Part III: Functional Group Chemistry | Chapter 1

Carbonyl Chemistry

The carbonyl group as the central reactive motif of organic chemistry: nucleophilic additions, enolate chemistry, the aldol and Claisen condensations, Grignard and Wittig reactions, and their stereochemical control

1. Introduction: The Carbonyl Group

The Most Important Functional Group

The carbonyl group (C=O) is arguably the single most important functional group in organic chemistry. It appears in aldehydes, ketones, carboxylic acids, esters, amides, acid halides, and anhydrides — functional groups that collectively account for the majority of reactions in organic synthesis, biochemistry, and industrial chemistry.

What makes C=O so reactive? The answer lies in the large electronegativity difference between carbon and oxygen. Oxygen ($\chi = 3.44$) is far more electronegative than carbon ($\chi = 2.55$), creating a strongly polarized double bond with significant $\delta^+$ character on carbon and $\delta^-$ character on oxygen.

This chapter derives the key reactions of the carbonyl group from first principles — orbital interactions, thermodynamic and kinetic arguments, and stereochemical models — building a unified framework for understanding carbonyl reactivity.

The carbonyl $\pi$-bond consists of a $\sigma$-bond formed by sp$^2$ overlap and a $\pi$-bond formed by side-on overlap of p-orbitals. The $\pi^*$ antibonding orbital is the key electrophilic site: it is the LUMO of the carbonyl and sits at relatively low energy due to the electronegativity of oxygen pulling electron density away from carbon.

We can write the resonance contributors for a generic carbonyl:

$\text{C=O} \longleftrightarrow \text{C}^+\text{—O}^-$

The zwitterionic contributor emphasizes the electrophilic character of the carbonyl carbon.

The dipole moment of a typical carbonyl is approximately $\mu \approx 2.5\text{–}2.8 \text{ D}$, confirming the substantial charge separation. This polarization is the driving force behind nucleophilic addition — the most fundamental reaction mode of the carbonyl group.

The $\pi^*_{C=O}$ orbital energy depends on the substituents. Electron-donating groups (alkyl groups in ketones) raise the LUMO energy, making the carbonyl less electrophilic. Electron-withdrawing groups (halogens in acid halides) lower the LUMO, increasing electrophilicity. This simple orbital picture explains the reactivity ordering:

$\text{Acid halides} > \text{Anhydrides} > \text{Aldehydes} > \text{Ketones} > \text{Esters} > \text{Amides}$

2. Derivation 1: Nucleophilic Addition to Aldehydes and Ketones

2.1 The Orbital Picture

When a nucleophile (Nu$^-$) approaches a carbonyl compound, it donates electron density from its HOMO into the $\pi^*_{C=O}$ LUMO. The geometry of this approach is not random — it follows a well-defined trajectory first characterized by Bürgi and Dunitz in their landmark crystallographic studies.

The Bürgi-Dunitz Trajectory

The nucleophile approaches the carbonyl carbon at an angle of approximately $107°$ relative to the C=O bond axis (not $90°$ and not $180°$). This angle arises from the requirement to maximize overlap with the $\pi^*$ antibonding orbital while minimizing steric and electronic repulsion with the $\sigma$-bonding electrons.

The $\pi^*_{C=O}$ orbital has its largest lobe on carbon, tilted away from the oxygen. The optimal HOMO-LUMO overlap requires the nucleophile to approach from the side opposite to the oxygen lone pairs, at the Bürgi-Dunitz angle:

$\theta_{\text{BD}} \approx 107° \pm 5°$

This trajectory was established by H.B. Bürgi and J.D. Dunitz through analysis of crystal structures showing incipient nucleophilic attack on carbonyl groups (Bürgi, Dunitz, & Shefter, 1973).

2.2 The Tetrahedral Intermediate

As the nucleophile forms a new $\sigma$-bond to the carbonyl carbon, the carbon rehybridizes from sp$^2$ to sp$^3$, generating a tetrahedral alkoxide intermediate:

$\text{Nu}^- + \text{R}_2\text{C=O} \longrightarrow \text{R}_2\text{C(Nu)(O}^-\text{)}$

The energy profile for this process involves a single transition state in which the C—Nu bond is partially formed and the C=O $\pi$-bond is partially broken. The activation energy depends on:

Nucleophile strength: Better nucleophiles (higher HOMO energy) have lower activation barriers
Carbonyl electrophilicity: Lower LUMO energy leads to smaller HOMO-LUMO gap and faster reaction
Steric factors: Bulky substituents on either the nucleophile or carbonyl raise the barrier

2.3 Relative Reactivity: Aldehydes vs. Ketones

Aldehydes are more reactive toward nucleophilic addition than ketones. This difference arises from two reinforcing effects:

Steric Argument

In an aldehyde (RCHO), one substituent is hydrogen — the smallest possible group. In a ketone (R$_2$CO), both substituents are carbon-based, creating greater steric crowding around the electrophilic carbon. As the nucleophile approaches along the Bürgi-Dunitz trajectory, the 1,3-diaxial-like interactions with the ketone substituents raise the transition state energy.

