Introduction

The C(sp³)–H bonds of inert alkanes generally have high bond dissociation energies, making their direct conversion into carbon-centered radicals capable of participating in bond-forming reactions difficult. Photocatalytic C(sp³)–H functionalization therefore often relies on photocatalysts to generate highly reactive radicals, followed by hydrogen atom transfer to activate the C–H bond.

Huang et al. reported an unusual class of reactions in which cyclooctane and electron-deficient alkenes undergo C(sp³)–H alkylation in the presence of chloride-containing components and light, even without the addition of an external photocatalyst.[1] Subsequent studies further showed that this phenomenon is associated with trace amounts of iron that are difficult to eliminate completely from the reaction system. Fe forms photochemically active Fe–Cl coordination species with chloride-containing components, which generate chlorine radicals through ligand-to-metal charge transfer; the chlorine radicals then activate alkane C–H bonds.[1]

This study revealed that ppm-level metals, under appropriate coordination environments, irradiation wavelengths, and reaction times, can transition from ordinary impurities into active components that make a measurable catalytic contribution.

1. C(sp³)–H Functionalization in the Absence of an Added Photocatalyst

1.1 Light irradiation and chloride-containing components jointly determine whether the reaction proceeds

The study used a Giese-type radical addition between cyclooctane and 2-benzylidenemalononitrile as the model reaction.

When tetrabutylammonium chloride (TBACl) was used under 365 nm irradiation, a 44% gas chromatography yield was obtained even without the addition of a conventional photocatalyst. When trimethylsilyl chloride (TMSCl) was subsequently used as the chloride-containing component under 400 nm irradiation, the model product was obtained in 98% gas chromatography yield and 95% isolated yield.[1]

Changes in the reaction conditions showed a clear pattern.[1]

No target product was formed in the absence of either light irradiation or the chloride-containing component; when the irradiation wavelength was changed from 400 nm to 450 nm, the reaction also essentially ceased.[1] These results indicate that C(sp³)–H functionalization depends on a photochemical process that responds to near-ultraviolet to violet light and is also associated with the chloride-containing component.

2. Chlorine Radicals Activate Alkane C(sp³)–H Bonds

2.1 Radical-trapping experiments support the involvement of carbon-centered radicals

When 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) or butylated hydroxytoluene (BHT) was added to the model system, the reaction was markedly inhibited. High-resolution mass spectrometry also detected the product formed by TEMPO trapping of an alkyl radical.[1]

These results support a reaction pathway involving a carbon-centered radical intermediate.

Additional evidence for the involvement of chlorine radicals in C–H activation came from the regioselectivity observed with n-pentane. After reaction of n-pentane, the ratio of functionalization products at the 1-, 2-, and 3-positions was approximately:

1.0 : 2.4 : 1.1

This distribution is consistent with the known selectivity of chlorine-radical-mediated hydrogen atom transfer.[1,3]

The reaction of a chlorine radical with an alkane can be represented as:

Cl· + R–H → HCl + R·

This step is a hydrogen atom transfer (HAT) process.

Cl· abstracts a hydrogen atom from the C–H bond, forming HCl while simultaneously generating the alkyl radical R·. Chlorine radicals have strong hydrogen-abstraction ability and can therefore participate in the cleavage of unactivated aliphatic C(sp³)–H bonds.[2,3]

2.2 The alkyl radical subsequently forms a new C–C bond

The resulting R· adds to an electron-deficient alkene to form a new carbon–carbon bond:

R· + electron-deficient alkene → carbon-centered radical addition intermediate

Substrate studies in the paper showed that alkenes with stronger electron-withdrawing character are better suited to this transformation. Electron-withdrawing groups facilitate radical addition to the alkene, stabilize the resulting carbon-centered radical, and favor the subsequent single-electron reduction process.[1]

Accordingly, the core sequence from an alkane to C–C bond formation can be summarized as:

R–H → R· → radical addition → C–C bond formation

Cl· is the species that directly performs the C–H bond activation.

The next key question is to identify the photochemical source of Cl·.

3. Trace Iron Is the Hidden Source of Photocatalytic Activity

3.1 The light-absorption characteristics of electron-deficient alkenes do not match the reaction wavelengths

The conversion of Cl⁻ to Cl· involves electron transfer. One possible pathway would therefore be for the electron-deficient alkene to first absorb light and form an excited state, then accept an electron from chloride ion and generate a chlorine radical.

