The functional group of alkyl halides is a carbon-halogen bond, the common halogens being fluorine, chlorine, bromine and iodine. With the exception of iodine, these halogens have electronegativities significantly greater than carbon. Consequently, this functional group is polarized so that the carbon is electrophilic and the halogen is nucleophilic, as shown in the drawing on the right. Two characteristics other than electronegativity also have an important influence on the chemical behavior of these compounds. The first of these is covalent bond strength. The strongest of the carbon-halogen covalent bonds is that to fluorine. Remarkably, this is the strongest common single bond to carbon, being roughly 30 kcal/mole stronger than a carbon-carbon bond and about 15 kcal/mole stronger than a carbon-hydrogen bond. Because of this, alkyl fluorides and fluorocarbons in general are chemically and thermodynamically quite stable, and do not share any of the reactivity patterns shown by the other alkyl halides. The carbon-chlorine covalent bond is slightly weaker than a carbon-carbon bond, and the bonds to the other halogens are weaker still, the bond to iodine being about 33% weaker. The second factor to be considered is the relative stability of the corresponding halide anions, which is likely the form in which these electronegative atoms will be replaced. This stability may be estimated from the relative acidities of the H-X acids, assuming that the strongest acid releases the most stable conjugate base (halide anion). With the exception of HF (pK_a = 3.2), all the hydrohalic acids are very strong, small differences being in the direction HCl < HBr < HI.

Substitution and Elimination

The characteristics noted above lead us to anticipate certain types of reactions that are likely to occur with alkyl halides. In describing these, it is useful to designate the halogen-bearing carbon as alpha and the carbon atom(s) adjacent to it as beta, as noted in the first four equations shown below. Replacement or substitution of the halogen on the α-carbon (colored maroon) by a nucleophilic reagent is a commonly observed reaction, as shown in equations 1, 2, 5, 6 & 7 below. Also, since the electrophilic character introduced by the halogen extends to the β-carbons, and since nucleophiles are also bases, the possibility of base induced H-X elimination must also be considered, as illustrated by equation 3. Finally, there are some combinations of alkyl halides and nucleophiles that fail to show any reaction over a 24 hour period, such as the example in equation 4. For consistency, alkyl bromides have been used in these examples. Similar reactions occur when alkyl chlorides or iodides are used, but the speed of the reactions and the exact distribution of products will change.

In order to understand why some combinations of alkyl halides and nucleophiles give a substitution reaction, whereas other combinations give elimination, and still others give no observable reaction, we must investigate systematically the way in which changes in reaction variables perturb the course of the reaction. The following general equation summarizes the factors that will be important in such an investigation.

One conclusion, relating the structure of the R-group to possible products, should be immediately obvious. If R- has no beta-hydrogens an elimination reaction is not possible, unless a structural rearrangement occurs first. The first four halides shown on the left below do not give elimination reactions on treatment with base, because they have no β-hydrogens. The two halides on the right do not normally undergo such reactions because the potential elimination products have highly strained double or triple bonds.
It is also worth noting that sp² hybridized C–X compounds, such as the three on the right, do not normally undergo nucleophilic substitution reactions, unless other functional groups perturb the double bond(s).

Using the general reaction shown above as our reference, we can identify the following variables and observables.

When several reaction variables may be changed, it is important to isolate the effects of each during the course of study. In other words: only one variable should be changed at a time, the others being held as constant as possible. For example, we can examine the effect of changing the halogen substituent from Cl to Br to I, using ethyl as a common R–group, cyanide anion as a common nucleophile, and ethanol as a common solvent. We would find a common substitution product, C₂H₅–CN, in all cases, but the speed or rate of the reaction would increase in the order: Cl < Br < I. This reactivity order reflects both the strength of the C–X bond, and the stability of X^(–) as a leaving group, and leads to the general conclusion that alkyl iodides are the most reactive members of this functional class.

1. Nucleophilicity

Recall the definitions of electrophile and nucleophile:

Electrophile:   An electron deficient atom, ion or molecule that has an affinity for an electron pair, and will bond to a base or nucleophile.
Nucleophile:   An atom, ion or molecule that has an electron pair that may be donated in forming a covalent bond to an electrophile (or Lewis acid).

If we use a common alkyl halide, such as methyl bromide, and a common solvent, ethanol, we can examine the rate at which various nucleophiles substitute the methyl carbon. Nucleophilicity is thereby related to the relative rate of substitution reactions at the halogen-bearing carbon atom of the reference alkyl halide. The most reactive nucleophiles are said to be more nucleophilic than less reactive members of the group. The nucleophilicities of some common Nu:^(–) reactants vary as shown in the following chart.