Electronic Argument

Alkyl groups are weakly electron-donating via hyperconjugation ($\sigma_{C-H} \to \pi^*_{C=O}$ interactions). A ketone with two alkyl groups has more electron density donated into the $\pi$-system, raising the LUMO energy and reducing the electrophilicity of the carbonyl carbon. We can quantify this using infrared stretching frequencies:

$\tilde{\nu}_{C=O}(\text{HCHO}) \approx 1750 \text{ cm}^{-1}, \quad \tilde{\nu}_{C=O}(\text{(CH}_3\text{)}_2\text{CO}) \approx 1715 \text{ cm}^{-1}$

The lower stretching frequency for acetone reflects a weaker (more electron-rich) C=O bond, consistent with reduced electrophilicity.

Additionally, the product stability differs. The tetrahedral intermediate from a ketone has two alkyl groups creating 1,3-interactions in the sp$^3$ geometry. Thus both the transition state (kinetics) and the product (thermodynamics) favor aldehyde addition:

$K_{\text{eq}}(\text{hydration of HCHO}) \approx 2000 \gg K_{\text{eq}}(\text{hydration of (CH}_3\text{)}_2\text{CO}) \approx 0.002$

3. Derivation 2: The Aldol Reaction

3.1 Base-Catalyzed Aldol: Enolate Formation

The aldol reaction is one of the most powerful carbon–carbon bond-forming reactions in organic chemistry. It begins with formation of an enolate — the conjugate base of a carbonyl compound formed by deprotonation of an $\alpha$-hydrogen.

The acidity of $\alpha$-hydrogens arises from stabilization of the resulting anion by conjugation with the carbonyl:

$\text{R—CH}_2\text{—C(=O)—R'} + \text{B}^- \rightleftharpoons \text{R—CH=C(O}^-\text{)—R'} + \text{BH}$

Typical p$K_a$ values: ketones $\approx 20$, esters $\approx 25$, 1,3-dicarbonyls $\approx 9\text{–}13$

The enolate is an ambident nucleophile with significant electron density on both carbon and oxygen. The HOMO of the enolate is the $\pi$-system extending across C—C—O, with the largest coefficient on carbon (soft nucleophilic site) and oxygen (hard nucleophilic site).

3.2 Kinetic vs. Thermodynamic Enolates

For unsymmetrical ketones, deprotonation can occur on either side of the carbonyl. The two enolates have different stabilities and are formed at different rates:

Derivation of Selectivity

Kinetic enolate (less substituted): Formed preferentially at low temperature ($-78°\text{C}$) with a strong, sterically demanding base such as LDA (lithium diisopropylamide). The less hindered$\alpha$-proton is abstracted faster because the transition state has less steric strain. The kinetic enolate has the double bond to the less substituted side.

Thermodynamic enolate (more substituted): Formed under equilibrating conditions — weaker base (e.g., NaOEt), higher temperature, or longer reaction time. The more substituted enolate is more stable by analogy with alkene stability (hyperconjugative stabilization):

$\Delta G^\circ_{\text{thermo}} < \Delta G^\circ_{\text{kinetic}}$

but $\Delta G^\ddagger_{\text{kinetic}} < \Delta G^\ddagger_{\text{thermo}}$ (less steric strain in TS)

3.3 Nucleophilic Addition: The Aldol Step

The enolate, acting as a carbon nucleophile, attacks the electrophilic carbonyl of a second molecule. This produces a $\beta$-hydroxy carbonyl compound — the aldol product:

$\text{R—CH=C(O}^-\text{)—R'} + \text{R''CHO} \longrightarrow \text{R''CH(O}^-\text{)—CH(R)—C(=O)—R'}$

Protonation of the alkoxide gives the $\beta$-hydroxy aldehyde or ketone.

3.4 Aldol Condensation: Dehydration

Under more vigorous conditions (heat, excess base, or acid), the $\beta$-hydroxy carbonyl undergoes elimination of water to form an $\alpha,\beta$-unsaturated carbonyl compound. This overall process — aldol addition followed by dehydration — is the aldol condensation:

$\text{R—CH(OH)—CH}_2\text{—C(=O)—R'} \xrightarrow{-\text{H}_2\text{O}} \text{R—CH=CH—C(=O)—R'}$

The driving force for dehydration is the formation of an extended conjugated system — the$\alpha,\beta$-unsaturated carbonyl is stabilized by $\pi$-conjugation. The E1cb mechanism predominates under basic conditions: the $\alpha$-proton is removed first to form an enolate, followed by loss of hydroxide.

3.5 Stereoselectivity: The Zimmerman-Traxler Model

Chair-Like Transition State

Zimmerman and Traxler (1957) proposed that the aldol reaction proceeds through a six-membered, chair-like transition state in which the metal cation (Li$^+$, Zn$^{2+}$, B, etc.) coordinates both the enolate oxygen and the aldehyde oxygen:

$\text{M—O—C=C—C—O—M (six-membered ring, chair conformation)}$

In this chair-like TS, substituents can occupy axial or equatorial positions. The major product arises from the TS with bulky groups in equatorial positions (minimizing 1,3-diaxial interactions):

(Z)-enolate: The substituent on the enolate double bond is forced into an axial-like orientation if it is placed equatorial relative to R on the aldehyde. Analysis shows (Z)-enolates preferentially give the syn (or erythro) aldol product through a TS with both R groups equatorial.

