids to give amides. However, all of these reagents are bases, and salt formation with the carboxylic acid occurs first, with basicity prevailing over nucleophilicity. The negatively charged carboxylate is correspondingly unreactive towards nucleophiles. A nucleophilic attack only occurs upon heating the ammonium salt, resulting in the overall dehydration of the salt. Consequently, it is usual to prepare amides by using a more favorable substrate than the carboxylic acid, one that is more reactive towards nucleophiles by possessing a better-leaving group, and where salt formation does not hinder the reaction.

Acyl halides and anhydrides are the most reactive class of carboxylic acid derivatives and readily react with amines to give amides. It should be noted that in both cases the leaving group is a conjugate base that, upon protonation during the reaction, will become an acid. Consequently, this acid forms a salt with the amine reagent, and the reaction will tend to stop. For success, the reaction thus requires the use of 2 mol of amine or some alternative base must be added.

Esters also react smoothly with amines, which is a useful reaction if the corresponding acyl halides or anhydrides are not easily available. The reaction proceeds through the anticipated tetrahedral anionic intermediate. There are two possible leaving groups in this tetrahedral intermediate: alkoxide anion or amide anion. Since ammonia is a considerably weaker acid than an alcohol, the preferred leaving group is the weaker base alkoxide.

• The consequence of this is that an ester can react with ammonia to give an amide, but the reverse reaction does not occur; simply treating amides with alcohols does not produce esters. Production of esters from amides requires acid or base catalysis. It should also be noted that amines are better nucleophiles than alcohols (see Section 6.1.2) and that the addition of nitrogen to the carbonyl compound does not usually require acid or base catalysis.
• Indeed, acid conditions would protonate the amine and destroy its nucleophilicity (see Section 7.7.1). Note that, in all of these reaction mechanisms, a proton needs to be removed from the nitrogen nucleophile. Hence ammonia, primary amines, and secondary amines, but not tertiary amines, can function as nucleophiles.
• Where we appeared to exploit the nucleophilicity of a tertiary amine (pyridine) towards carboxylic acid derivatives forming an acylpyridinium ion, the amine was subsequently lost as a leaving group (see Section 7.9.1). Pyridine behaved as a nucleophilic catalyst, and a permanent N–C bond was not produced.

Synthesis O F Paracetamol: An Example Of Selective Reactivities

The different reactivities associated with nucleophiles and leaving groups are nicely exemplified in the synthesis of the analgesic drug paracetamol (USA: acetaminophen) from 4-aminophenol. If 4-aminophenol is treated with an excess of acetic anhydride, acetylation of both amino and phenol groups is observed, and the product is the diacetate. Paracetamol is the N-acetate of 4-aminophenol, so how might mono-acetylation be achieved? There are two approaches.

• One method is to treat 4-aminophenol with just one molar equivalent of acetic anhydride. The main product is paracetamol, which is produced almost selectively since –NH₂ is a better nucleophile than –OH. We can predict this from their pKa values as bases, about 5 for the conjugate acid of a typical aromatic amine, and about −7 for a phenol, i.e. the amine is the stronger base.
• Although the heteroatoms are not the same (see Section 6.1.2), the pKa values are significantly different and allow us to predict that the amine is also going to be the better nucleophile. The higher the pKa of the conjugate acid, the better the nucleophile.
• However, the second method of synthesizing paracetamol is to hydrolyze the diacetate of 4-aminophenol carefully using aqueous NaOH. In this case, hydrolysis of the amide is slower than that of the ester, since the ArNH− ion is a poorer leaving group than ArO−. Again, this can be predicted from pKa values. ArOH (pKa 10) is a stronger acid than ArNH₂ (pKa 28). The lower the pKa of the conjugate acid, the better the leaving group.
• We observed that cyclic esters (lactones) may be formed when the carboxyl electrophile and hydroxyl nucleophile are in the same molecule (see Section 7.9.1). Similarly, cyclic amides are produced when carboxyl and amine groups are in the same molecule and are again most favored when this results in the generation of strain-free five- or six-membered rings. Cyclic esters are termed lactones, whereas cyclic amides are in turn called lactams.
• The nomenclature of lactams is similar to that used for lactones. Thus, 4-aminobutyric acid may form a five-membered lactam, which is termed a γ-lactam, its name coming from the alternative nomenclature γ-aminobutyric acid for the acyclic compound. Similarly, six-membered lactams are termed δ-lactams. The fully systematic -aza nomenclature for the nitrogen heterocycle in lactams is much easier to use. In practice, we would probably name lactams as ketone derivatives of heterocycles.