Nucleophilicity:   CH₃CO₂^(–) < Cl^(–) < Br^(–) < N₃^(–) < CH₃O^(–) < CN^(–) < I^(–) < SCN^(–) < I^(–) < CH₃S^(–)

The reactivity range encompassed by these reagents is over 5,000 fold, thiolate being the most reactive. Note that by using methyl bromide as the reference substrate, the complication of competing elimination reactions is avoided. The nucleophiles used in this study were all anions, but this is not a necessary requirement for these substitution reactions. Indeed reactions 6 & 7, presented at the beginning of this section, are examples of neutral nucleophiles participating in substitution reactions. The cumulative results of studies of this kind has led to useful empirical rules pertaining to nucleophilicity:

(i) For a given element, negatively charged species are more nucleophilic (and basic) than are equivalent neutral species.
(ii) For a given period of the periodic table, nucleophilicity (and basicity) decreases on moving from left to right.
(iii) For a given group of the periodic table, nucleophilicity increases from top to bottom (i.e. with increasing size), although there is a solvent dependence due to hydrogen bonding. Basicity varies in the opposite manner.

2. Solvent Effects

Solvation of nucleophilic anions markedly influences their reactivity. The nucleophilicities cited above were obtained from reactions in methanol solution. Polar, protic solvents such as water and alcohols solvate anions by hydrogen bonding interactions, as shown in the diagram on the right. These solvated species are more stable and less reactive than the unsolvated "naked" anions. Polar, aprotic solvents such as DMSO (dimethyl sulfoxide), DMF (dimethylformamide) and acetonitrile do not solvate anions nearly as well as methanol, but provide good solvation of the accompanying cations. Consequently, most of the nucleophiles discussed here react more rapidly in solutions prepared from these solvents. These solvent effects are more pronounced for small basic anions than for large weakly basic anions. Thus, for reaction in DMSO solution we observe the following reactivity order:

Nucleophilicity:   I^(–) < SCN^(–) < Br^(–) < Cl^(–) ≈ N₃^(–) < CH₃CO₂ ^(–) < CN^(–) ≈ CH₃S^(–) < CH₃O^(–)

Note that this order is roughly the order of increasing basicity.

3. The Alkyl Moiety

Some of the most important information concerning nucleophilic substitution and elimination reactions of alkyl halides has come from studies in which the structure of the alkyl group has been varied. If we examine a series of alkyl bromide substitution reactions with the strong nucleophile thiocyanide (SCN) in ethanol solvent, we find large decreases in the rates of reaction as alkyl substitution of the alpha-carbon increases. Methyl bromide reacts 20 to 30 times faster than simple 1Ί-alkyl bromides, which in turn react about 20 times faster than simple 2Ί-alkyl bromides, and 3Ί-alkyl bromides are essentially unreactive or undergo elimination reactions. Furthermore, β-alkyl substitution also decreases the rate of substitution, as witnessed by the failure of neopentyl bromide, (CH₃)₃CCH₂-Br (a 1Ί-bromide), to react.
Alkyl halides in which the alpha-carbon is a chiral center provide additional information about these nucleophilic substitution reactions. Returning to the examples presented at the beginning of this section, we find that reactions 2, 5 & 6 demonstrate an inversion of configuration when the cyanide nucleophile replaces the bromine. Other investigations have shown this to be generally true for reactions carried out in non-polar organic solvents, the reaction of (S)-2-iodobutane with sodium azide in ethanol being just one example ( in the following equation the alpha-carbon is maroon and the azide nucleophile is blue). Inversion of configuration during nucleophilic substitution has also been confirmed for chiral 1Ί-halides of the type RCDH-X, where the chirality is due to isotopic substitution.