(E)-enolate: With the opposite double bond geometry, the most favorable chair places substituents to give the anti (or threo) aldol product.

$(Z)\text{-enolate} \xrightarrow{\text{Zimmerman-Traxler}} syn\text{-aldol (major)}$

$(E)\text{-enolate} \xrightarrow{\text{Zimmerman-Traxler}} anti\text{-aldol (major)}$

4. Derivation 3: Grignard and Organolithium Reactions

4.1 The Grignard Reagent

Grignard reagents (RMgX, where X = Cl, Br, I) are formed by reaction of an organic halide with magnesium metal in anhydrous ether or THF. The resulting organomagnesium compound is a powerful nucleophile and a strong base due to the highly polar C—Mg bond ($\Delta\chi \approx 1.3$):

$\text{R—X} + \text{Mg} \xrightarrow{\text{Et}_2\text{O}} \text{R—MgX}$

4.2 Mechanism of Addition

The Grignard reagent adds to a carbonyl via a four-centered transition state. The magnesium coordinates to the carbonyl oxygen (Lewis acid activation), while the nucleophilic carbon attacks the electrophilic carbonyl carbon:

Step-by-Step Mechanism

Step 1 — Coordination: Mg$^{2+}$ coordinates to the lone pair on the carbonyl oxygen, lowering the LUMO energy of the C=O and activating it toward nucleophilic attack.

Step 2 — Nucleophilic addition: The carbanion character of R$^{\delta-}$in RMgX attacks the carbonyl carbon via a four-membered cyclic transition state:

$\text{R—MgX} + \text{R'CHO} \longrightarrow \text{R'CH(OMgX)R}$

Step 3 — Aqueous workup: Protonation of the magnesium alkoxide gives the alcohol:$\text{R'CH(OMgX)R} \xrightarrow{\text{H}_3\text{O}^+} \text{R'CH(OH)R}$

4.3 Retrosynthetic Disconnections

The Grignard reaction provides a powerful retrosynthetic disconnection for alcohols. Any secondary alcohol can be disconnected at a C—C bond adjacent to the OH to reveal a Grignard reagent and an aldehyde:

$\text{R'CH(OH)R} \Longrightarrow \text{R'CHO} + \text{RMgX}$

Or equivalently: $\text{R'CH(OH)R} \Longrightarrow \text{RCHO} + \text{R'MgX}$

For tertiary alcohols, three distinct disconnections are possible (using ketones or esters). This flexibility makes the Grignard reaction one of the most versatile tools in retrosynthetic analysis.

4.4 Limitations

Key Limitations of Grignard Reagents

Acidic protons: Grignard reagents are strong bases (p$K_a$ of conjugate acid$\approx 44$). They react immediately with water, alcohols, carboxylic acids, terminal alkynes, and amines. All protic functional groups must be absent or protected.
Double addition to esters: Esters undergo two successive additions with Grignard reagents. The first equivalent adds to the ester carbonyl, the tetrahedral intermediate collapses to expel the alkoxide leaving group to give a ketone, and the second equivalent adds to this ketone:
$\text{RCO}_2\text{R''} + 2\text{ R'MgX} \longrightarrow \text{R(R')}_2\text{COH}$ (tertiary alcohol)
Steric limitations: Very bulky Grignard reagents (e.g., tert-butylMgBr) may undergo reduction (hydride transfer) instead of addition when paired with sterically hindered ketones.
Organolithium alternatives: RLi reagents are more reactive than RMgX due to the greater ionic character of the C—Li bond ($\Delta\chi \approx 1.6$). They add to more hindered carbonyls but are also less selective.

5. Derivation 4: The Wittig Reaction

5.1 Phosphonium Ylides

The Wittig reaction converts a carbonyl compound to an alkene using a phosphorus ylide (Ph$_3$P=CR$_2$). The ylide is prepared in two steps:

Step 1: S$_\text{N}$2 reaction of triphenylphosphine with an alkyl halide:

$\text{Ph}_3\text{P} + \text{R—CH}_2\text{—X} \longrightarrow [\text{Ph}_3\text{P—CH}_2\text{R}]^+ \text{X}^-$

Step 2: Deprotonation with a strong base (n-BuLi, NaH, or NaHMDS):

$[\text{Ph}_3\text{P—CH}_2\text{R}]^+ + \text{B}^- \longrightarrow \text{Ph}_3\text{P=CHR} + \text{BH}$

The ylide can be described by two resonance forms:

$\text{Ph}_3\text{P=CHR} \longleftrightarrow \text{Ph}_3\text{P}^+\text{—CHR}^-$

The ylidic (zwitterionic) form emphasizes the carbanion character that enables nucleophilic attack.