Ultraviolet–visible absorption spectroscopy (UV–Vis) showed that the two representative electron-deficient alkenes examined in the study did not exhibit corresponding absorption above 350 nm.[1]

This result does not match the actual irradiation conditions of 365–400 nm. Direct photoexcitation of the alkene therefore cannot reasonably account for the sustained generation of Cl· during the reaction.[1]

3.2 ICP-MS detects trace Fe in the reaction system

The study used inductively coupled plasma mass spectrometry (ICP-MS) to analyze Fe, Cu, Ce, and Ni.

Across different reagents, the detected levels of these metals ranged approximately from 10 ppb to 6 ppm. The Fe and Cu contents of the complete reaction mixture were substantially higher than the levels detected in the individual reagents, corresponding to approximately the 10⁻⁴ mol% range for Fe or Cu after conversion.[1]

These results provide quantitative evidence for the presence of trace metals, but the mere presence of an element does not demonstrate that it performs a catalytic role. The key evidence comes from comparisons of metal activity at low concentrations and from spike-in experiments performed at concentrations matching those actually detected.

3.3 Fe at concentrations close to the measured level can already alter reaction kinetics

Fe, Cu, Ce, and Ni are all metals that could potentially participate in photochemical processes, but their behavior at low concentrations differs substantially.

As the concentration of the metal catalyst decreased, the rate of the model reaction decreased; within the low-concentration range investigated, Fe exhibited higher catalytic activity than Cu, Ce, and Ni.[1] On this basis, the paper concluded that Fe impurities play a major role in this apparently photocatalyst-free C(sp³)–H functionalization.[1]

Using a new batch of reaction vessels, the researchers added 10⁻⁴ mol% FeCl₃ to the model reaction. This amount was of the same order of magnitude as the Fe content detected in the complete reaction mixture. Addition of FeCl₃ markedly increased the rate of cyclooctane alkylation.[1]

These observations link elemental analysis to catalytic activity:

Trace Fe is quantitatively detected

→ Fe exhibits relatively high catalytic activity at low concentration

→ Addition of FeCl₃ at a concentration close to the measured level increases the reaction rate

Compared with simply detecting the presence of a metal, this set of experiments further demonstrates that Fe within the concentration range present in the actual reaction system already exerts a measurable chemical effect.

The 10⁻⁴ mol% FeCl₃ spike-in experiment reflects the effect of Fe at this concentration scale on the reaction rate rather than providing a simple stoichiometric explanation for the entire reaction. The total amount and origin of Fe in the actual system, as well as changes in Fe during the course of the reaction, were not fully determined.[1]

3.4 Trace Fe may be introduced through multiple experimental sources

The trace Fe present in the reaction system was not traced to any single piece of equipment or individual reagent. The researchers conducted experiments using reaction vessels made of different materials and with different geometries. Product formation was observed in glass, quartz, and polytetrafluoroethylene (PTFE) vessels; the reaction also proceeded when the magnetic stir bar was removed.[1]

In addition, the model reaction showed similar behavior in two other laboratories located in different cities.[1] These results indicate that the observed trace Fe should not simply be attributed to a particular reaction vessel or magnetic stir bar.

Reagents, laboratory equipment, gas lines, and the experimental environment may all introduce small amounts of metals. The Fe contribution from any single source may be extremely low, but once these sources enter the same reaction system, their combined contributions can constitute the actual metal background.[1]

Therefore, when assessing whether trace metals affect a reaction, the metal content of the complete reaction mixture is more representative of the actual reaction conditions than the metal level in any single reagent or piece of equipment. This is also an important reason why the original study further used ICP-MS to quantitatively analyze the complete reaction system.

4. Fe–Cl Coordination Species Generate Chlorine Radicals through LMCT

4.1 A chloride-containing environment imparts photochemical activity to trace Fe

The influence of trace Fe on the reaction cannot be attributed merely to the presence of elemental iron in the system. Experimental results indicate that the role of Fe is closely associated with the Fe–Cl coordination environment formed under chloride-containing conditions.

