J. Phys. Chem. A 2010, 114, 1783–1789
1783
Low-Energy Reaction Rate Constants for the Ni+-Assisted Decomposition of Acetaldehyde: Observation of C-H and C-C Activation S. Jason Dee, Vanessa A. Castleberry, Otsmar J. Villarroel, Ivanna E. Laboren, and Darrin J. Bellert* Department of Chemistry and Biochemistry, Baylor UniVersity, One Bear Place 97348, Waco, Texas 76798-7348 ReceiVed: October 30, 2009; ReVised Manuscript ReceiVed: December 11, 2009
Rate constants for the low-energy Ni+-assisted dissociative reaction of acetaldehyde have been measured under jet-cooled conditions in the gas phase. The rate constants are acquired through monitoring the time dependence of fragment Ni+CO formation. The decomposition of the precursor Ni+-acetaldehyde cluster ion proceeds via consecutive, parallel reaction coordinates that originate with the Ni+-assisted cleavage of either a C-C or an aldehyde C-H bond. The energies used to initiate these reactions are well below that required to cleave σ-bonds in the isolated acetaldehyde molecule. Direct measurement of the reaction kinetics over a range of energies indicates that the rate-limiting step in the dissociative mechanism changes at cluster ion internal energies ) 17 200 ( 400 cm-1. Arguments are presented that this energy marks the closure of the dissociative coordinate that initiates with C-H σ-bond activation and thus provides a measure of the activation energy of this dissociative pathway. Introduction The use of transition metals in catalyzing organic bond fragmentation reactions in the gas phase is well-documented within the research community.1-12 A fundamental objective in researching such systems is to appreciate factors involved in facilitating bond activation.13-25 In particular, transition metal ions have been found, both experimentally26-31 and theoretically,32-35 to cleave C-H and/or C-C bonds within ketones and aldehydes. We have recently been successful at measuring the kinetics associated with oxidative addition of bare Ni+ cations into C-C σ-bonds.36-38 These reactions occur in the gas phase and with internal energies well below that required to cleave C-C bonds in an isolated organic molecule.39 The systems we have studied thus far have involved Ni+-assisted dissociation of organic ketones. We have found that reaction products always consist of a stable, neutral organic molecule (ethane, ethene, etc.) and the corresponding ionic fragment (which is sampled by this technique). These studies have established the rate-limiting step along the reaction coordinate that connects reactants to products. Our measurements often suggest a likely mechanism for these simple reactions. Gasphase studies are particularly amenable to measuring the mechanistic details of the reaction coordinate as the reaction environment and energy content are controlled. Our studies serve as models for the catalytic production of gaseous combustibles from organic liquids. Furthermore, the sizes of our molecular complexes lend themselves to high-level computational studies and our results should provide benchmarks to the theoretical research community. This study represents our first kinetic measurement of oxidative addition to a C-H bond. Here we present the Ni+assisted decomposition of acetaldehyde; this reaction consists of two parallel paths (Ni+ oxidative addition across either the C-C or the aldehyde C-H bond), and the observed fragmenta* Corresponding author: fax 254-710-4272; e-mail Darrin_Bellert@ baylor.edu.