(S)-CH₃CHICH₂CH₃   +   NaN₃   ——>  (R)-CH₃CHN₃CH₂CH₃   +   NaI

We can now piece together a plausible picture of how nucleophilic substitution reactions of 1Ί and 2Ί-alkyl halides take place. The nucleophile must approach the electrophilic alpha-carbon atom from the side opposite the halogen. As a covalent bond begins to form between the nucleophile and the carbon, the carbon halogen bond weakens and stretches, the halogen atom eventually leaving as an anion. The diagram on the right shows this process for an anionic nucleophile. We call this description the S_N2 mechanism, where S stands for Substitution, N stands for Nucleophilic and 2 stands for bimolecular (defined below). In the S_N2 transition state the alpha-carbon is hybridized sp² with the partial bonds to the nucleophile and the halogen having largely p-character. Both the nucleophile and the halogen bear a partial negative charge, the full charge being transferred to the halogen in the products. The consequence of rear-side bonding by the nucleophile is an inversion of configuration about the alpha-carbon. Neutral nucleophiles react by a similar mechanism, but the charge distribution in the transition state is very different.
This mechanistic model explains many aspects of the reaction. First, it accounts for the fact that different nucleophilic reagents react at very different rates, even with the same alkyl halide. Since the transition state has a partial bond from the alpha-carbon to the nucleophile, variations in these bond strengths will clearly affect the activation energy, ΔE^‡, of the reaction and therefore its rate. Second, the rear-side approach of the nucleophile to the alpha-carbon will be subject to hindrance by neighboring alkyl substituents, both on the alpha and the beta-carbons. The following models clearly show this "steric hindrance" effect.

The two models displayed below start as methyl bromide, on the left, and ethyl bromide, on the right. These may be replaced by isopropyl, tert-butyl, neopentyl, and benzyl bromide models by pressing the appropriate buttons. (note that when first activated, this display may require clicking twice on the selected button.) In each picture the nucleophile is designated by a large deep violet sphere, located 3.75 Angstroms from the alpha-carbon atom (colored a dark gray), and located exactly opposite to the bromine (colored red-brown). This represents a point on the trajectory the nucleophile must follow if it is to bond to the back-side of the carbon atom, displacing bromide anion from the front face. With the exception of methyl and benzyl, the other alkyl groups present a steric hindrance to the back-side approach of the nucleophile, which increases with substitution alpha and beta to the bromine. The hydrogen (and carbon) atoms that hinder the nucleophile's approach are colored a light red. The magnitude of this steric hindrance may be seen by moving the models about in the usual way, and is clearly greatest for tert-butyl and neopentyl, the two compounds that fail to give substitution reactions.

4. Molecularity

If a chemical reaction proceeds by more than one step or stage, its overall velocity or rate is limited by the slowest step, the rate-determining step. This "bottleneck concept" has analogies in everyday life. For example, if a crowd is leaving a theater through a single exit door, the time it takes to empty the building is a function of the number of people who can move through the door per second. Once a group gathers at the door, the speed at which other people leave their seats and move along the aisles has no influence on the overall exit rate. When we describe the mechanism of a chemical reaction, it is important to identify the rate-determining step and to determine its "molecularity". The molecularity of a reaction is defined as the number of molecules or ions that participate in the rate determinining step. A mechanism in which two reacting species combine in the transition state of the rate-determining step is called bimolecular. If a single species makes up the transition state, the reaction would be called unimolecular. The relatively improbable case of three independent species coming together in the transition state would be called termolecular.

5. Kinetics

One way of investigating the molecularity of a given reaction is to measure changes in the rate at which products are formed or reactants are lost, as reactant concentrations are varied in a systematic fashion. This sort of study is called kinetics, and the goal is to write an equation that correlates the observed results. Such an equation is termed a kinetic expression, and for a reaction of the type: A + B  –––>  C + D it takes the form:  Reaction Rate = k[A] ⁿ[B] ^m, where the rate constant k is a proportionality constant that reflects the nature of the reaction, [A] is the concentration of reactant A, [B] is the concentration of reactant B, and n & m are exponential numbers used to fit the rate equation to the experimental data. Chemists refer to the sum n + m as the kinetic order of a reaction. In a simple bimolecular reaction n & m would both be 1, and the reaction would be termed second order, supporting a mechanism in which a molecule of reactant A and one of B are incorporated in the transition state of the rate-determining step. A bimolecular reaction in which two molecules of reactant A (and no B) are present in the transition state would be expected to give a kinetic equation in which n=2 and m=0 (also second order). The kinetic expressions found for the reactions shown at the beginning of this section are written in blue in the following equations. Each different reaction has its own distinct rate constant, k_#. All the reactions save 7 display second order kinetics, reaction 7 is first order.

It should be recognized and remembered that the molecularity of a reaction is a theoretical term referring to a specific mechanism. On the other hand, the kinetic order of a reaction is an experimentally derived number. In ideal situations these two should be the same, and in most of the above reactions this is so. Reaction 7 above is clearly different from the other cases reported here. It not only shows first order kinetics (only the alkyl halide concentration influences the rate), but the chiral 3Ί-alkyl bromide reactant undergoes substitution by the modest nucleophile water with extensive racemization. Note that the acetonitrile cosolvent does not function as a nucleophile. It serves only to provide a homogeneous solution, since the alkyl halide is relatively insoluble in pure water.
One of the challenges faced by early workers in this field was to explain these and other differences in a rational manner.