Previous studies have shown that chloride-coordinated Fe species formed between Fe(III) and Cl⁻ can undergo ligand-to-metal charge transfer (LMCT). Upon photoexcitation, an electron is transferred from the chloride ligand to the Fe(III) center, simultaneously generating a highly reactive chlorine radical.[2]

The experiments reported by Huang et al. are consistent with this chemical process.[1]

In a model system containing Fe³⁺ together with chloride-containing components, the UV–Vis spectrum displayed a characteristic absorption at approximately 365 nm, consistent with the reported spectral characteristics of tetrachloroferrate, [FeCl₄]⁻.[1]

Different Fe salts, when combined with TMSCl, were also able to promote C(sp³)–H functionalization. This indicates that the anion initially associated with the added iron salt is not the primary factor determining reaction activity; rather, the coordination structures formed by Fe in a chloride-containing environment are more closely associated with the observed activity.[1]

Ligand-interference experiments further support this interpretation. Addition of pyridine-containing ligands decreased the reaction yield, and a similar trend was observed when 1,10-phenanthroline was added to a reaction system containing FeCl₃.[1] These molecules can coordinate strongly with Fe, altering the coordination environment around the Fe center and thereby affecting the formation and reactivity of photochemically active Fe–Cl species.

4.2 LMCT converts light energy into radical reactivity

The core of Fe–Cl photochemistry is photoinduced electron transfer.

The key process can be simplified as:

Fe(III)–Cl + hν → Fe(II) + Cl·

Here, hν represents the energy of the absorbed photon.

Upon photoexcitation, the Cl ligand transfers an electron to the Fe(III) center:

① Fe(III) accepts the electron and forms a reduced iron species;

② the Fe-coordinated Cl loses an electron to form Cl·;

③ Cl· subsequently activates an alkane C–H bond through HAT.[2]

This process links light absorption to cleavage of the alkane C–H bond.

4.3 Mechanistic sequence from trace Fe to C–C bond formation

The principal reaction pathway supported by the available experimental results can be represented as follows:

① Fe–Cl photoexcitation and LMCT

Fe(III)–Cl + hν → Fe(II) + Cl·

② HAT by the chlorine radical

Cl· + R–H → HCl + R·

③ Addition of the alkyl radical to an electron-deficient alkene

R· + electron-deficient alkene → carbon-centered radical addition intermediate. For example:

Taking 2-benzylidenemalononitrile as an example:

R· + PhCH=C(CN)₂ → PhCH(R)–C·(CN)₂

④ The radical intermediate undergoes subsequent electron transfer and protonation to form the product

The experimental evidence presented in the paper is concentrated mainly on the sequence of Fe–Cl photoactivation → Cl· formation → HAT → alkyl-radical formation. The specific electron-transfer and protonation steps following radical addition were proposed mainly on the basis of the reaction results and related mechanistic considerations; the individual intermediates were not directly structurally characterized.[1]

4.4 [FeCl₄]⁻ is a representative Fe–Cl species rather than an established unique active structure

The absorption characteristics of [FeCl₄]⁻ are consistent with the approximately 365 nm absorption observed experimentally, and previous studies of Fe–Cl photocatalysis have also demonstrated that such chloroferrate species can generate Cl· through LMCT.[1,2]

However, the Fe content in the actual reaction is at the ppm level, and the study did not directly determine the complete coordination structures of all active Fe species. The authors also explicitly stated that the specific form of Fe in the trace-level system remains unclear.[1]

Accordingly, the experimental evidence supports the formation of LMCT-active Fe–Cl coordination species between Fe and chloride-containing components. [FeCl₄]⁻ provides a representative model for understanding their light absorption and reaction behavior, but the current evidence does not restrict the active Fe species in the actual system to a single [FeCl₄]⁻ structure.

This distinction also illustrates that, in studies of trace catalysis, establishing a reaction mode and directly identifying a single catalytic species represent different levels of mechanistic evidence.

5. Why ppm-Level Fe Can Still Make a Significant Contribution to the Reaction

A common question in trace-metal studies is: if the Fe content is already at the ppm level or even lower, why can it still substantially alter an organic reaction?

The reason is that, in a catalytic system, there is no simple one-time stoichiometric relationship between the metal concentration and the final amount of substrate that can be converted.

5.1 Catalytic centers can generate reactive species repeatedly

In Fe–Cl photochemistry, the primary role of Fe is not to act as a stoichiometric reagent that binds to one substrate molecule and is then permanently consumed. Instead, Fe participates in photoinduced redox processes and generates Cl·.