tion pathway depends upon the supplied energy. Our group is not the first to study this gas-phase reaction. In fact, early collision-induced dissociation (CID) experiments have confirmed that the only low-energy reaction channel involved loss of CH4 gas.30 However, these experiments were unable to ascertain whether the addition reaction occurred at the C-C or the aldehyde C-H bond. Additionally, recent theoretical investigations into this reaction coordinate have been accomplished and our results are compared to this density functional theory (DFT) calculation.35 Experimental Section The instrument, and the method by which the data is analyzed, has been described previously;36,37 therefore, only a brief description is provided here. In general, a supersonic source chamber is connected to a custom time-of-flight mass spectrometer (TOF-MS). Clusters generated in the source are massanalyzed in the TOF-MS. Source conditions are optimized to form the coldest ionic clusters possible. This is done by minimizing the distribution of precursor ion velocities in the beam. Under such cold conditions, Mach numbers as high as 70 are fairly easily realized for small ionic clusters. Additionally, under optimal conditions, substantial clustering of the acetaldehyde monomer units onto a single Ni+ cation further illustrates the effective cooling of the supersonic expansion. Figure 1 shows such a mass spectrum where up to four acetaldehyde molecules clustered onto a Ni+ cation are clearly visible. Specifically, the molecular beam is generated through pulsedlaser (248 nm) vaporization of a relatively pure (99%), rotating nickel rod under high vacuum conditions. High-pressure, pulsed helium gas (doped with the vapor pressure of pure acetaldehyde) is timed to entrain the Ni+ cations into the doped plume. The vaporization products are then cooled through supersonic expansion into vacuum. Collisions between the acetaldehyde vapor and Ni+ cations can form the singly complexed cluster ion, Ni+(acetaldehyde) [Ni+(Aald)], under jet-cooled conditions.
10.1021/jp910396t 2010 American Chemical Society Published on Web 01/05/2010
1784
J. Phys. Chem. A, Vol. 114, No. 4, 2010
Dee et al. TABLE 1: Rate Constants (×104 s-1) Measured for Ni+-Assisted Dissociation of Organic Molecules at Low Internal Energies internal energy (cm-1) 18 200 18 000 17 700 16 800 16 700 16 400 16 100 15 600
Figure 1. Ni+(Aald) precursor ion time-of-flight mass spectrum.
The supersonic expansion is skimmed twice and the beam approaches a Wiley-McLaren40-type orthogonal accelerator (OA). Along this approach, pulsed laser radiation counterpropagates to the molecular beam direction and is timed to intersect the clusters before pulsed extraction in the OA. Those ions that absorb the radiation and dissociate before entering the OA are accelerated along with all the other ions. These early fragment ions cannot be selectively transmitted through a hemispherical kinetic energy analyzer (sector), as they will receive the full kinetic energy supplied by the OA voltage. Thus, early dissociation events are not sampled. The precursor ions that dissociate following orthogonal extraction, but before entering the sector, will yield fragment ions that are selectively transmitted to a microchannel plate detector located at the terminus of the sector. Monitoring only those fragment ions from specific precursor ions dissociating between the OA and sector provides the signals acquired in this study. Although the dissociative event does not significantly impact the ion’s velocity, the mass change alters the kinetic energy of the dissociated charged fragment. The optimal fragment transmission voltage across the halves of the sector is determined through application of eq 1:
mf V ) Vf mp p
(1)
where mf and mp are the masses of the fragment ion and precursor complex, respectively. Vp and Vf are the optimal potential differences applied to the sector to allow for transmission of the precursor complex and the specific dissociative fragment. Data Analysis. The unimolecular decay of precursor ions is sampled directly by this technique. We monitor the temporal response of product ion formation following the reaction initiated by photon absorption. Zero time is defined when the complex ions are coincident with the laser field within the OA. Thus, at zero time, the complex ions absorb the laser radiation and are immediately pulse-accelerated into field free flight zone. Fragment production is sampled from the time the precursor ions exit the OA to the time just before they enter into the field of the sector. Negative values of time indicate the temporal displacement of the precursor ions from the OA following photon absorption. Experimentally, the triggering of the pump laser is delayed from zero time; thus dye laser radiation intersects the molecular beam at continually greater distances from the OA. This allows time for the precursor ions to dissociate into undetectable fragments before pulsed orthogonal extraction. We therefore observe the exponential growth of fragment ions that initiate at negative times and terminate at time zero. Fragments produced at positive
a
Ni+(Ac-h6)36
Ni+(Ac-d6)37
kisoCH3
kisoCH3
kisoCH3
9.68 ( 0.03 9.27 ( 0.03
1.60 ( 0.08
11.9 ( 0.2 10.9 ( 0.4 9.8 ( 0.4
6.26 ( 0.02 5.90 ( 0.02 5.80 ( 0.03 5.50 ( 0.03
1.17 ( 0.06
Ni+(Aald)a kactC-H + kactC-C
15.5 ( 0.4 14.0 ( 0.5 13.5 ( 0.5 13.0 ( 0.5
This work.
