CH₃I is the simplest system in the CX₃I series of molecules, and its photodissociation process has received the most attention from experimental and theoretical studies. The CH₃ group in the parent molecule has a trigonal pyramidal structure, but the equilibrium configuration of the CH₃ radical fragment is planar, with its umbrella vibration (Umbrella, ν₂) being considered the most likely and strongly excited vibration during photodissociation. Due to the lack of vibrationally resolved results, early experimental studies had disagreements over the highest populated vibrational state of the CH₃ fragment produced by the photodissociation of CH₃I. In 1985, Qihuo Zhu et al.^[14] used a self-developed rotatable laser photolysis molecular beam translational energy spectrometer to conduct a photodissociation experiment on CH₃I at 248 nm, obtaining a time-of-flight spectrum (TOF) for the iodine atom fragments (Figure 5). This spectrum showed two distinct peaks corresponding to the I* and I dissociation channels, and the quantum yield Φ(I*) was measured. Analysis indicated that in the I* channel, the highest populated peak of the CH₃ umbrella vibration excitation was at v₂ = 2, while in the I channel, it was at v₂ = 4. The experiment also measured the C-I bond dissociation energy of the CH₃I molecule as D₀ = 56 ± 1 kcal/mol. In subsequent experiments, spectra with better signal-to-noise ratios were obtained, and the ratio of internal energy excitation of the fragments to the available energy in the photodissociation process was determined, calculating an anisotropy parameter β = 1.85 for the spatial angular distribution of the photodissociation fragments, further confirming that both the I* and I channels at 248 nm originated from the parallel transition dissociation of the ³Q₀ state^[37⇓−39]. In 1986, Mr. Qihuo Zhu visited Professor Yuan Tseh Lee's laboratory at the University of California, Berkeley, and repeated the photodissociation experiment of CH₃I, acquiring vibrationally resolved translational energy spectra of the photodissociation fragments. It was experimentally confirmed that at 248 nm photodissociation, the highest populated vibrational excitation peak of the CH₃ fragment in the I* channel was v₂ = 0, and in the I channel, it was v₂ = 1; moreover, weak excitation of the C-H symmetric stretching vibration (ν₁) of the CH₃ radical was observed (the research results were not publicly published). Subsequent experimental and theoretical studies supported this conclusion^{[24,25,40-41]}.

Using a self-developed high-resolution miniature laser photolysis fragment translational energy spectrometer, Zhu Qihao et al.^[42−43] systematically carried out the study of photodissociation dynamics of CH₃I across the entire A band. At 277, 280, 297, and 304 nm in the middle and red end of the A band^[42], vibrational-resolved photofragment translational energy spectra (Figure 6) were obtained, yielding a series of photodissociation kinetic parameters (Table 1). Within the 277-304 nm range, only the excitation of the umbrella vibration of the CH₃ radical was observed in the I* channel, with its highest populated peak always remaining at v₂ = 0. Although the internal energy of CH₃ decreases as the photon energy of the dissociation laser decreases, the ratio of internal energy to available energy E_int/E_avl remains approximately 0.03, essentially unchanged. In the I channel, where the available energy is much greater than in the I* channel, weaker excitation of the symmetric stretching vibration of CH (ν₁) was observed, with the maximum reaching v₁ = 1. At shorter wavelengths such as 277.87 and 279.71 nm, the highest populated peak of CH₃ vibrational excitation is at (v₁ = 0, v₂ = 1). At 298.23 and 304.67 nm, the highest populated peak of CH₃ vibrational excitation is at (v₁ = 0, v₂ = 0). In this wavelength range, when the symmetric stretching vibration of CH is excited, the highest populated peak of the umbrella vibration is always at (v₁ = 1, v₂ = 0). As the wavelength increases, the available energy decreases, leading to a decrease in the internal energy of the CH₃ fragment, and the ratio E_int/E_avl drops from around 0.09 at 277 nm to around 0.06 at 304 nm. The total population of v₁ = 1 decreases from around 0.20 at 277 nm to around 0.06 and 0.03 at 304 nm. At the vertical transition of 304.67 nm (³Q₁ channel), the internal energy of the CH₃ fragment is slightly lower than that of the corresponding parallel transition (¹Q₁ $\leftarrow$ ³Q₀ channel). By measuring the anisotropy parameters of the spatial angular distribution of photofragments, it was found that within the 277-304 nm range, the I* channel all comes from contributions via the ³Q₀ potential surface; for dissociation wavelengths < 280 nm, the I channel all comes from contributions via the ¹Q₁ $\leftarrow$ ³Q₀ dissociation pathway, and for dissociation wavelengths > 280 nm, the contribution from ³Q₁ gradually increases, consistent with the composition of the A band absorption spectrum. At 277 and 304 nm, the branching ratios, I* quantum yield Φ(I*), and potential curve crossing probability P_cc for each channel were also measured, as shown in Table 1. According to the one-dimensional Landau-Zener formula, the energy E_c of the conical intersection point between the ¹Q₁ and ³Q₀ potential surfaces was calculated to be approximately 32 740 cm⁻¹.