Amides and β -lactams:  Semi-synthesis and hydrolysis of penicillins and cephalosporins

Penicillin and cephalosporin antibiotics possess an unusual and highly strained four-membered lactam ring. These and related antibiotics are commonly called β-lactam antibiotics. In both penicillins and cephalosporins, the β-lactam ring is fused through the nitrogen and adjacent carbon to a sulfur-containing ring. This is a five-membered thiazolidine ring in penicillins and a six-membered dihydrothiazine ring in cephalosporins.

The penicillins are the oldest of the clinical antibiotics and are still the most widely used. Early examples, such as benzylpenicillin, were decomposed by gastric acid and, consequently, could not be administered orally. Modern penicillins have been developed to overcome this sensitivity towards acid by changing the nature of the carboxylic acid that features in the acyclic amide linkage. It was found that introducing electron-withdrawing heteroatoms into the side chain significantly inhibited sensitivity to acid hydrolysis.

Benzylpenicillin, produced in fermentors by cultures of the fungus Penicillium chrysogenum, is converted into 6-aminopenicillanic acid by a suitable bacterial enzyme system. This enzyme selectively hydrolyses the acyclic amide, without affecting the cyclic amide.

This selectivity is not achievable by simple chemical hydrolysis, since the strained β-lactam ring is much more susceptible to nucleophilic attack than the unstrained side-chain amide function. Normally, the electron-donating effect from the lone pair of the adjacent nitrogen stabilizes the carbonyl against nucleophilic attack (see Section 7.9.2); this is not possible with the β-lactam ring because of the geometric restrictions.

It is feasible to convert benzylpenicillin into 6-aminopenicillins acid chemically but by a procedure involving several steps. After removal of the carboxylic acid portion of the original amide, a new amide linkage is generated, e.g. by reaction with a suitable acyl chloride. One of the first commercial semi-synthetic penicillins, methicillin, was produced as shown. Other agents, e.g. ampicillin, may be produced by similar means, though sensitive functional groups in the new side-chain will need suitable protection.

An additional disadvantage with many penicillin and cephalosporin antibiotics is that bacteria have developed resistance to the drugs by producing enzymes capable of hydrolyzing the β-lactam ring; these enzymes are called β-lactamases. This type of resistance still poses serious problems. Indeed, methicillin is no longer used, and antibiotic-resistant strains of the most common infective bacterium Staphylococcus aureus are commonly referred to as MRSA (methicillin-resistant Staphylococcus aureus). The action of β-lactamase enzymes resembles simple base hydrolysis of an amide.

• It is known that the nucleophilic species in a β-lactamase enzyme is the hydroxyl group of a serine residue in the protein and that this attacks the β-lactam carbonyl, followed by loss of the leaving group and consequent opening of the four-membered ring. The ring-opened penicillin (or cephalosporin) becomes bound to the enzyme through an ester linkage and is no longer active.
• The ester linkage is subsequently hydrolyzed to release an inactive penicilloic acid derivative and regenerate the functional β-lactamase enzyme. Note again that the strained β-lactam ring is more susceptible to nucleophilic attack than the unstrained sidechain amide function.
• However, by increasing the steric bulk of the side-chain, the approach of a β-lactamase enzyme to the β-lactam ring is hindered in the semi-synthetic antibiotic, giving it more resistance to enzymic hydrolysis.