values of time result from collision-induced dissociation of the precursor ions. Sampling fragment ion production throughout the entire length of the TOF-MS effectively integrates the observed signals over the 38-µs field free flight. This significantly enhances the acquired product ion intensities. Ion Internal Energy. The title complex ion is formed under jet-cooled conditions. The internal energy is minimized by monitoring the complex ions’ velocity distribution within the beam and optimizing source conditions by narrowing this distribution. Under these cold conditions, cluster ions built from multiple monomer units are observed. Energy is deposited into the cold cluster ion through absorption of laser photon energy. Upon absorption, the cluster ion is promoted to an electronic excited state. There are no low-lying electronic states in the acetaldehyde molecule; therefore, the Ni+ cation is the chromophore in the electronic transition. The lowest-lying, excited electronic state of Ni+ is 4F(3d8 4s) with lowest energy spin orbit component (J ) 9/2) lying 8393.9 cm-1 above the Ni+ 2D ground state. Electronic transition into this 4F manifold of states initiates the dissociative chemical reactions observed in this study. This Ni+-centered electronic transition {4F(3d8 4s) r 2 D(3d9)} is both spin- and parity-forbidden. The prepared, excited quartet electronic state of Ni+(Aald) is metastable; hence, the absorbed photon energy is insufficient to cause direct dissociation into Ni+ and acetaldehyde fragments as the energy of the prepared state is below the adiabatic bond energy of the complex. Moreover, coupling to the ground state through photon emission is optically forbidden. Rather, the excited quartet state intersystem-crosses to the doublet electronic state and deposits the energy of the electronic transition (or the photon energy) into the high vibrational levels of the ground state. Energy deposition in these high vibrational states provides the activation energy for the unimolecular dissociation reaction. Results Rate constants for the unimolecular decomposition reaction, Ni+(Aald) f Ni+CO + CH4, have been measured over a range of internal energies (Table 1). The energies used to initiate the dissociative reaction are well below the energy required to cleave the C-C or C-H σ-bond of isolated acetaldehyde. The Ni+ cation thus assists in the cleavage reaction by lowering the energy required to activate the σ-bond. Figure 2 shows fragment production waveforms measured for the Ni+-assisted decomposition of acetaldehyde into Ni+CO and CH4. The reaction in each panel is initiated through absorption of a different energy photon (A, 18 200 cm-1; B, 16 800 cm-1). These waveforms are acquired by scanning the time delay between the reaction initiation laser firing and the orthogonal accelerator pulse.
Ni+-Assisted Decomposition of Acetaldehyde
Figure 2. Fragment production waveforms measured for the decomposition of Ni+(Aald) at two different precursor ion internal energies. The rate constants shown in each panel indicate the laser wavelength that supplied the activation energy for the unimolecular decomposition reaction. The rate constants are extracted from a linear analysis of the waveform intensity and are a measurement of the rate-limiting step in the dissociative mechanism.