In the blue end of the A band of CH₃I, its photodissociation exhibits significant differences from the middle and red ends. Zhu Qihao et al.^[43]measured the translational energy spectrum of photodissociation fragments at 225 nm, which showed good vibrational state resolution in both I and I* channels, through which the vibrational state population of CH₃fragments was obtained. The experiment found that regardless of whether in the I* or I channel, the symmetric stretching vibration of C-H in the CH₃fragment was strongly excited. In the I* channel, the ν ₁vibration of the CH₃fragment was maximally excited to v ₁ = 1; in the I channel, the ν ₁vibration of the CH₃fragment was maximally excited to v ₁ = 2. The peaks on the spectra exhibited a clear clustering phenomenon, indicating that the umbrella vibration of CH₃is not easily populated to v ₂ = 4 and above; only the I channel from the ¹Q₁$\leftarrow$³Q₀dissociation path did not show obvious clustering. The ratio of internal energy to available energy in the I* channel is significantly lower than that in the I channel. It was also found that the internal energy excitation of CH₃generated by parallel transitions initially excited to the ³Q₀state is higher than that generated by vertical transitions initially excited to the ¹Q₁state, possibly because the ³Q₀state has a relatively low energy, and when the parent molecule is excited to the ³Q₀state, the umbrella and C-H stretching vibrations of the CH₃group are more strongly excited, exhibiting a certain degree of vibrational adiabatic process. The branching ratios for the four dissociation pathways ³Q₀, ³Q₀ ← ¹Q₁, ¹Q₁, and ¹Q₁ ← ³Q₀were measured as 0.090.030.340.54, with the quantum yield of I* Φ(I*) = 0.12, β = 1.25 for the I* channel, and β = 0.84 for the I channel. The experimental crossing probability P _ccfor ¹Q₁ ← ³Q₀was 0.86, and for ³Q₀ ← ¹Q₁it was 0.08, proving that the one-dimensional Landau-Zener equation is not suitable for predicting the crossing probabilities between potential energy surfaces during the photodissociation of CH₃I at the blue end of the A band, as the one-dimensional Landau-Zener equation predicts equal P _ccfor the crossings of ³Q₀ ←¹Q₁and ¹Q₁ ← ³Q₀. The P _cc= 0.08 for ³Q₀←¹Q₁indicates that it is difficult for CH₃I molecules in the ¹Q₁state to cross over to the ³Q₀state.

Zhu Qihao et al. also observed the photodissociation signal of CH₃I parent molecules in a vibrationally excited state, which were excited by thermal excitation leading to C—I stretching vibrations (ν₃ʹ). It was found that the proportion of photodissociation signals from the ν₃ʹ vibration of the parent molecule gradually increased from 277 to 304 nm, with the thermal spectral effect of the I channel being lower than that of the corresponding I* channel; and after the vibrationally excited parent molecule absorbs a photon, its 1Q₁ ← 3Q₀ crossing probability P_cc decreases^[42]. The photodissociation of vibrationally excited parent molecules often exhibits significant differences compared to that of ground-state parent molecules^[44⇓-46], making it an important subject in state-to-state reaction dynamics and a crucial foundation for controlling chemical reactions in a favorable direction. In atmospheric and environmental chemistry, under natural conditions, there are many reactant molecules populating vibrational excited states, where the impact of initial vibrational excited state molecular photodissociation is more pronounced. To further study the photodissociation dynamics of vibrationally excited CH₃I molecules, Zhu Qihao et al. used infrared lasers to pre-excite the C-H symmetric stretching vibration to v₁ʹ = 1, and then studied its photodissociation dynamics in the ultraviolet region. While previous researchers mostly focused on the influence of bond-breaking vibrational excitation on photodissociation, Zhu Qihao et al. chose to explore other dimensions of vibrational excitation, which is of great significance for understanding the transfer and transformation of energy during the photodissociation process.