Two discrete mechanisms for nucleophilic substitution reactions will be described in the next section.

Reactions of Alkyl Halides with Reducing Metals

The alkali metals (Li, Na, K etc.) and the alkaline earth metals (Mg and Ca, together with Zn) are good reducing agents, the former being stronger than the latter. Sodium, for example, reduces elemental chlorine to chloride anion (sodium is oxidized to its cation), as do the other metals under varying conditions. In a similar fashion these same metals reduce the carbon-halogen bonds of alkyl halides. The halogen is converted to halide anion, and the carbon bonds to the metal (the carbon has carbanionic character). Halide reactivity increases in the order: Cl < Br < I. The following equations illustrate these reactions for the commonly used metals lithium and magnesium (R may be hydrogen or alkyl groups in any combination). The alkyl magnesium halides described in the second reaction are called Grignard Reagents after the French chemist who discovered them. The other metals mentioned above react in a similar manner, but the two shown here are the most widely used. Although the formulas drawn here for the alkyl lithium and Grignard reagents reflect the stoichiometry of the reactions and are widely used in the chemical literature, they do not accurately depict the structural nature of these remarkable substances. Mixtures of polymeric and other associated and complexed species are in equilibrium under the conditions normally used for their preparation.

The metals referred to here are insoluble in most organic solvents, hence these reactions are clearly heterogeneous, i.e. take place on the metal surface. The conditions necessary to achieve a successful reaction are critical.
First, the metal must be clean and finely divided so as to provide the largest possible surface area for reaction.
Second, a suitable solvent must be used. For alkyl lithium formation pentane, hexane or ethyl ether may be used; but ethyl ether or THF are essential for Grignard reagent formation.
Third, since these organometallic compounds are very reactive, contaminants such as water, alcohols and oxygen must be avoided.

These reactions are obviously substitution reactions, but they cannot be classified as nucleophilic substitutions, as were the earlier reactions of alkyl halides. Because the functional carbon atom has been reduced, the polarity of the resulting functional group is inverted (an originally electrophilic carbon becomes nucleophilic). This change, shown below, makes alkyl lithium and Grignard reagents unique and useful reactants in synthesis.

Reactions of organolithium and Grignard reagents reflect the nucleophilic (and basic) character of the functional carbon in these compounds. Many examples of such reactions will be encountered in future discussions, and five simple examples are shown below. The first and third equations demonstrate the strongly basic nature of these compounds, which bond rapidly to the weakly acidic protons of water and methyl alcohol (colored blue). The nucleophilic carbon of these reagents also bonds readily with electrophiles such as iodine (second equation) and carbon dioxide (fifth equation). The polarity of the carbon-oxygen double bonds of CO₂ makes the carbon atom electrophilic, shown by the formula in the shaded box, so the nucleophilic carbon of the Grignard reagent bonds to this site. As noted above, solutions of these reagents must also be protected from oxygen, since peroxides are formed (equation 4).

Another important reaction exhibited by these organometallic reagents is metal exchange. In the first example below, methyl lithium reacts with cuprous iodide to give a lithium dimethylcopper reagent, which is referred to as a Gilman reagent. Other alkyl lithiums give similar Gilman reagents. A useful application of these reagents is their ability to couple with alkyl, vinyl and aryl iodides, as shown in the second equation. Later we shall find that Gilman reagents also display useful carbon-carbon bond forming reactions with conjugated enones and with acyl chlorides.

The formation of organometallic reagents from alkyl halides is more tolerant of structural variation than were the nucleophilic substitutions described earlier. Changes in carbon hybridization have little effect on the reaction, and 1Ί, 2Ί and 3Ί-alkyl halides all react in the same manner. One restriction, of course, is the necessary absence of incompatible functional groups elsewhere in the reactant molecule. For example, 5-bromo-1-pentanol fails to give a Grignard reagent (or a lithium reagent) because the hydroxyl group protonates this reactive function as soon as it is formed.