Hydride As A Nucleophile: Reduction Of Carboxylic Acid Derivatives

We have already noted the ability of complex metal hydrides like lithium aluminum hydride and sodium borohydride to reduce the carbonyl group of aldehydes and ketones, giving alcohols (see Section 7.5). These reagents deliver hydride in such a manner that it appears to act as a nucleophile. However, as we have seen, the aluminum hydride anion is responsible for the transfer of the hydride and the hydride itself never acts as a nucleophile because of its small size and high charge density.

Acyl halides, anhydrides, esters, and acids all react with LAH to give a primary alcohol. Amides (see later) behave differently. The initial reaction is effectively the same as with an aldehyde or ketone, in that hydride is transferred from the reducing agent, and that the tetrahedral anionic intermediate then complexes with the Lewis acid aluminium hydride. However, the typical reactivity of the carboxylic acid derivatives arises because of the presence of a leaving group.

• The first-formed product is an aldehyde, resulting from loss of the leaving group and regeneration of the carbonyl. It is not normally possible to isolate this aldehyde product, because it reacts rapidly with the reducing agent, more rapidly in fact than the original carboxylic acid derivative.
• As a result, the aldehyde is further reduced, and after treatment with a proton source is converted into a primary alcohol. The Lewis acid aluminum hydride released during regeneration of the carbonyl will complex with the leaving group and continue as a source of hydride.
• Although LAH will reduce carboxylic acids, it is not usually employed for this purpose, since salt formation can interfere with the reduction process. LAH is a strong base, and the lithium salt of the carboxylic acid typically precipitates out of the solution.
• The usual approach to reducing carboxylic acids is to employ a two-stage process, first making an ester and then reducing this derivative. A feature of ester reduction is that it generates two molecules of alcohol, one from the acyl group and one from the leaving group.

Selective Reduction Of Carbonyl Groups

Sodium borohydride is a weaker hydride donor than lithium aluminum hydride (see Section 7.5) and it is only effective for reducing acyl halides, the most reactive of the carboxylic acid derivatives. However, this difference in reactivity of the reducing agents can be very useful, allowing selectivity and reduction of one group in the presence of other susceptible groups. For example, NaBH4 will reduce the more reactive aldehyde and ketone groups but not reduce the less reactive ester group.

Therefore, it is possible to reduce both carbonyl groups in the ketoester methyl 4-oxo cyclohexane carboxylate using LAH. Sodium borohydride will reduce only the ketone, giving the hydroxyester. However, by using a ketal as a protecting group (see Section 7.2) it is possible to reduce just the ester with LAH, and the original ketone can be regenerated by hydrolysis of the ketal.

Amides behave differently towards LAH than the other carboxylic acid derivatives, and the overall reaction observed is reduction of the carbonyl to a methylene group, with retention of the amino group.

This unusual behavior may be explained simply as a consequence of alternative leaving groups being present in the addition intermediate. After the transfer of hydride to the carbonyl with the formation of a tetrahedral anionic complex, there are two potential leaving groups, i.e. R₂N− and the aluminate anion (OAlH₃)₂−.

The aluminate anion is a better-leaving group than the amide, and this leads to the formation of an iminium ion. This behavior can thus be seen to be analogous to the dehydration of hydroxylamines to imines during the reaction of aldehydes or ketones with amines (see Section 7.7.1). There, we also had an alternative leaving group present.

In the LAH reduction sequence, the C=N double bond in the iminium ion now behaves just as the C=O bond of a carbonyl (see Section 7.7.1) and is also reduced by the transfer of hydride from a further equivalent of LAH. The final product is thus an amine. These reactions also provide us with a convenient way of making secondary and tertiary amines. Thus, a primary amine may be converted into an amide by reaction with an acyl chloride, then LAH reduction leads to a secondary amine. We are effectively introducing an RCH₂– group via the corresponding R_CO– acyl group.