At 277 nm^[47], by detecting the iodine atoms and CH₃ fragment signals from the photodissociation of CH₃I (v₁ʹ = 1), it was found that the impact of initial v₁ʹ = 1 vibrational excitation on the photodissociation cross section of CH₃I is relatively small, with most of the initial v₁ʹ=1 vibrational energy being retained in the CH₃ fragments (Figure7). The umbrella vibrational excitation of CH₃ (v₁ = 1, v₂) is similar to that of the ground state CH₃I molecule when it undergoes photodissociation near 277 nm; and compared to the experimental results of the photodissociation of the ground state CH₃I at shorter wavelengths (225, 248 nm), it can be clearly confirmed that the v₁ = 1 vibrational excitation of the CH₃ fragment originates from the retention of the initial vibrational excitation, rather than due to an increase in dissociation energy. At this point, the initial vibrational excitation plays a role akin to a "spectator" in the photodissociation, and the process can be considered approximately vibrationally adiabatic. However, at 304 nm, the photodissociation of CH₃I (v₁ʹ = 1) is significantly different from that at 277 nm^[48]. By detecting the iodine atom fragments produced during the photodissociation, it was found that the photodissociation cross section of CH₃I (v₁ʹ = 1) is about 2.67 times larger than that of the ground state methyl iodide, showing a significant enhancement. The quantum yield Φ(I*) = 0.24 for the photoproduct I* is much higher than that for the photodissociation of the ground state methyl iodide; its ¹Q₁←³Q₀ crossing probability P_cc = 0.72 is lower than the P_cc value for the photodissociation of the ground state methyl iodide at this wavelength^[42,48]. By detecting the methyl and iodine fragments, the vibrational excitation modes and vibrational state populations of the CH₃ (ν₁, ν₂) fragments produced during the photodissociation of CH₃I (v₁ʹ = 1) were determined; it was found that in the I channel, the proportion of CH₃ fragments excited to v₁ = 1 is much higher than in the I* channel, indicating that in the I* channel, the initial vibrational excitation of the parent molecule is more difficult to retain in the fragments. Comparing the photodissociation data at 277 nm, it was found that at low photon energy photodissociation, the initial vibrational excitation of the parent molecule is more difficult to retain in the dissociation fragments. Whether at 277 or 304 nm, the ¹Q₁←³Q₀ crossing probability P_cc in the photodissociation of CH₃I (v₁ʹ = 1) is lower than that in the photodissociation of the ground state CH₃I molecule, which also leads to a higher quantum yield Φ(I*) for the dissociation product I*. Currently, both experimental and theoretical studies on the photodissociation of vibrationally excited CH₃I molecules are still limited and not systematic, making it impossible to comprehensively and accurately explain the kinetic mechanisms, thus further in-depth research is needed.

Zhu Qihet al also studied the photodissociation of the iodomethane dimer (CH₃I)₂ to produce (CH₃)₂I + I, finding that the translational energy of the produced fragment products was very small, revealing that van der Waals forces play an important role in the photodissociation process^[49].

When the H atom in CH₃I is replaced by F, the molecular symmetry remains unchanged. CF₃I, during photodissociation at the A band, involves potential energy surfaces and dissociation pathways consistent with those of CH₃I. However, the equilibrium configuration of the dissociation fragment CF₃ radical in its ground state is a trigonal pyramidal structure, which significantly differs from the planar structure of the CH₃ radical in its ground state; as F atoms are much heavier than H atoms, the vibrational modes and frequencies of the CF₃ radical also differ greatly from those of CH₃. Based on theoretical calculations by Clary^[33] and Bowman et al.^[50], this paper identifies the vibration of the CF₃ radical at 701 cm⁻¹ as the symmetric stretching (ν₁) mode of CF, and the vibration at 1086 cm⁻¹ as the umbrella ν₂ mode of the CF₃ fragment (this identification may be opposite to that in some literature, so readers should distinguish carefully). Clary^[33] and van Veen et al.^[32] predicted through theoretical calculations that, due to the presence of vibrational interactions on the excited state potential energy surface during the A-band photodissociation of CF₃I, the CF₃ fragment would exhibit significant dual-mode vibrational excitation (ν₁ and ν₂), but experimental studies have never been able to observe this phenomenon.