5.2 Mechanism via Betaine and Oxaphosphetane

Detailed Mechanism

Step 1 — [2+2] Cycloaddition: The ylide carbon (nucleophile) attacks the carbonyl carbon (electrophile), while the phosphorus interacts with the carbonyl oxygen. This forms a four-membered ring intermediate called an oxaphosphetane:

$\text{Ph}_3\text{P=CHR} + \text{R'CHO} \longrightarrow \underset{\text{oxaphosphetane}}{\begin{bmatrix} \text{Ph}_3\text{P—O} \\ | \quad\quad\quad | \\ \text{RHC—CHR'} \end{bmatrix}}$

Whether the reaction proceeds through a discrete betaine intermediate (Ph$_3$P$^+$—CHR—CHR'—O$^-$) or directly to the oxaphosphetane has been debated. Current evidence (primarily from Vedejs and coworkers) supports direct [2+2] cycloaddition for non-stabilized ylides, while stabilized ylides may proceed via a betaine.

Step 2 — Retro-[2+2] cycloreversion: The oxaphosphetane fragments in a concerted, suprafacial process to give the alkene and triphenylphosphine oxide:

$\text{Oxaphosphetane} \longrightarrow \text{RHC=CHR'} + \text{Ph}_3\text{P=O}$

The driving force is the very strong P=O bond ($\approx 544 \text{ kJ mol}^{-1}$), making the reaction essentially irreversible.

5.3 E/Z Selectivity

The stereochemical outcome of the Wittig reaction depends critically on the nature of the ylide:

Selectivity Rules

Unstabilized ylides (no electron-withdrawing groups on the ylidic carbon, e.g., Ph$_3$P=CH$_2$ or Ph$_3$P=CHCH$_3$): These are highly reactive and form the oxaphosphetane under kinetic control. The cis-oxaphosphetane is formed preferentially (less steric strain in the early, puckered TS), giving predominantly the (Z)-alkene.

Stabilized ylides (electron-withdrawing group conjugated with the carbanion, e.g., Ph$_3$P=CHCO$_2$Et): These are less reactive and the reaction is reversible. Equilibration favors the thermodynamically more stable trans-oxaphosphetane, giving predominantly the (E)-alkene.

$\text{Unstabilized ylide} \longrightarrow (Z)\text{-alkene (kinetic control)}$

$\text{Stabilized ylide} \longrightarrow (E)\text{-alkene (thermodynamic control)}$

The Horner-Wadsworth-Emmons (HWE) modification uses phosphonate esters instead of phosphonium ylides and reliably gives (E)-alkenes. The Still-Gennari modification of HWE can give (Z)-alkenes selectively.

6. Derivation 5: Enolate Chemistry and the Claisen Condensation

6.1 Acidity of $\alpha$-Hydrogens

The acidity of hydrogens $\alpha$ to a carbonyl group is a cornerstone of carbonyl chemistry. These protons are far more acidic than typical C—H bonds ($\text{p}K_a \approx 50$) because the resulting carbanion is stabilized by delocalization into the carbonyl:

$\text{p}K_a \text{ values of } \alpha\text{-hydrogens:}$

$\text{Ketones} \approx 20, \quad \text{Esters} \approx 25, \quad \text{Nitriles} \approx 25$

$\text{1,3-Diketones} \approx 9, \quad \text{1,3-Ketoesters} \approx 11, \quad \text{1,3-Diesters} \approx 13$

6.2 Derivation of Enolate Stability

Why are enolates so stable? We can derive this from molecular orbital theory. Consider the deprotonation of acetaldehyde (CH$_3$CHO):

MO Analysis of Enolate Stabilization

Before deprotonation, the $\alpha$-C—H $\sigma$-bond is orthogonal to the carbonyl $\pi$-system. Upon removal of H$^+$, the remaining lone pair on carbon can conjugate with the C=O $\pi$-system, forming a three-center, four-electron allylic-type system:

$\psi_1 = c_1\phi_{\text{O}} + c_2\phi_{\text{C(=O)}} + c_3\phi_{\text{C}_\alpha}$

$\psi_2 = c_1'\phi_{\text{O}} + 0 \cdot \phi_{\text{C(=O)}} - c_3'\phi_{\text{C}_\alpha}$

$\psi_3 = c_1''\phi_{\text{O}} - c_2''\phi_{\text{C(=O)}} + c_3''\phi_{\text{C}_\alpha}$

The four electrons occupy $\psi_1$ (bonding) and $\psi_2$ (nonbonding). The stabilization energy relative to the localized carbanion is approximately:

$\Delta E_{\text{stab}} \approx \frac{\beta^2}{\Delta\epsilon}$

where $\beta$ is the resonance integral and $\Delta\epsilon$ is the energy gap between the carbon p-orbital and the carbonyl $\pi^*$. The stabilization is substantial ($\sim 60\text{–}80 \text{ kJ mol}^{-1}$), lowering the p$K_a$ by roughly$25\text{–}30$ units relative to a simple alkane.

For 1,3-dicarbonyl compounds, the carbanion is stabilized by two flanking carbonyls, leading to a five-center delocalized system with even greater stabilization. This explains the dramatically lower p$K_a$ values ($\approx 9\text{–}13$).