The Importance Of Leaving Groups: Linking Up The Chemistry Of Amides With Imine F Ormation

Although we have discussed all of the following reactions, they have been covered in separate sections, and this box is included to draw them together and demonstrate that they can all be rationalized via a common theme, namely the nature of leaving groups. A fundamental consideration is that addition to the carbonyl group is reversible if the nucleophile can subsequently be lost from the addition product as a leaving group. This explains why halides do not react as nucleophiles towards carbonyl compounds; halides are such good leaving groups that the reverse reaction always predominates. The forward reaction is completed by protonation of the oxyanion in the case of aldehydes and ketones, or loss of a leaving group for carboxylic acid derivatives.

We saw that the reaction of amines with aldehydes or ketones led to imine formation, rather than the simple amino alcohol addition product (see Section 7.7.1). This was because, in an acidic solution, the protonated amino alcohol had two possible leaving groups, and water rather than amine was the better-leaving group. Dehydration occurs, leading to the imine.

Amide formation involved the same considerations. Thus, esters are readily converted into amides by treatment with ammonia (see Section 7.10). The intermediate anion has two potential leaving groups, alkoxide RO− and amide NH2−, and alkoxide is the better-leaving group. The converse of this is that treatment of an amide with alcohol does not lead to an amide; we generate the same intermediate anion, so the reverse reaction, loss of alkoxide, predominates.

Reducing aldehydes and ketones with a complex metal hydride gives an alcohol (see Section 7.5). Such reactions are not reversible because hydride is a very poor leaving group, so we eventually get protonation of the alkoxide system. Acyl derivatives generally have a good leaving group and this is lost, restoring the carbonyl group, and producing an aldehyde. Of course, this reacts further with a reducing agent, and the final product is a primary alcohol.

Amides seem to behave differently, with complex metal hydride reduction giving an amine, effectively converting the carbonyl group to methylene.

This behavior results from the initial formation of an intermediate with two potential leaving groups, an amide anion R₂N− and the aluminate anion (OAlH₃)₂−. Aluminate is the better-leaving group, and its loss produces an iminium cation that is also subject to further reduction. This gives us the amine product.

Although at first glance the behavior of some of these carbonyl compounds toward nucleophiles might seem anomalous, closer consideration shows there is a logical explanation for the reactions observed. Furthermore, if we understand the underlying mechanisms, these reactions become predictable.

Carbon As A Nucleophile : Grignardre Agents

The reaction of carbon nucleophiles derived from organometallics with carboxylic acid derivatives follows closely the reactions we have already encountered in Sections 6.3.2 and 7.6.2. Organometallics such as Grignard reagents are conveniently regarded as sources of carbanion equivalents, and these add to the carbonyl, followed by loss of the leaving group. As with other examples, a tetrahedral anionic complex with the metal is likely to be produced. Regeneration of the carbonyl with loss of the leaving group produces an intermediate ketone.

Now we see an analogy with the LAH reduction sequence (see Section 7.11), in that this ketone intermediate also reacts with the organometallic reagent, rather more readily than the initial carboxylic acid derivative, so that this ketone cannot usually be isolated.

The final product is thus a tertiary alcohol, which contains two alkyl or aryl groups from the organometallic reagent. Note that derivatives of formic acid, HCO₂H, will be converted into secondary alcohols by this double reaction with a Grignard reagent.

Nucleophilic Substitution On Derivatives Of Sulfuric And Phosphoric Acids

Phosphorus and sulfur are immediately below nitrogen and oxygen in the periodic table; therefore, we might expect them to have properties akin to nitrogen and oxygen. This is true in principle, so that PH₃ and H₂S are going to be analogs of NH₃ and H₂O. We have already met several sulfur derivatives, and have seen how thiols can be considered to behave in much the same way as alcohols (see Sections 6.3.2 and 7.4).

• However, a major difference that is encountered between phosphorus and sulfur arises from the fact that both can accommodate more than eight electrons in the outer electron shell. Thecanto makes use of d orbitals in bonding, and this leads to a greater versatility in bonding and a range of valencies is available to them.
• In the common acids phosphoric acid H₃PO₄ and sulfuric acid H₂SO₄, phosphorus is pentavalent and sulfur is hexavalent. Interestingly, we now find that these atoms have more in common with carbon than with nitrogen and oxygen.
• The chemical reactivity of organic derivatives of phosphoric and sulfuric acids in most aspects parallels that of carboxylic acid derivatives, so this is a particularly convenient place to describe some of the reactions and to emphasize the similarities.
• Most reactions of sulfuric and phosphoric acid derivatives can be rationalized by considering that the S=O and P=O functionalities are equivalent to the carbonyl group, and that polarization in these groups allows similar nucleophilic reactions to occur. Initial nucleophilic addition will then be followed by loss of an appropriate leaving group and regeneration of the S=O or P=O.