Zhu Qihao et al. conducted a photodissociation study of CF₃I at 248 nm using a molecular beam rotatable laser photolysis fragment translational energy spectrometer^[15], obtaining vibrationally resolved photofragment translational energy spectra in the I* channel. By identifying the spectra, they concluded that the peak vibrational population of the CF₃ fragment is at v₁ = 5, and based on this, calculated the C—I bond dissociation energy D₀ = 53.8 ± 0.6 kcal/mol, thus determining the ratio of internal energy to available energy E_int/E_avl for the photofragments in the I* channel. They also investigated the photofragment translational energy spectra in the I* channel at 281.73 nm^[51], identifying the peak vibrational population of the CF₃ fragment at v₁ = 2, with its E_int/E_avl being approximately 0.21. The spectra also showed a peak suspected to originate from bimodal vibrational excitation of the CF₃ fragment, but due to the limitations of spectral resolution and signal-to-noise ratio, it could not be clearly confirmed.

To further explore the dual-mode vibrational excitation of CF₃ fragments, Zhu Qihao et al. used a high-resolution micro laser photolysis fragment translational energy spectrometer to systematically study the photodissociation dynamics of CF₃I across the entire A band^[52⇓-54] (Table2). At 248 nm^[53], high-vibration-state-resolved I* channel photofragment translational energy spectra (Figure8) were obtained, where the spectral peaks could be clearly divided into two alternating groups with stronger and slightly weaker intensities. The energy interval ΔE_V between the higher peaks was about 700 cm⁻¹, which matches the vibration frequency of the ν₁ symmetric stretching mode of the CF₃ radical, and was identified as the (v₁ = 0 ~ 7, v₂ = 0) vibrational states of the CF₃ fragment, with the highest vibrational population peak at (v₁ = 4, v₂ = 0). The slightly weaker group of spectral peaks, located in the middle between two (v₁, v₂ = 0) peaks, had an energy gap consistent with the difference in vibration frequencies for (v₁, v₂ = 1), and was identified as the (v₁ = 0 ~ 5, v₂ = 1) states of the CF₃ fragment. This was the first clear experimental observation of dual-mode vibrational excitation of the CF₃ radical during the photodissociation of CF₃I. In the I* channels at 266 and 277 nm, two alternating groups of vibrational spectral peaks, also attributed to the dual-mode vibrational excitation of the CF₃ fragment, were observed; the highest vibrational population peaks were at (v₁ = 2, v₂ = 0) for 266 nm photolysis and at (v₁ = 1, v₂ = 0) for 277 nm. As the wavelength increased and photon energy decreased, the population of v₂ = 1 vibrational excitation significantly decreased. During photolysis at 248, 266, and 277 nm, the ratios of internal energy to available energy E_int/E_avl were 0.220, 0.145, and 0.108, respectively, and the ratios of ν₂ umbrella vibrational excitation ∑P(v₁, v₂ = 1)/∑P(v₁, v₂) were 0.41, 0.35, and 0.28, all of which rapidly decreased as the available energy decreased. Due to resolution limitations, it is not ruled out that the ν₂ umbrella vibration of the CF₃ fragment might be excited to even higher quantum states. When the photolysis photon energy was further reduced to around 304 nm^[52], many closely spaced spectral peaks were observed on the translational energy spectrum in the I* channel. Pressure experiments revealed that these peaks originated from the ν₁ symmetric stretching vibrational excitation (v₂ = 0) of CF₃ fragments produced by the photolysis of vibrationally ground state or C-I stretching vibrationally excited CF₃I (v₃ʹ = 0, 1, 2) molecules, with no clear ν₂ vibrational excitation of the CF₃ fragment observed. At this point, the highest vibrational population peak of the CF₃ fragment was at v₁ = 1. The anisotropy parameter β of the I* product angular distribution approached 2, indicating direct dissociation via the ³Q₀ potential energy surface.