Plotting the natural logarithm of the intensity versus time yields values for the rate constants observed in the figure. Additionally, the fit parameters are used to construct the solid curve drawn through the data points in each panel. The majority of the waveform in panel A is well-described by a single-exponential growth curve, while the entirety of the waveform in panel B is well-described as a single exponential. In panel A, the data at time values from 0 to 6 µs deviates from this single-exponential description, which indicates a more complicated, energydependent fragmentation pathway. The waveforms observed in Figure 2 represent a type of limiting behavior observed for the Ni+-assisted dissociation of acetaldehyde. All fragment production waveforms acquired for the unimolecular decomposition of Ni+(Aald) at energies less than 16 800 cm-1 have the same single-exponential form. Measurements made at energies greater than 17 700 cm-1 possess the same biexponential behavior as observed in Figure 2A. The change that occurs between biexponential and singleexponential growth is subtle, and waveforms measured at reaction initiation energies between these two limits express characteristics of both waveforms. Therefore, useful rate information could not be extracted from studies between these two limits. The rate constants provided in Figure 2 are for the ratelimiting step thath connects reactants to products in the Ni+assisted dissociative reaction mechanism. Since the waveform appears different at the two reaction initiation energies used in Figure 2, this suggests that the rate-limiting step is different at these two energies. Comparisons between the magnitudes of the measured rate constants provide additional support for this claim. From Table 1, the observed rate constants for the Ni+assisted dissociation of acetone decrease systematically with internal energy as expected. However, there is an obvious discontinuity in the measured rate-limiting rate constants that govern the decomposition of Ni+(Aald) and occurs at energies ∼17 200 ( 400 cm-1. At energies greater than this limit, the measured rate constants are smaller than the rate constants measured at energies below this value. In order to understand the dynamics associated with the Ni+assisted dissociation of acetaldehyde, comparisons are drawn from the well-characterized, simpler Ni+(acetone) [Ni+(Ac)] system.36 These comparisons are made at comparable initiation energies and are shown in Figures 3 and 4. The Ni+-assisted dissociation of acetone follows consecutive steps in a reaction mechanism. The reaction commences with C-C bond activation
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1785
Figure 3. Comparison of the fragment production waveform measured for the Ni+-assisted dissociation of acetaldehyde (top panel) and acetone (bottom panel) at comparable photon energies (∼18 000 cm-1). The solid curve through the data points is the simulated contour from the rate constants indicated. The value k′ is the sum of the C-C and C-H activation rate constants: k′ ) kactC-C + kactC-H.
Figure 4. Same type of comparison as given in Figure 3, except the internal energy of the precursor ion is ∼16 800 cm-1.
and CH3 isomerization follows, forming products C2H6 and Ni+CO. Although the results of deuterium37-labeled experiments indicate that methyl isomerization is the rate-limiting step along the reaction coordinate, we find that both bond activation and isomerization are kinetically important. It is the similarity between these two rate constants that results in the unique shape of the waveforms in the lower panels of Figures 3 and 4. The plateau area results from delayed product (Ni+CO) formation due to an induction period (τind). Correlating this induction period to the maximum in intermediate concentration allows adequate simulation of the experimentally observed waveforms. The solid curves in Figures 3 and 4 were simulated from the results of a differential analysis of the rate equations that govern product formation, with the introduction of an induction period that delays this product formation until an intermediate has reached maximum concentration. The simulation and how these conclusions were reached is described in greater detail below. The waveforms representing Ni+CO growth from Ni+-assisted dissociation of acetone and acetaldehyde (Figure 3) are similar at a reaction initiation energy ∼18 000 cm-1. Again, the rate constants indicated in the figure are measured directly from linear fits of the waveform intensity. First-order, exponential growth of the product yields similar rate-limiting rate constants for the different systems. Additionally, the induction period in the formation of fragment Ni+CO within the Ni+(Ac) system nearly correlates with the biexponential behavior observed in the Ni+(Aald) system (top panel). Clearly, at this reaction initiation energy, similar dissociation dynamics dominate each reaction coordinate. It is known that methyl isomerization rate limits the production of fragment Ni+CO within the Ni+(Ac) system.36,37 We therefore assume, at internal energies ∼18 000