Sulfuric Acid Derivatives

Sulfuric acid can form ester derivatives with alcohols, though since it is a dibasic acid (pKa − 3, 2) it can form both mono- and di-esters. Thus, the acid-catalyzed reaction of methanol with sulfuric acid initially gives methyl hydrogen sulfate, and with a second mole of alcohol the diester dimethyl sulfate. Though not shown here, the mechanism will be analogous to the acid-catalyzed formation of carboxylic acid esters (see Section 7.9).

The simple diesters dimethyl sulfate and diethyl sulfate are convenient and useful reagents for alkylation reactions. As derivatives of sulfuric acid, the alkyl sulfate anions are also the conjugate bases of strong acids and are consequently good leaving groups.

• Compounds that are even better analogs of carboxylic acids are produced when an alkyl or aryl group replaces one of the hydroxyls in sulfuric acid. This provides compounds called sulfonic acids, which in turn give rise to a range of derivatives exactly comparable to those we have met as carboxylic acid derivatives (Table 7.4).
• As with the carboxylic acid group, the reactivity of these sulfonic acid derivatives may be predicted from the properties of the leaving group, and sulfonyl chlorides are the most reactive (see Section 7.8).
• Other classes of derivatives are thus most conveniently prepared from the sulfonyl chloride. Reaction with alcohol leads to the formation of a sulfonate ester. Two common sulfonyl chloride reagents employed to make sulfonate esters from alcohols are p-toluenesulfonyl chloride, known as tosyl chloride, and methanesulfonyl chloride, known as mesyl chloride (see Section 6.1.4). Note the nomenclature tosyl and mesyl for these groups, which may be abbreviated to Ts and Ms respectively

The reaction of tosyl chloride with an alcohol is easily represented by the standard nucleophilic substitution sequence, and gives a sulfonate ester called a tosyl ester.

Tosyl esters are good alkylating agents, rather like dimethyl sulfate above, and for the same reason, i.e. the presence of a good resonance-stabilized leaving group, the tosylate anion. This is the conjugate base of p-toluenesulfonic acid, a strong acid (pKa − 1.3). Thus, tosyl chloride may be used to facilitate nucleophilic substitutions.

Hydroxide is a poor leaving group, and nucleophilic reactions on alcohols are not particularly favorable unless acidic conditions are used to protonate the hydroxyl and produce a better-leaving group, namely water (see Section 6.1.4). An alternative is to convert the alcohol into its tosylate ester using tosyl chloride and then carry out the nucleophilic substitution on this ester, where there is now an excellent leaving group in the tosylate anion. This strategy may also be used to facilitate elimination reactions by providing a better-leaving group.

• Mesyl chloride may be employed in the same manner as tosyl chloride. Methanesulfonic acid is also a strong acid (pKa − 1.2). Using amines as nucleophiles, sulfonyl chlorides are readily converted into sulfonamides, exemplified here by the formation of op-amino benzenesulfonamide (sulfanilamide).

• It is relatively easy to predict many properties of sulfonamides just by thinking about the corresponding amides. For example, just as amide nitrogens are not basic, so the sulfonamide nitrogen is not basic; indeed, it is more likely to be acidic (see Section 4.5.4).
• This is because of resonance stabilization involving the nitrogen lone pair feeding back towards the oxygen. Resonance is also responsible for stabilizing the anion resulting frothe m loss of a proton from this nitrogen. Thus, pKa values for sulfanilamide are 2.0 and 10.5.
• The 2.0 value relates to the aromatic amino group, which is somewhat less basic than aniline (pKa 4.6; see Section 4.5.4) due to the contribution from the electron-withdrawing para sulfonyl group. The sulfonamide amine is rather more acidic than a carboxylic amide (pKa about 18; see Section 10.7), a feature of the enhanced resonance stabilization conferred by two S=O systems.