For the photodissociation I channel, no significant signal was observed at 248 nm^[53]. This is mainly because the photon energy at 248 nm is far from the absorption centers of the ¹Q₁ and ³Q₁ states, and CF₃I molecules cannot absorb photons to transition to these two states; while excitation to the ³Q₀ state results in fast photodissociation, making the probability of dissociation through the ¹Q₁←³Q₀ transition very small. In the I channel photodissociation at 266 and 277 nm, experiments have measured the translational energy spectra of photofragments with vibrational state resolution^[53]. The spectra reveal that in the I channel, the bimodal vibration of the photofragment CF₃ is also simultaneously excited, with the highest vibrational peak populations being (v₁ = 5, v₂ = 0) for 266 nm photodissociation and (v₁ = 3, v₂ = 0) for 277 nm photodissociation. The ratios of internal energy to available energy E_int/E_avl are 0.202 and 0.154, respectively, and the excitation ratios of the umbrella mode vibration ∑P(v₁, v₂ = 1)/∑P(v₁, v₂) are 0.47 and 0.46, respectively. These kinetic parameter values are all higher than those in the I* channel, due to the larger available energy in the I channel. Near 304 nm, the I channel has both a direct dissociation path from the ³Q₁ state (vertical transition) and a dissociation path via the ¹Q₁←³Q₀ transition (parallel transition). Translational energy spectra with good vibrational state resolution were obtained on both dissociation paths^[52], showing strong excitation of the ν₁ symmetric stretching vibration of the CF₃ fragment, without noticeable bimodal vibration excitation of the CF₃ fragment. On these two dissociation paths, the highest vibrational state population of the CF₃ fragment is v₁ = 1; however, the ratio of internal energy to available energy E_int/E_avl for the ^{1Q₁←3Q₀ dissociation path is slightly higher than that for direct dissociation from the 3Q₁ state, consistent with CH₃I photodissociation at 304 nm[42]. For the I channel photodissociation from 266 to 304 nm, the anisotropy parameters β of the product spatial angular distribution are 1.03, 0.98, and -0.45, indicating that the proportion of products from direct dissociation of the 3Q₁ state gradually increases as the photon energy decreases. At 277 nm photodissociation, the quantum yield Φ(I*) of I* was measured to be 0.87, with branching ratios for the three dissociation paths 3Q₀, 1Q₁←3Q₀, and 3Q₁ being 0.870.0860.044, and the crossing probability P_cc for the potential energy surface 1Q₁←3Q₀ being 0.09; at 304 nm photodissociation[57], Φ(I*) = 0.06, with branching ratios for the dissociation paths 3Q₀, 1Q₁←3Q₀, and 3Q₁ being 0.060.150.79, and the crossing probability P_cc for 1Q₁←3Q₀ being 0.71.}

In the blue end of the A absorption band, there are four photodissociation pathways for CF₃I molecules, namely, the I* channel via ³Q₀ and ³Q₀←¹Q₁, and the I channel via ¹Q₁ and ¹Q₁←³Q₀. Zhu Qihao et al. conducted a photodissociation study of CF₃I at 238 nm^[54]; only in the ³Q₀ dissociation pathway was a vibrationally resolved translational energy spectrum of photofragments obtained; the signals from the other three pathways were too weak to yield spectra with high resolution. In the I* channel of the ³Q₀ dissociation pathway, bimodal vibrational excitation of the CF₃ fragment was also observed, with the highest vibrational state population being (v₁ = 6, v₂ = 0), E_int/E_avl = 0.266, and ∑P(v₁, v₂=1)/∑P(v₁, v₂) = 0.48. They measured the branching ratios of the four dissociation pathways, and the anisotropy parameter β values for the angular distributions of products in the I* and I channels were 1.70 and −0.04, respectively. The crossing probability P_cc of the potential energy surface ¹Q₁←³Q₀ was found to be 0.112, while that for ³Q₀←¹Q₁ was 0.294, showing a significant difference; this revealed that when dealing with the very important non-adiabatic interactions between potential energy surfaces at conical intersections, the direction of crossing should be considered as an important factor.