BrCH₂CH₂CH₂CH₂CH₂OH   +   Mg   ——>  [ BrMgCH₂CH₂CH₂CH₂CH₂OH ]   ——>  HCH₂CH₂CH₂CH₂CH₂OMgBr

Reactions of Dihalides

If two halogen atoms are present in a given compound, reactions with reducing metals may take different paths depending on how close the carbon-halogen bonds are to each other. If they are separated by four or more carbons, as in the first example below, a bis-organometallic compound may be formed. However, if the halogens are bonded to adjacent (vicinal) carbons, an elimination takes place with formation of a double bond. Since vicinal-dihalides are usually made by adding a halogen to a double bond, this reaction is mainly useful for relating structures to each other. The last example, in which two halogens are bonded to the same carbon, referred to as geminal (twinned), gives an unusual reagent which may either react as a carbon nucleophile or, by elimination, as a carbene. Such reagents are often termed carbenoid.

The solution structure of the Simmons-Smith reagent is less well understood than that of the Grignard reagent, but the formula given here is as useful as any that have been proposed. Other alpha-halogenated organometallic reagents, such as ClCH₂Li, BrCH₂Li, Cl₂CHLi and Cl₃CLi, have been prepared, but they are substantially less stable and must be maintained at very low temperature (ca. -100 Ί C) to avoid loss of LiX. The stability and usefulness of the Simmons-Smith reagent may be attributed in part to the higher covalency of the carbon-zinc bond together with solvation and internal coordination of the zinc. Hydrolysis (reaction with water) gives methyl iodide, confirming the basicity of the carbon; and reaction with alkenes gives cyclopropane derivatives, demonstrating the carbene-like nature of the reagent. The latter transformation is illustrated by the equation on the right.

Elimination reactions of the stereoisomeric 1,2-dibromo-1,2-diphenylethanes provide a nice summary of the principles discussed above. The following illustration shows first the meso-diastereomer and below it one enantiomer of the racemic-diastereomer. In each case two conformers are drawn within parentheses, and the anti-relationship of selected vicinal groups in each is colored green. The reaction proceeding to the left is a dehydrohalogenation induced by treatment with KOH in alcohol. Since this is a stereospecific elimination, each diastereomer gives a different stereoisomeric product. The reaction to the right is a dehalogenation (the reverse of halogen addition to an alkene), caused by treatment with iodide anion. Zinc dust effects the same reaction, but with a lower degree of stereospecificity. The mechanism of the iodide anion reaction is shown by red arrows in the top example. A similar mechanism explains the comparable elimination of the racemic isomer. In both reactions an anti-transition state is observed.

The two stereoisomers of 1-bromo-1,2-diphenylethene (shown on the left of the diagram) undergo a second dehydrobromination reaction on more vigorous treatment with base, as shown in the following equation. This elimination generates the same alkyne (carbon-carbon triple bond) from each of the bromo-alkenes. Interestingly, the (Z)-isomer (lower structure) eliminates more rapidly than the (E)-isomer (upper structure), again showing a preference for anti-orientation of eliminating groups.

C₆H₅CH=CBrC₆H₅   +   KOH   ——>    C₆H₅C≡CC₆H₅   +   KBr   +   H₂O

Preparation of Alkynes by Dehydrohalogenation

The last reaction shown above suggests that alkynes might be prepared from alkenes by a two stage procedure, consisting first of chlorine or bromine addition to the double bond, and secondly a base induced double dehydrohalogenation. For example, reaction of 1-butene with bromine would give 1,2-dibromobutane, and on treatment with base this vicinal dibromide would be expected to yield 1-bromo-1-butene followed by a second elimination to 1-butyne.

CH₃CH₂CH=CH₂   +   Br₂   ——>  CH₃CH₂CHBr–CH₂Br   +   base   ——>  CH₃CH₂CH=CHBr   +   base   ——>  CH₃CH₂C≡CH

In practice this strategy works, but it requires care in the selection of the base and solvent. If KOH in alcohol is used, the first elimination is much faster than the second, so the bromoalkene may be isolated if desired. Under more extreme conditions the second elimination takes place, but isomerization of the triple bond also occurs, with the more stable isomer (2-butyne) being formed along with 1-butyne, even becoming the chief product. To facilitate the second elimination and avoid isomerization the very strong base sodium amide, NaNH₂, may be used. Since ammonia is a much weaker acid than water (by a factor of 10¹⁸), its conjugate base is proportionally stronger than hydroxide anion (the conjugate base of water), and the elimination of HBr from the bromoalkene may be conducted at relatively low temperature. Also, the acidity of the sp-hybridized C-H bond of the terminal alkyne traps the initially formed 1-butyne in the form of its sodium salt.