Sulfonamide A Ntibiotics And Diuretics

Sulfanilamide was the first of a range of synthetic antibacterial drugs known collectively as sulfa drugs. These agents are antibacterial because they mimic in size, shape, and polariton the carboxylic acid p-aminobenzoic acid.

p-Aminobenzoic acid is used by bacteria for the synthesis of folic acid, and sulfanilamide acts as a competitive inhibitor of an enzyme involved in folic acid biosynthesis.

• Folic acid is vital for both humans and bacteria. Bacteria synthesize this compound, but humans are unable to synthesize it and, consequently, obtain the necessary amounts from the diet, principally from green vegetables and yeast. This allows selectivity of action. Therefore, sulfa drugs are toxic to bacteria because folic acid biosynthesis is inhibited, whereas they produce little or no ill effects in humans. The structural relationships between carboxylic acids and sulfonic acids that we have observed in rationalizing chemical reactivity are now seen to extend to some biological properties.
• The use of sulfa drugs as antibacterial drugs has diminished over the years as even better agents have been discovered, but in their time they were crucial to medical health. An interesting additional property of many sulfa drugs has been developed further, however. Many sulfonamides display diuretic activity, and sulfonamide diuretics are still a major drug group. Two widely used examples are furosemide and bendroflumethazide.
• Furosemide (frusemide) is an aromatic sulfonamide that can be seen to be a structural variant of sulfanilamide. Bendroflumethazide (bendrofluazide) also contains an aromatic sulfonamide grouping, but in addition contains a second sulfonamide group as part of a ring system (compare lactams, section 7.10). This drug is a member of the thiazide diuretics, so named because of the ring system containing this cyclic sulfonamide.

Phosphoric Acid Derivatives

Derivatives of phosphoric acid are of particular significance in biochemical reactions, in that many metabolic intermediates are phosphates. The phosphate group introduces polarity, makes the compound water soluble, and provides a group that facilitates binding to proteins, especially enzymes.

Phosphoric acid (pKa 2.1) is a weaker acid than sulfuric acid (pKa − 3), but stronger than a typical carboxylic acid (pKa 5). It is also a tribasic acid; ionizations and pKa values are as shown. It follows that, at pH 7, there will be considerable ionization, and by application of the Henderson–Hasselbalch equation (see Section 4.9) the major species at pH 7 can also be determined.

As a tribasic ac, id it has three replaceable hydroxy, st mono-, di-, and tri-substituted derivatives are possible.

The Nucleic Acids DNA and RNA Feature Diesters Of Phosphoric Acid

Whilst many biochemicals are mono-esters of phosphoric acid, the nucleic acids DNA and RNA (see Section 14.2) provide us with good examples of diesters. A short portion of one strand of a DNA molecule is shown here; the most significant difference in RNA is the use of ribose rather than deoxyribose as the sugar unit.

A feature of phosphoric acid is that it forms a series of polymeric anhydrides that resemble carboxylic acid anhydrides in structure and reactivity. Diphosphoric acid (formerly called pyrophosphoric acid) and triphosphoric acid are the simplest examples, and derivatives of these are the ones we meet in biochemistry.

Since phosphoric acid, diphosphoric acid, and triphosphoric acid are reasonably strong acids, their anions are good leaving groups and biochemical reactions frequently exploit this leaving group capacity. Phosphate derivatives retaining one or more unsubstituted hydroxyls will usually be significantly ionized at physiological pH so these compounds will be water-soluble, which is an important property for substrates in metabolic processes.