After the subsequent reaction steps, the iron center can re-enter the catalytic process, allowing the same Fe center to participate in multiple substrate-conversion events.

In another study in which an iron salt was explicitly used as a photocatalyst, Jin et al. used Fe(III)–Cl LMCT to generate chlorine radicals and achieved C(sp³)–H functionalization of alkanes including ethane and propane, reporting turnover numbers (TONs) of up to approximately 8,000.[2] This demonstrates that Fe–Cl photochemistry is capable of continuously generating radicals and completing multiple catalytic turnovers at relatively low iron loadings.

The 10⁻⁴ mol% FeCl₃ spike-in experiment reported by Huang et al., meanwhile, demonstrated within the present system itself that Fe at a concentration close to the measured level can already alter the reaction rate.[1]

Taken together, these observations explain why very low Fe concentrations cannot be dismissed solely on the basis of their abundance.

5.2 Continuous irradiation supplies energy to the catalytic process

Fe–Cl LMCT requires photon absorption. The extent of the effect of trace Fe is therefore also influenced by the irradiation conditions.

Huang et al. compared different light-source powers.[1]

Under 400 nm irradiation:

① a 5 W light source gave 22% of the target product;

② a 20 W light source gave 50%;

③ the standard 50 W conditions gave 98%;

④ the 100 W conditions gave 70%.

These endpoint data cannot simply be interpreted as showing that “higher power gives higher yield,” because excessively high light power may also affect other processes in the system. However, kinetic experiments showed that reducing the light power substantially slowed product formation. No product was detected during the first 10 h of irradiation with the 5 W LED.[1]

This indicates that the effect of trace Fe is not determined by concentration alone; photon input also determines the extent to which Fe–Cl photochemistry can proceed within a given period.

5.3 Long reaction times allow the contribution of a slow catalytic process to accumulate

Many substrate reactions in this study employed prolonged irradiation, with some reactions extending to 48 h.[1]

When the Fe concentration is very low, the number of reactive radicals generated from trace Fe per unit time may be limited. Continuous irradiation allows Fe–Cl photochemistry to occur repeatedly, so that product generated through a slow catalytic process gradually accumulates.

Accordingly, the influence of trace metals on the reaction can be understood in terms of four interrelated factors:

The study therefore shows that “very low concentration” does not necessarily correspond to “very little chemical effect.” When a low-concentration metal can participate in an efficient photocatalytic cycle and is continuously exposed to light of an appropriate wavelength over an extended period, its catalytic contribution can gradually become experimentally observable.[1,2]

6. How to Experimentally Assess Trace-Metal Effects

Another important contribution of this study is the establishment of a relatively comprehensive approach for evaluating trace catalysis. For reactions that appear to proceed without an added metal catalyst, neither a single blank experiment nor elemental analysis alone is sufficient to determine the true source of catalytic activity; different types of evidence must be considered together.

6.1 Elemental quantification answers the question: “How much metal is actually present in the system?”

ICP-MS can convert the possibility of “metal contamination” into quantitative data.

The analysis should not be limited to a single reagent. Huang et al. showed that Fe and Cu levels in the complete reaction mixture were higher than those measured in the individual reagents.[1]

Therefore, when investigating trace catalysis, elemental analysis can cover:

① key starting materials and additives;

② solvents;

③ the complete reaction mixture;

④ experimental systems from different batches.

The resulting data more closely reflect the actual metal environment experienced by the catalytic reaction.

6.2 Concentration-matched spike-in experiments answer the question: “Is the measured concentration already chemically active?”

If ppm-level Fe is detected in a system but the validation experiment uses 1 mol% FeCl₃, even a substantial acceleration of the reaction cannot demonstrate that ppm-level Fe can produce the same effect.

The experimental design used by Huang et al. was more targeted: the FeCl₃ concentration was controlled at approximately 10⁻⁴ mol%, close to the measured Fe level in the reaction system, and the resulting changes in reaction kinetics were then examined.[1]

Such concentration-matched experiments establish a direct connection between “elemental presence” and “reactivity at the actual concentration.”

6.3 Time–yield kinetics provide more information than a single endpoint yield

Trace catalysts often affect the reaction rate first.

Two systems may show similar final conversions after 48 h, yet their product-formation rates during the first few hours may differ substantially. Comparing endpoint yields alone can obscure such differences.