CH₃CH₂C≡CH   +   NaNH₂   ——>  CH₃CH₂C≡C:^(–) Na⁽⁺⁾   +   NH₃

An additional complication of this procedure is that the 1-bromo-1-butene product of the first elimination (see previous equations) is accompanied by its 2-bromo-1-butene isomer, CH₃CH₂CBr=CH₂, and elimination of HBr from this bromoalkene not only gives 1-butyne (base attack at C-1) but also 1,2-butadiene, CH₃CH=C=CH₂, by base attack at C-3. Dienes of this kind, in which the central carbon is sp-hybridized, are called allenes and are said to have cumulated double bonds. They are usually less stable than their alkyne isomers.

Elimination Reactions

1. The E2 Reaction

We have not yet considered the factors that influence elimination reactions, such as example 3 in the group presented at the beginning of this section.

(3)   (CH₃)₃C-Br   +   CN^(–)   ——>  (CH₃)₂C=CH₂   +   Br^(–)   +   HCN

We know that t-butyl bromide is not expected to react by a S_N2 mechanism. Furthermore, the ethanol solvent is not sufficiently polar to facilitate a S_N1 reaction. The other reactant, cyanide anion, is a good nucleophile; and it is also a decent base, being about ten times weaker than bicarbonate. Consequently, a base-induced elimination seems to be the only plausible reaction remaining for this combination of reactants. To get a clearer picture of the interplay of these factors consider the reaction of a 2Ί-alkyl halide, isopropyl bromide, with two different nucleophiles.

In the methanol solvent used here, methanethiolate has greater nucleophilicity than methoxide by a factor of 100. Methoxide, on the other hand is roughly 10⁶ times more basic than methanethiolate. As a result, we see a clearcut difference in the reaction products, which reflects nucleophilicity (bonding to an electrophilic carbon) versus basicity (bonding to a proton). Kinetic studies of these reactions show that they are both second order (first order in R–Br and first order in Nu:^(–)), suggesting a bimolecular mechanism for each. The substitution reaction is clearly S_N2. The corresponding designation for the elimination reaction is E2. An energy diagram for the single-step bimolecular E2 mechanism is shown on the right. We should be aware that the E2 transition state is less well defined than is that of S_N2 reactions. More bonds are being broken and formed, with the possibility of a continuum of states in which the extent of C–H and C–X bond-breaking and C=C bond-making varies. For example, if the R–groups on the beta-carbon enhance the acidity of that hydrogen, then substantial breaking of C–H may occur before the other bonds begin to be affected. Similarly, groups that favor ionization of the halogen may generate a transition state with substantial positive charge on the alpha-carbon and only a small degree of C–H breaking. For most simple alkyl halides, however, it is proper to envision a balanced transition state, in which there has been an equal and synchronous change in all the bonds. Such a model helps to explain an important regioselectivity displayed by these elimination reactions.
If two or more structurally distinct groups of beta-hydrogens are present in a given reactant, then several constitutionally isomeric alkenes may be formed by an E2 elimination. This situation is illustrated by the 2-bromobutane and 2-bromo-2,3-dimethylbutane elimination examples given below.

By using the strongly basic hydroxide nucleophile, we direct these reactions toward elimination. In both cases there are two different sets of beta-hydrogens available to the elimination reaction (these are colored red and orange and the alpha carbon is blue). If the rate of each possible elimination was the same, we might expect the amounts of the isomeric elimination products to reflect the number of hydrogens that could participate in that reaction. For example, since there are three 1Ί-hydrogens (red) and two 2Ί-hydrogens (orange) on beta-carbons in 2-bromobutane, statistics would suggest a 3:2 ratio of 1-butene and 2-butene in the products. This is not observed, and the latter predominates by 4:1. This departure from statistical expectation is even more pronounced in the second example, where there are six 1Ί-beta-hydrogens compared with one 3Ί-hydrogen. These results point to a strong regioselectivity favoring the more highly substituted product double bond, an empirical statement generally called the Zaitsev Rule.