• When we draw the structures of phosphate derivatives in metabolic transformations, we should strictly show these compounds as anions, but, in general, the additional negative charges complicate the structures and interfere with our understanding of mechanistic electron movements. As a result, non-ionized acids may be shown to simplify structures and mechanisms and avoid the need for counter-ions; this is the convention we shall use.
• It is also very common to see abbreviations for phosphate-based structures, such as OP for phosphate, and OPP for diphosphate, which are convenient to use when mechanisms do not involve the P=O system. When writing such phosphates, drawing a ring around the P is a speedy and accepted way of abbreviating the structure.
• Nucleophilic reactions on phosphate derivatives follow the general mechanisms seen with carboxylic acid derivatives, namely initial attack on the P=O double bond followed by loss of the leaving group. In the following example, we employ a diphosphate system as the electrophile. Note that there are two types of linkage in this diphosphate, i.e. an anhydride and an ester.
• Nucleophilic attack results in cleavage of the anhydride bond (phosphate is a good leaving group) and not the ester bond (RO− is a poor leaving group). A nucleophilic attack followed by cleavage of the anhydride bond could also result if the alternative P=O was the electrophile. This is an equally valid mechanism, but it is not as common in enzyme-controlled reactions as an attack on the terminal phosphate.

Adenosine T Riphosphate

One of the most important molecules in biochemical metabolism is adenosine triphosphate (ATP). Hydrolysis of ATP to adenosine diphosphate (ADP) liberates energy, which can be coupled to energy-requiring processes. Alternatively, the synthesis of ATP from ADP can be coupled to energy-releasing processes. ATP thus provides nature with a molecule for energy storage; we also consider it to be the currency unit for energy

Hydrolysis of ATP to ADP is rationalized as a nucleophilic attack of water onto the terminal P=O double bond, followed by cleavage of the anhydride bond and expulsion of ADP as the leaving group.

• Note that there are two anhydride linkages in ATP and one ester linkage. We know that hydrolysis of anhydride bonds is more favorable than hydrolysis of ester bonds because of the nature of the leaving group (see Section 7.8). In the enzyme-controlled reaction.
• The nucleophilic attack usually occurs on the terminal P=O (hydrolysis of ATP to ADP), but very occasionally we encounter an attack on the central P=O (hydrolysis of ATP to adenosine monophosphate, AMP).
• Both reactions yield the same amount of energy,
  G = −34 kJ mol−1. This is not surprising, since the same type of bond is being hydrolyzed in each case. The further hydrolysis of AMP to adenosine breaks an ester linkage anliberateste only a fraction of the energy,
  G = −9 kJ mol1, so this reaction is not biochemically important.

Inhibitors of acetylcholinesterase

The neurotransmitter acetylcholine is both a quaternary ammonium compound (see Box 6.7) and an ester. After interaction with its receptor, acetylcholine is normally degraded by hydrolysis in a reaction catalyzed by the enzyme acetylcholinesterase. This enzyme contains a serine residue that acts as the nucleophile, hydrolyzing the ester linkage in acetylcholine (see Box 13.4).

This effectively acetylates the serine hydroxyl and is an example of transesterification (see Section 7.9.1). Fothe r continuation of acetylcholine degradation, the original form of the enzyme must be regenerated by a further ester hydrolysis reaction.

Acetylcholinesterase is a remarkably efficient enzyme; turnover has been estimated as over 10,000 molecules per second at a single active site. This also makes it a key target for drug action, and acetylcholinesterase inhibitors are of considerable importance.

Some natural and synthetic toxins also function by inhibiting this enzyme. The natural alkaloid physostigmine (eserine) and its synthetic analog neostigmine inhibit acetylcholinesterase by forming a covalent intermediate that hydrolyzes very much more slowly than the normal substrate. These drugs are carbamoyl esters rather than acetyl esters.

The carbamoyl group is transferred to the serine hydroxyl in the enzyme, but the resultant carbamoyl–enzyme intermediate then hydrolyses only very slowly (minutes rather than microseconds), effectively blocking the active site for most of the time. The slower rate of hydrolysis of the serine carbamate ester is a consequence of decreased carbonyl character resulting from resonance stabilization, as shown.