6.3 The Claisen Condensation

The Claisen condensation is the ester analog of the aldol reaction. An ester enolate attacks the carbonyl of a second ester molecule, with subsequent loss of alkoxide to form a $\beta$-ketoester:

Claisen Condensation Mechanism

Step 1: Alkoxide base deprotonates the $\alpha$-position of the ester:

$\text{RCH}_2\text{CO}_2\text{R'} + \text{R'O}^- \rightleftharpoons \text{RCH=C(O}^-\text{)OR'} + \text{R'OH}$

Step 2: The ester enolate attacks a second ester molecule at the carbonyl carbon:

$\text{Enolate} + \text{RCH}_2\text{CO}_2\text{R'} \longrightarrow \text{Tetrahedral intermediate}$

Step 3: The tetrahedral intermediate collapses, expelling alkoxide (R'O$^-$) as a leaving group:

$\text{Tetrahedral intermediate} \longrightarrow \text{RCH}_2\text{COCH(R)CO}_2\text{R'} + \text{R'O}^-$

Step 4 (Driving force): The $\beta$-ketoester product is deprotonated by alkoxide at the highly acidic position between the two carbonyls (p$K_a \approx 11$). This final, irreversible deprotonation drives the equilibrium to completion.

6.4 Dieckmann Cyclization

The intramolecular Claisen condensation is called the Dieckmann cyclization. A diester with an appropriate chain length ($\geq 5$ carbons between the ester groups) undergoes ring closure to form a cyclic $\beta$-ketoester:

$\text{R'O}_2\text{C—(CH}_2\text{)}_n\text{—CO}_2\text{R'} \xrightarrow{\text{R'O}^-} \text{Cyclic } \beta\text{-ketoester}$

Most favorable for five- and six-membered ring formation (n = 3 or 4), following Baldwin's rules for ring closure.

The Dieckmann cyclization is a key strategy in the synthesis of cyclopentanone and cyclohexanone derivatives, which are ubiquitous intermediates in natural product synthesis.

7. Applications

7.1 Total Synthesis Strategies

Carbonyl chemistry forms the backbone of nearly every total synthesis campaign. The aldol reaction, Wittig olefination, and Grignard addition are three of the most commonly employed C—C bond-forming reactions in the construction of complex natural products.

Retrosynthetic Logic

The aldol disconnection reveals a $\beta$-hydroxy carbonyl pattern in the target. The Wittig disconnection reveals an alkene adjacent to a carbonyl and phosphonium salt. The Grignard disconnection reveals an alcohol adjacent to a C—C bond. Together, these three disconnections provide routes to an enormous range of structural motifs.

Notable syntheses that rely heavily on carbonyl chemistry include E.J. Corey's synthesis of erythromycin (multiple aldol steps), R.B. Woodward's synthesis of cholesterol (Wittig olefinations), and K.C. Nicolaou's synthesis of taxol (Grignard additions and aldol reactions).

7.2 Pharmaceutical Synthesis

Statins

The statin family of drugs (atorvastatin/Lipitor, rosuvastatin/Crestor) are HMG-CoA reductase inhibitors that lower cholesterol. Their common pharmacophore is a $\beta$-hydroxy-$\delta$-lactone (or the open-chain dihydroxy acid), which is assembled by asymmetric aldol reactions in modern industrial syntheses. Evans oxazolidinone auxiliaries provide the stereochemical control needed for the syn-aldol disconnection.

Steroids

Steroid synthesis has historically been a showcase for carbonyl chemistry. The Robinson annulation — a Michael addition followed by an intramolecular aldol condensation — constructs the cyclohexenone ring system central to steroids. This elegant cascade creates a new six-membered ring with defined stereochemistry in a single operation.

7.3 Industrial Chemistry

Carbonyl compounds are produced on enormous industrial scales:

Formaldehyde ($\sim 52$ million tonnes/year): Produced by oxidation of methanol over silver or iron-molybdenum oxide catalysts. Used in phenol-formaldehyde resins (Bakelite), urea-formaldehyde adhesives, and as a disinfectant.
Acetaldehyde ($\sim 1.5$ million tonnes/year): Produced by the Wacker process (PdCl$_2$/CuCl$_2$ catalyzed oxidation of ethylene). Key intermediate for acetic acid, pyridine, and pentaerythritol production. The Wacker process involves nucleophilic addition of water to the Pd-coordinated alkene, followed by $\beta$-hydride elimination.
Acetone ($\sim 7$ million tonnes/year): Produced by the cumene process (oxidation of cumene to cumene hydroperoxide, acid-catalyzed Hock rearrangement). Co-produced with phenol in a 1:1 ratio.

8. Historical Context

Victor Grignard (1871–1935)

Victor Grignard discovered the organomagnesium reagents that bear his name while working as a doctoral student under Philippe Barbier at the University of Lyon. Barbier had shown that magnesium turnings could mediate the addition of organic halides to carbonyl compounds, but the yields were irreproducible. Grignard's key insight was to pre-form the organomagnesium compound in ether before adding the carbonyl substrate, giving reliable, high-yielding reactions.