The main factor contributing to Zaitsev Rule behavior is the stability of the alkene. We noted earlier that carbon-carbon double bonds are stabilized (thermodynamically) by alkyl substituents, and that this stabilization could be evaluated by appropriate heat of hydrogenation measurements. Since the E2 transition state has significant carbon-carbon double bond character, alkene stability differences will be reflected in the transition states of elimination reactions, and therefore in the activation energy of the rate-determining steps. From this consideration we anticipate that if two or more alkenes may be generated by an E2 elimination, the more stable alkene will be formed more rapidly and will therefore be the predominant product. This is illustrated for 2-bromobutane by the energy diagram on the right. The propensity of E2 eliminations to give the more stable alkene product also influences the distribution of product stereoisomers. In the elimination of 2-bromobutane, for example, we find that trans-2-butene is produced in a 6:1 ratio with its cis-isomer.
The Zaitsev Rule is a good predictor for simple elimination reactions of alkyl chlorides, bromides and iodides as long as relatively small strong bases are used. Thus hydroxide, methoxide and ethoxide bases give comparable results. Bulky bases such as tert-butoxide tend to give higher yields of the less substituted double bond isomers, a characteristic that has been attributed to steric hindrance. In the case of 2-bromo-2,3-dimethylbutane, described above, tert-butoxide gave a 4:1 ratio of 2,3-dimethyl-1-butene to 2,3-dimethyl-2-butene ( essentially the opposite result to that obtained with hydroxide or methoxide). This point will be discussed further once we know more about the the structure of the E2 transition state.

Bredt's Rule

The importance of maintaining a planar configuration of the trigonal double-bond carbon components must never be overlooked. For optimum pi-bonding to occur, the p-orbitals on these carbons must be parallel, and the resulting doubly-bonded planar configuration is more stable than a twisted alternative by over 60 kcal/mole. This structural constraint is responsible for the existence of alkene stereoisomers when substitutuion patterns permit. It also prohibits certain elimination reactions of bicyclic alkyl halides, that might be favorable in simpler cases. For example, the bicyclooctyl 3Ί-chloride shown below appears to be similar to tert-butyl chloride, but it does not undergo elimination, even when treated with a strong base (e.g. KOH or KOC₄H₉). There are six equivalent beta-hydrogens that might be attacked by base (two of these are colored blue as a reference), so an E2 reaction seems plausible. The problem with this elimination is that the resulting double bond would be constrained in a severely twisted (non-planar) configuration by the bridged structure of the carbon skeleton. The carbon atoms of this twisted double-bond are colored red and blue respectively, and a Newman projection looking down the twisted bond is drawn on the right. Because a pi-bond cannot be formed, the hypothetical alkene does not exist. Structural prohibitions such as this are often encountered in small bridged ring systems, and are referred to as Bredt's Rule.

Bredt's Rule should not be applied blindly to all bridged ring systems. If large rings are present their conformational flexibility may permit good overlap of the p-orbitals of a double bond at a bridgehead. This is similar to recognizing that trans-cycloalkenes cannot be prepared if the ring is small (3 to 7-membered), but can be isolated for larger ring systems. The anti-tumor agent taxol has such a bridgehead double bond (colored red), as shown in the following illustration. The bicyclo[3.3.1]octane ring system is the smallest in which bridgehead double bonds have been observed. The drawing to the right of taxol shows this system. The bridgehead double bond (red) has a cis-orientation in the six-membered ring (colored blue), but a trans-orientation in the larger eight-membered ring.

2. Stereochemistry of the E2 Reaction

E2 elimination reactions of certain isomeric cycloalkyl halides show unusual rates and regioselectivity that are not explained by the principles thus far discussed. For example, trans-2-methyl-1-chlorocyclohexane reacts with alcoholic KOH at a much slower rate than does its cis-isomer. Furthermore, the product from elimination of the trans-isomer is 3-methylcyclohexene (not predicted by the Zaitsev rule), whereas the cis-isomer gives the predicted 1-methylcyclohexene as the chief product. These differences are described by the first two equations in the following diagram.
Unlike open chain structures, cyclic compounds generally restrict the spatial orientation of ring substituents to relatively few arrangements. Consequently, reactions conducted on such substrates often provide us with information about the preferred orientation of reactant species in the transition state. Stereoisomers are particularly suitable in this respect, so the results shown here contain important information about the E2 transition state.