By markedly slowing down the degradation of acetylcholine, these drugs are used to prolong the effects of endogenous acetylcholine. Physostigmine and neostigmine have ophthalmic use as a miotic, contracting the pupil of the eye, often to combat the effects of mydriatics such as atropine.

• They could be used as an antidote to anticholinergic poisons, such as atropine (see Box 10.9), or to reverse the effects of muscle relaxants that block acetylcholine receptors, such as tubocurarine and atracurium (see Boxes 6.7 and 6.9). Other acetylcholinesterase inhibitors, e.g. rivastigmine, are found to be of value in treating Alzheimer’s disease, by increasing memory function.
• Many phosphorus derivatives function as irreversible inhibitors of acetylcholinesterase, and are thus potentially toxic. These include a range of organophosphorus insecticides, such as malathion and parathion, and nerve gases such as sarin.
• In contrast to the inhibitors such as neostigmine and related compounds described above, where the intermediate complexes hydrolyze slowly, these toxic compounds form complexes that do not hydrolyze. The enzyme becomes irreversibly bound to the toxin and, as a result, ceases to function. These agents all have left groups that can be displaced by the serine hydroxyl of the enzyme, leading to stable addition products.

Malathion and parathion contain a P=S grouping, exemplifying a further carbonyl analogue, in which phosphorus replaces carbon, and sulfur replaces oxygen. Nevertheless, the same type of chemistry occurs, in which the serine hydroxyl of the insect’s acetylcholinesterase attacks this P=S electrophile, followed by the expulsion of the leaving group, here a thiolate.

• The esterified enzyme, however, is not hydrolyzed back to the original form of the enzyme, and its action is thus totally inhibited. The insect becomes subjected to a build-up of acetylcholine that eventually proves fatal. Malathion is much less toxic to mammals because of the two carboxylic ester functions that allow metabolism through hydrolysis to inactive products.
• Even more reactive towards acetylcholinesterase are the organophosphorus derivatives developed as chemical warfare nerve agents, e.g. sarin. Such compounds react readily with the enzyme and form very stable addition intermediates. It is unusual to see fluoride as a leaving group, as in sarin, but its presence provides a huge inductive effect, thus accelerating the initial nucleophilic addition step

Acylphosphates: Mixed Anhydrides Of Phosphoric And Carboxylic Acids

We are now familiar with anhydrides of carboxylic acids, e.g. acetic anhydride, and of phosphoric acid, e.g. ATP. In each case we can rationalize their reactivity by considering nucleophilic attack onto the C=O or P=O, followed by loss of a leaving group, carboxylate, or a phosphate derivative.

As we consider biochemical processes in Chapter 15, we shall encounter some metabolic intermediates that are acylphosphates, i.e. hybrid mixed anhydrides of phosphoric and carboxylic acids. The simplest example of an acylphosphate is acetylphosphate, employed by some bacteria as an energy-rich metabolite.

Examples of considerably more consequence are 1,3-diphosphoglyceric acid in the glycolytic pathway, and succinyl phosphate in the Krebs cycle. These compounds should not trouble us, since their reactivity is easily explained in terms of the above processes.

• In both cases, the mixed anhydride is used to synthesize ATP from ADP. Hydrolysis of the anhydride liberates more energy than the hydrolysis of ATP to ADP and, therefore, can be linked to the enzymic synthesis of ATP from ADP.
• This may be shown mechanistically as a hydroxyl group on ADP acting as a nucleophile towards the mixed anhydride, and in each case, a new phosphoric anhydride is formed. In the case of succinyl phosphate, it turns out that GDP rather than ADP attacks the acyl phosphate, and ATP production is a later step (see Section 15.3).
• These are enzymic reactions; therefore, the reaction and the nature of the product are closely controlled. We need not concern ourselves why an attack should be on the P=O rather than on the C=O. Further examples of acyl phosphates are found in fatty acyl-AMPs (see Section 15.4.1) and aminoacyl-AMPs.
• Activated intermediates in the metabolism of fatty acids and formation of peptides respectively. Each of these is attacked on the C=O by an appropriate S or O nucleophile, displacing the phosphate derivative AMP.