Grignard was awarded the 1912 Nobel Prize in Chemistry (shared with Paul Sabatier) “for the discovery of the so-called Grignard reagent, which in recent years has greatly advanced the progress of organic chemistry.” The Grignard reaction remains one of the most taught and most used reactions in organic chemistry over a century later.

Georg Wittig (1897–1987)

Georg Wittig discovered the reaction that bears his name in 1954 while at the University of Heidelberg. The initial discovery was somewhat serendipitous: Wittig was studying the reactivity of pentaphenylphosphorane (Ph$_5$P) when he observed unexpected olefin formation upon treatment with benzophenone. He and his student Geissler quickly recognized the generality of the reaction.

Wittig received the 1979 Nobel Prize in Chemistry (shared with H.C. Brown) “for his development of the use of phosphorus-containing compounds into important reagents in organic synthesis.” The Wittig reaction's power lies in its regiospecificity — the new double bond forms exactly where the carbonyl was, with no possibility of the double bond migrating.

The Aldol Reaction: A Brief History

The aldol reaction was independently discovered by Charles-Adolphe Wurtz (1872) and Alexander Borodin (the composer-chemist!) in the 1870s. Wurtz observed that acetaldehyde treated with dilute acid gave a “aldol” (aldehyde-alcohol), naming both the product and the reaction.

The modern renaissance of the aldol reaction began in the 1970s–1980s when Heathcock, Evans, Masamune, and others developed methods for controlling the stereochemistry of the aldol using chiral auxiliaries, chiral Lewis acids, and chiral catalysts. Today, asymmetric aldol reactions (including proline-catalyzed direct aldol reactions, developed by List, Barbas, and Lerner in 2000) are among the most important tools in asymmetric synthesis.

9. Python Simulation

The following simulation generates two plots: (1) a reaction coordinate energy diagram comparing kinetic vs. thermodynamic control in the aldol reaction, and (2) a visualization of the Bürgi-Dunitz trajectory angle for nucleophilic attack on a carbonyl.

Simulation

Python

script.py235 lines

#!/usr/bin/env python3
"""
carbonyl_simulations.py
1. Reaction coordinate diagram: kinetic vs thermodynamic control in the aldol.
2. Burgi-Dunitz trajectory visualization.
Uses numpy only (no scipy).
"""
import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# =====================================================
# FIGURE 1: Aldol reaction coordinate energy diagram
# Kinetic vs Thermodynamic control
# =====================================================

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 6))

# --- Smooth energy profiles using Gaussian-like bumps ---
xi = np.linspace(0, 10, 500)

def gaussian(x, mu, sigma, amp):
    return amp * np.exp(-0.5 * ((x - mu) / sigma) ** 2)

# Kinetic pathway (lower barrier, higher product energy)
E_kinetic_base = 0.0
# Reactant well at x~1, TS at x~3.5, product at x~6
E_kin = (
    gaussian(xi, 1.0, 0.6, -2.0)   # reactant well
    + gaussian(xi, 3.5, 0.5, 6.0)  # TS (kinetic)
    + gaussian(xi, 6.0, 0.8, -3.5) # kinetic product well
    + 3.0                           # baseline shift
)

# Thermodynamic pathway (higher barrier, lower product energy)
E_thermo = (
    gaussian(xi, 1.0, 0.6, -2.0)   # same reactant
    + gaussian(xi, 4.0, 0.5, 7.5)  # TS (thermodynamic) - higher barrier
    + gaussian(xi, 7.0, 0.8, -5.5) # thermodynamic product well - lower energy
    + 3.0
)

ax1.plot(xi, E_kin, 'b-', linewidth=2.5, label='Kinetic pathway')
ax1.plot(xi, E_thermo, 'r-', linewidth=2.5, label='Thermodynamic pathway')

# Mark key points
# Reactant
ax1.annotate('Reactants\n(enolate + aldehyde)', xy=(1.0, 1.0), fontsize=9,
             ha='center', va='top', color='white',
             bbox=dict(boxstyle='round,pad=0.3', fc='gray', alpha=0.5))

# Kinetic TS
kin_ts_idx = np.argmax(E_kin[100:250]) + 100
ax1.annotate('TS (kinetic)\n$\\Delta G^{\\ddagger}_{kin}$',
             xy=(xi[kin_ts_idx], E_kin[kin_ts_idx]),
             xytext=(xi[kin_ts_idx]-0.5, E_kin[kin_ts_idx]+0.8),
             fontsize=9, color='cyan',
             arrowprops=dict(arrowstyle='->', color='cyan'),
             bbox=dict(boxstyle='round,pad=0.3', fc='darkblue', alpha=0.5))

# Thermodynamic TS
thermo_ts_idx = np.argmax(E_thermo[100:300]) + 100
ax1.annotate('TS (thermo)\n$\\Delta G^{\\ddagger}_{thermo}$',
             xy=(xi[thermo_ts_idx], E_thermo[thermo_ts_idx]),
             xytext=(xi[thermo_ts_idx]+0.5, E_thermo[thermo_ts_idx]+0.8),
             fontsize=9, color='orange',
             arrowprops=dict(arrowstyle='->', color='orange'),
             bbox=dict(boxstyle='round,pad=0.3', fc='darkred', alpha=0.5))