The most sensible interpretation of the elimination reactions of 2- and 4-substituted halocyclohexanes is that this reaction prefers an anti orientation of the halogen and the beta-hydrogen which is attacked by the base. These anti orientations are colored in red in the above equations. The compounds used here all have six-membered rings, so the anti orientation of groups requires that they assume a diaxial conformation. The observed differences in rate are the result of a steric preference for equatorial orientation of large substituents, which reduces the effective concentration of conformers having an axial halogen. In the case of the 1-bromo-4-tert-butylcyclohexane isomers, the tert-butyl group is so large that it will always assume an equatorial orientation, leaving the bromine to be axial in the cis-isomer and equatorial in the trans. Because of symmetry, the two axial beta-hydrogens in the cis-isomer react equally with base, resulting in rapid elimination to the same alkene (actually a racemic mixture). This reflects the fixed anti orientation of these hydrogens to the chlorine atom. To assume a conformation having an axial bromine the trans-isomer must tolerate serious crowding distortions. Such conformers are therefore present in extremely low concentration, and the rate of elimination is very slow. Indeed, substitution by hydroxide anion predominates.
A similar analysis of the 1-chloro-2-methylcyclohexane isomers explains both the rate and regioselectivity differences. Both the chlorine and methyl groups may assume an equatorial orientation in a chair conformation of the trans-isomer, as shown in the top equation. The axial chlorine needed for the E2 elimination is present only in the less stable alternative chair conformer, but this structure has only one axial beta-hydrogen (colored red), and the resulting elimination gives 3-methylcyclohexene. In the cis-isomer the smaller chlorine atom assumes an axial position in the more stable chair conformation, and here there are two axial beta hydrogens. The more stable 1-methylcyclohexene is therefore the predominant product, and the overall rate of elimination is relatively fast.
An orbital drawing of the anti-transition state is shown on the right. Note that the base attacks the alkyl halide from the side opposite the halogen, just as in the S_N2 mechanism. In this drawing the α and β carbon atoms are undergoing a rehybridization from sp³ to sp² and the developing π-bond is drawn as dashed light blue lines. The symbol R represents an alkyl group or hydrogen. Since both the base and the alkyl halide are present in this transition state, the reaction is bimolecular and should exhibit second order kinetics. We should note in passing that a syn-transition state would also provide good orbital overlap for elimination, and in some cases where an anti-orientation is prohibited by structural constraints syn-elimination has been observed.
It is also worth noting that anti-transition states were preferred in several addition reactions to alkenes, so there is an intriguing symmetry to these inverse structural transformations.Having arrived at a useful and plausible model of the E2 transition state, we can understand why a bulky base might shift the regioselectivity of the reaction away from the most highly substituted double bond isomer. Steric hindrance to base attack at a highly substituted beta-hydrogen site would result in preferred attack at a less substituted site.

To see the effect of steric hindrance at a beta carbon on the E2 transition state  

3. The E1 Reaction

Just as there were two mechanisms for nucleophilic substitution, there are two elimination mechanisms. The E1 mechanism is nearly identical to the S_N1 mechanism, differing only in the course of reaction taken by the carbocation intermediate. As shown by the following equations, a carbocation bearing beta-hydrogens may function either as a Lewis acid (electrophile), as it does in the S_N1 reaction, or a Brψnsted acid, as in the E1 reaction.

Thus, hydrolysis of tert-butyl chloride in a mixed solvent of water and acetonitrile gives a mixture of 2-methyl-2-propanol (60%) and 2-methylpropene (40%) at a rate independent of the water concentration. The alcohol is the product of an S_N1 reaction and the alkene is the product of the E1 reaction. The characteristics of these two reaction mechanisms are similar, as expected. They both show first order kinetics; neither is much influenced by a change in the nucleophile/base; and both are relatively non-stereospecific.

(CH₃)₃C–Cl   +   H₂O   ——>  [ (CH₃)₃C(+) ]  +   Cl^(–)  +   H₂O   ——>  (CH₃)₃C–OH   +   (CH₃)₂C=CH₂   +   HCl   +   H₂O

To summarize, when carbocation intermediates are formed one can expect them to react further by one or more of the following modes:

1. The cation may bond to a nucleophile to give a substitution product.
2. The cation may transfer a beta-proton to a base, giving an alkene product.
3. The cation may rearrange to a more stable carbocation, and then react by mode #1 or #2.

Since the S_N1 and E1 reactions proceed via the same carbocation intermediate, the product ratios are difficult to control and both substitution and elimination usually take place.

Having discussed the many factors that influence nucleophilic substitution and elimination reactions of alkyl halides, we must now consider the practical problem of predicting the most likely outcome when a given alkyl halide is reacted with a given nucleophile. As we noted earlier, several variables must be considered, the most important being the structure of the alkyl group and the nature of the nucleophilic reactant. The nature of the halogen substituent on the alkyl halide is usually not very significant if it is Cl, Br or I. In cases where both S_N2 and E2 reactions compete, chlorides generally give more elimination than do iodides, since the greater electronegativity of chlorine increases the acidity of beta-hydrogens. Indeed, although alkyl fluorides are relatively unreactive, when reactions with basic nucleophiles are forced, elimination occurs (note the high electronegativity of fluorine).

 

Go to Main Menu

---

© M.EL-Fellah ,Chemistry Department, Garyounis University