When one H in the CH₃I molecule is replaced by another halogen atom, the C_3v symmetry of the dihalo I-CH₂X molecule is disrupted, and the dissociation dynamics also show significant differences. After dissociation, the eccentric equilibrium configuration of the CH₂X fragment differs from the planar configuration of the methyl group or the trigonal pyramidal configuration of the CF₃ radical. The most easily excited vibrational mode may no longer be the umbrella vibration or symmetric stretching vibration, and under the influence of bond-breaking recoil, the rotational excitation of the fragments will also be more strongly induced. Dihalo-methanes have important implications in atmospheric chemistry and marine chemistry^[3,55], but there are relatively few studies on their gas-phase photodissociation dynamics.

Zhu Qihao et al.^[37,39] used a molecular beam rotatable laser photolysis fragment translational energy spectrometer to conduct a study on the photodissociation of I-CH₂Cl molecules at 248 nm, observing the C—I bond breaking process. The quantum yield Φ(I*) = 0.70 for the I* channel was measured, and the ratio of internal energy to available energy E_int/E_avl for the photofragments in both the I* and I channels was approximately 0.69. However, due to resolution limitations, it was not possible to obtain vibrationally resolved photofragment translational energy spectra, making it impossible to identify the vibrational excitation modes of the CH₂Cl fragments.

Using a high-resolution miniature laser photofragment translational spectroscopy, they conducted the photodissociation study of I-CH₂Cl molecules at 277 and 304 nm^[56], obtaining photofragment translational energy spectra with clear vibrational state resolution (Figure 9). Through the analysis of well-resolved I* channel spectra, it was identified that the maximum vibrational populations of the CH₂Cl fragment during photodissociation at 277 and 304 nm were v = 3 and 2, respectively, and it was calculated that the products had very high rotational excitation (average E_R/E_T ≈ 0.72, with rotational excited states reaching up to J = 95 and 79), and the energy distribution ratio among translational, rotational, and vibrational energies was about 0.48:0.34:0.18. By the energy spacing of vibrational peaks on the spectrum and considering different rotational excitations, it was confirmed that the C—Cl stretching vibration of the CH₂Cl fragment was the most likely vibrational mode to be excited during the photodissociation process. In the I channel, the ratio of product rotational to translational energy E_R/E_T was slightly less than in the I* channel, with the energy distribution ratio among translational, rotational, and vibrational energies being approximately 0.45:0.31:0.24. A large number of vibrational peaks appeared on the translational energy spectrum, attributed to the excitation of C—Cl stretching vibrations from the CH₂Cl fragment. Unlike in the I* channel, the vibrational peaks on the I channel's translational energy spectrum clearly showed components from two dissociation pathways (Figure 9), and their contributions changed with different photon energies. The component with smaller translational energy (left side of the spectrum) was assigned to the I channel produced by direct dissociation after transitioning to the 3A' state upon absorbing a photon, while the component with larger translational energy (right side of the spectrum) was assigned to the I channel produced after transitioning to the 4A' state and then dissociating through the conical intersection 5A'$\leftarrow$4A'. When the photon energy decreased from 277 to 304 nm, more I-CH₂Cl molecules absorbed photons and transitioned to the lower-energy 3A' state, leading to a higher population of the component with smaller translational energy on the translational energy spectrum. It was found that the anisotropy parameter β values for both I* and I channel fragment spatial angular distributions were close to 2, indicating that both reaction channels were mainly activated via parallel transitions. The quantum yields Φ(I*) for the I* channel during photodissociation at 277 and 304 nm were measured to be 0.36 and 0.26, respectively. As the photon energy for photodissociation decreased, the speed at which excited-state molecules passed through the potential crossing region decreased, and the time increased, resulting in more excited-state molecules crossing from 4A' to 5A' to produce ground-state I, and simultaneously, more molecules being excited to the 3A' state, which directly produces the ground-state I; all these factors would lead to a decrease in Ф(I^*). Compared to the photodissociation of CX₃I, the photodissociation of dihalomethanes shows significant differences, and the current research has not systematically revealed its photodissociation dynamics mechanism, requiring further in-depth exploration combining experiments and theory.