# Kinetic product
ax1.annotate('Kinetic product\n(less stable)', xy=(6.0, E_kin[300]-0.3),
             fontsize=9, ha='center', color='cyan',
             bbox=dict(boxstyle='round,pad=0.3', fc='darkblue', alpha=0.5))

# Thermodynamic product
ax1.annotate('Thermodynamic product\n(more stable)', xy=(7.0, E_thermo[350]-0.8),
             fontsize=9, ha='center', color='orange',
             bbox=dict(boxstyle='round,pad=0.3', fc='darkred', alpha=0.5))

# Annotations for conditions
ax1.text(0.5, 8.5, 'Low T, strong base (LDA)\n$\\rightarrow$ kinetic product',
         fontsize=8, color='cyan', style='italic')
ax1.text(5.5, 8.5, 'High T, weak base\n$\\rightarrow$ thermodynamic product',
         fontsize=8, color='orange', style='italic')

ax1.set_xlabel('Reaction Coordinate', fontsize=12, color='white')
ax1.set_ylabel('Free Energy (arbitrary units)', fontsize=12, color='white')
ax1.set_title('Aldol Reaction: Kinetic vs Thermodynamic Control', fontsize=13,
              color='white', fontweight='bold')
ax1.legend(fontsize=10, loc='upper left', facecolor='black', edgecolor='gray',
           labelcolor=['cyan', 'orange'])
ax1.set_facecolor('#1a1a2e')
ax1.tick_params(colors='white')
ax1.set_xlim(0, 10)
ax1.set_ylim(-1, 12)
ax1.spines['bottom'].set_color('gray')
ax1.spines['left'].set_color('gray')
ax1.spines['top'].set_visible(False)
ax1.spines['right'].set_visible(False)

# =====================================================
# FIGURE 2: Burgi-Dunitz Trajectory Visualization
# =====================================================

# Draw the carbonyl in the center
# C at origin, O along +y direction
C_pos = np.array([0.0, 0.0])
O_pos = np.array([0.0, 1.2])

# Substituents on C (sp2, 120 degree angles)
R1_pos = np.array([-1.0, -0.6])  # ~120 from C=O
R2_pos = np.array([1.0, -0.6])   # ~120 from C=O

# Draw bonds
ax2.plot([C_pos[0], O_pos[0]], [C_pos[1], O_pos[1]], 'w-', linewidth=4)
ax2.plot([C_pos[0]-0.04, O_pos[0]-0.04], [C_pos[1], O_pos[1]], 'w-', linewidth=2)
ax2.plot([C_pos[0]+0.04, O_pos[0]+0.04], [C_pos[1], O_pos[1]], 'w-', linewidth=2)
ax2.plot([C_pos[0], R1_pos[0]], [C_pos[1], R1_pos[1]], 'w-', linewidth=3)
ax2.plot([C_pos[0], R2_pos[0]], [C_pos[1], R2_pos[1]], 'w-', linewidth=3)

# Atom labels
ax2.text(C_pos[0]+0.15, C_pos[1]-0.1, 'C', fontsize=16, fontweight='bold',
         color='#4488ff', ha='center', va='center')
ax2.text(O_pos[0]+0.15, O_pos[1]+0.1, 'O', fontsize=16, fontweight='bold',
         color='#ff4444', ha='center', va='center')
ax2.text(R1_pos[0]-0.15, R1_pos[1]-0.1, 'R', fontsize=14, fontweight='bold',
         color='#88cc88', ha='center', va='center')
ax2.text(R2_pos[0]+0.15, R2_pos[1]-0.1, "R'", fontsize=14, fontweight='bold',
         color='#88cc88', ha='center', va='center')

# Charge annotations
ax2.text(C_pos[0]+0.35, C_pos[1]+0.15, '$\\delta^+$', fontsize=14, color='#6699ff')
ax2.text(O_pos[0]+0.3, O_pos[1]-0.05, '$\\delta^-$', fontsize=14, color='#ff6666')

# Draw the Burgi-Dunitz trajectory
# 107 degrees from the C=O bond, measured from C
# C=O points from C(0,0) to O(0,1.2), so the C=O direction is (0,1)
# The Nu approaches from below, at 107 degrees from C=O axis
bd_angle_deg = 107.0
bd_angle_rad = np.radians(bd_angle_deg)

# Direction opposite to O (downward), then rotated by (180-107)=73 degrees from -y axis
# The angle from the C=O bond axis: Nu approaches C, and the angle Nu-C-O = 107 degrees
# So from the C->O direction (upward), rotate 107 degrees clockwise
nu_direction = np.array([np.sin(bd_angle_rad), np.cos(bd_angle_rad)])
# This gives the direction FROM C TO Nu (approaching from upper-left)
# Actually let's think again: Burgi-Dunitz = angle(Nu-C-O) ~ 107
# Nu approaches from behind the carbonyl plane
# Let's place Nu approaching from the left-above
nu_direction = np.array([-np.sin(bd_angle_rad), np.cos(bd_angle_rad)])