In this study, time–yield relationships were used to compare different metals and different Fe concentrations, thereby identifying the effect of low-concentration Fe on the reaction rate.[1]

For studies of trace catalysis, kinetic profiles can therefore provide more direct information than a single endpoint measurement.

6.4 Spectroscopic and wavelength-dependent experiments support the involvement of Fe–Cl species in the photochemical reaction

For a proposed photochemical mechanism to be valid, the light-absorption process must also be compatible with the actual light source.

In this study:

① the electron-deficient alkenes showed no corresponding absorption above 350 nm;

② the Fe³⁺/Cl⁻ model system displayed an absorption at approximately 365 nm;

③ irradiation at 365–400 nm could drive the reaction;

④ under 450 nm irradiation, the reaction essentially ceased.[1]

The agreement between the spectroscopic data and wavelength dependence makes Fe–Cl photoactivation more consistent with the experimental results than direct photoabsorption by the substrate.

6.5 Coordination-interference experiments connect the metal element with its specific chemical form

Elemental analysis can detect Fe, but it cannot establish the chemical form in which Fe exerts its effect.

Pyridine-based ligands and 1,10-phenanthroline can alter the coordination environment of Fe and suppress the reaction.[1] When considered together with the Fe/Cl⁻ spectroscopic experiments, these results further associate the source of activity with Fe–Cl coordination chemistry rather than simply with the presence of iron.

For similar systems, a set of complementary criteria can be summarized as follows:

These experiments address different questions, including “how much is present,” “whether it is active,” “in what form it participates,” and “how it is photoexcited.” Only when multiple lines of evidence are mutually consistent can the identity of the catalyst be assigned with strong mechanistic explanatory power.

7. Representative Chemicals Related to Trace-Iron-Driven Alkane C(sp³)–H Photofunctionalization

Table 1. C(sp³)–H Substrates and Radical Acceptors

Table 2. Chloride-Containing Reaction Components, Solvents, and Analytical Reagents

Table 3. Iron Salts, Trace-Metal Controls, and Mechanistic Validation Reagents

Note: The products listed above are representative Aladdin products related to scientific research. Their specific applications should be determined in conjunction with product specifications, batch-specific COAs, and the target reaction or evaluation system. Additional information on product specifications, grades, and COAs can be searched on the Aladdin website using the “product name/CAS/catalog number.”

References

[1] Huang Y, Wang M, Liu W, Wu Q, Hu P. Unraveling the Prominent Existence of Trace Metals in Photocatalysis: Exploring Iron Impurity Effects. The Journal of Organic Chemistry, 2024, 89(6): 4156–4164. DOI: 10.1021/acs.joc.4c00155.

[2] Jin Y, Zhang Q, Wang L, Wang X, Meng C, Duan C. Convenient C(sp³)–H Bond Functionalisation of Light Alkanes and Other Compounds by Iron Photocatalysis. Green Chemistry, 2021, 23(18): 6984–6989. DOI: 10.1039/D1GC01563J.

[3] Deng H P, Zhou Q, Wu J. Microtubing-Reactor-Assisted Aliphatic C–H Functionalization with HCl as a Hydrogen-Atom-Transfer Catalyst Precursor in Conjunction with an Organic Photoredox Catalyst. Angewandte Chemie International Edition, 2018, 57(39): 12661–12665. DOI: 10.1002/anie.201804844.

For more related articles, see below:

Radical Initiator Selection Guide

Determination of the rate of oxygen radical production

Comparison of Principles, System Control, and Result Interpretation in Hydroxyl Radical Scavenging Capacity Assays

Structural and Catalytic Roles of Metal Ions in Protein Enzyme Activity

Experimental Selection and Outcome Assessment of Heterogeneous Metal Catalysts in Organic Synthesis

Experimental Selection Logic for Transition-Metal-Catalyzed C–H Functionalization: When It Is Worth Pursuing and How to Assess Site Selectivity and Conditions

Visible Light Photoredox Catalysts

Photoredox Iridium Catalyst for Single Electron Transfer (SET) Cross-Coupling

How to Choose Reagents for Inorganic Trace Analysis — Grade Definitions & Aladdin Product List

From Photocatalytic Ring Opening to Skeletal Reconstruction: Reaction Logic and Synthetic Applications of Oxygen-Atom Replacement in Oxetanes

Experimental Judgment and Condition Selection for Visible-Light Catalysis in Organic Synthesis