# Trajectory line
t_vals = np.linspace(0.5, 2.5, 100)
traj_x = C_pos[0] + nu_direction[0] * t_vals
traj_y = C_pos[1] + nu_direction[1] * t_vals

ax2.plot(traj_x, traj_y, '--', color='#ffaa00', linewidth=2.5, alpha=0.8)

# Arrow showing Nu approach direction
nu_start = C_pos + nu_direction * 2.3
nu_end = C_pos + nu_direction * 0.6
ax2.annotate('', xy=nu_end, xytext=nu_start,
             arrowprops=dict(arrowstyle='->', color='#ffaa00', lw=2.5))

# Nu label
nu_label_pos = C_pos + nu_direction * 2.6
ax2.text(nu_label_pos[0], nu_label_pos[1], '$\\mathrm{Nu}^-$', fontsize=16,
         fontweight='bold', color='#ffaa00', ha='center', va='center')

# Draw the angle arc
arc_radius = 0.7
# Arc from C->O direction to C->Nu direction
# C->O angle from +x axis: 90 degrees
co_angle = 90.0  # degrees, since O is directly above C
# C->Nu angle from +x axis
nu_angle_from_x = np.degrees(np.arctan2(nu_direction[1], nu_direction[0]))

# Draw arc from the C->O direction to the C->Nu direction
if nu_angle_from_x > co_angle:
    theta_arc = np.linspace(np.radians(co_angle), np.radians(nu_angle_from_x), 50)
else:
    theta_arc = np.linspace(np.radians(nu_angle_from_x), np.radians(co_angle), 50)

arc_x = C_pos[0] + arc_radius * np.cos(theta_arc)
arc_y = C_pos[1] + arc_radius * np.sin(theta_arc)
ax2.plot(arc_x, arc_y, '-', color='#ff6600', linewidth=2)

# Angle label
mid_theta = (np.radians(co_angle) + np.radians(nu_angle_from_x)) / 2
label_r = arc_radius + 0.25
ax2.text(C_pos[0] + label_r * np.cos(mid_theta),
         C_pos[1] + label_r * np.sin(mid_theta),
         '$107°$', fontsize=13, color='#ff6600', fontweight='bold',
         ha='center', va='center')

# Title and label
ax2.set_title('Burgi-Dunitz Trajectory ($\\theta_{BD} \\approx 107°$)',
              fontsize=13, color='white', fontweight='bold')
ax2.text(0.0, -1.5, 'Nucleophile approaches C=O at ~107°\nfrom the C=O axis (not 90° or 180°)',
         fontsize=10, color='#cccccc', ha='center', va='center', style='italic')

# Draw pi* orbital lobes (schematic)
# Two lobes above and below the C=O bond, tilted
lobe_angles = np.linspace(0, 2*np.pi, 100)
# Upper lobe (on C, tilted toward Nu approach)
lobe_cx = C_pos[0] - 0.15
lobe_cy = C_pos[1] + 0.2
lobe_rx, lobe_ry = 0.25, 0.4
lobe_x = lobe_cx + lobe_rx * np.cos(lobe_angles) * np.cos(np.radians(15)) - lobe_ry * np.sin(lobe_angles) * np.sin(np.radians(15))
lobe_y = lobe_cy + lobe_rx * np.cos(lobe_angles) * np.sin(np.radians(15)) + lobe_ry * np.sin(lobe_angles) * np.cos(np.radians(15))
ax2.fill(lobe_x, lobe_y, alpha=0.2, color='#66aaff')
ax2.plot(lobe_x, lobe_y, '-', color='#66aaff', linewidth=1, alpha=0.5)

# Lower lobe
lobe2_cy = C_pos[1] - 0.5
lobe2_x = lobe_cx + lobe_rx * np.cos(lobe_angles) * np.cos(np.radians(15)) - lobe_ry * np.sin(lobe_angles) * np.sin(np.radians(15))
lobe2_y = lobe2_cy - (lobe_rx * np.cos(lobe_angles) * np.sin(np.radians(15)) + lobe_ry * np.sin(lobe_angles) * np.cos(np.radians(15)))
ax2.fill(lobe2_x, lobe2_y, alpha=0.15, color='#ff6666')
ax2.plot(lobe2_x, lobe2_y, '-', color='#ff6666', linewidth=1, alpha=0.5)

ax2.text(lobe_cx - 0.45, lobe_cy + 0.3, '$\\pi^*$', fontsize=11, color='#66aaff', alpha=0.8)

ax2.set_facecolor('#1a1a2e')
ax2.set_xlim(-3.5, 3.5)
ax2.set_ylim(-2.0, 3.5)
ax2.set_aspect('equal')
ax2.axis('off')

fig.patch.set_facecolor('#0d0d1a')
plt.tight_layout(pad=2.0)
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0d0d1a')
plt.close()
print("Plots saved to output.png")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

Part III Overview Carboxylic Acid Derivatives