1786
J. Phys. Chem. A, Vol. 114, No. 4, 2010
SCHEME 1
cm-1, that CH3 isomerization represents the rate-limiting step in the reaction coordinate that connects Ni+(Aald) to Ni+CO + CH4. Figure 4 shows the fragment growth waveforms acquired from the Ni+-assisted dissociation of acetone (bottom panel) and acetaldehyde (top panel) measured at photon energy ∼16 750 cm-1. Here, comparison indicates that Ni+(Aald) no longer follows reaction dynamics comparable to those observed in the decomposition of Ni+(Ac). Within the Ni+(Ac) system, both the rate-limiting methyl isomerization rate constant and the C-C bond activation rate constant have decreased with decreasing internal energy. This is evident by the long exponential growth profile as well as the increased induction time (which is related to the C-C bond activation rate constant). This is also confirmed through regression analysis and simulation (solid curve). The product growth waveform acquired through monitoring the decomposition of Ni+(Aald) f Ni+CO at 16 800 cm-1 is described by a single exponential with a rate constant larger than that observed at higher energies. There is no obvious biexponential behavior and the waveform bears no resemblance to the Ni+(Ac) waveform in the bottom panel. The results indicate that the Ni+(Aald) decay mechanism exhibits two paths to dissociation. At higher energies, it is likely that both paths are available; however, the measurement is dominated by a rate-limiting methyl isomerization. At lower energies, it appears that the rate-limiting step has changed and we conclude that only a single path is followed to dissociation at lower energies. Thus, one of the dissociative channels available to the decomposition of Ni+(Aald) closes between 16 800 and 17 700 cm-1. Table 1 provides a summary of the rate constants measured in this study. Discussion Scheme 1 presents a mechanism detailing the Ni+-assisted decomposition of acetaldehyde. The reaction progresses through parallel consecutive steps as reactants are converted to products. Species A is the encounter complex that is formed cold within the supersonic expansion. The first intermediates, B and C, are accessed through rate constants k1 and k2 and represent the Ni+-
Dee et al. inserted species. Final products are Ni+CO (species D) and CH4, where formation of D is monitored over time. The rate constant that connects B to D is k3, the H-isomerization rate constant. The intermediate C is connected to D through the CH3isomerization rate constant, k4. DFT calculations have verified that the reaction coordinate progresses through intermediates B and C and ultimately concludes with formation of Ni+CO and CH4.35 The comparisons with Ni+(Ac) unimolecular decomposition (where deuterium labeling experiments have verified methyl migration as the rate-limiting step in the dissociative mechanism) at internal energies greater than 17 700 cm-1 suggest that methyl isomerization is also the rate-limiting step in the assisted dissociation of acetaldehyde. Thus k4 is the smallest rate constant in the mechanism of Scheme 1 at energies greater than 17 700 cm-1. Additionally, our results indicate that this path to dissociation closes at energies below 16 800 cm-1. Therefore, the dissociative reaction coordinate that initiates with aldehyde C-H bond cleavage has an activation energy of 17 200 ( 400 cm-1. To support this conclusion, a single model is developed that simulates the waveform shapes acquired in this study. At energies greater than 17 700 cm-1, the observed waveforms can be effectively simulated by equating k4 with the rate constant determined experimentally. At energies below 16 800 cm-1, the waveforms are simulated by setting the value of k4 to zero, effectively preventing dissociation from intermediate C (Scheme 1). Finally, we provide additional support for this conclusion through comparisons with a recent DFT calculation of the Ni+assisted dissociative reaction coordinate of acetaldehyde.35 The solutions to the differential rate equations that apply to Scheme 1 result in the following time-dependent concentration terms:
At ) A0e-k′t Bt )
Ct )
(2)
k1A0 (e-k′t - e-k3t) (k3 - k′)
(3)
k2A0 -k′t (e - e-k4t) k4 - k′
(4)
Dt ) A0 - At - Bt - Ct
(5)
where A0 equals the initial concentration of the photoexcited precursor complex ion and k′ ) (k1 + k2). Equations 2-5 provide a model that describes depletion of the cold precursor complex (At), the build-up and subsequent depletion of the Ni+ C-C and C-H inserted intermediates (Bt and Ct), and final production of product (Dt). This model provides the basis of a computer program that is used to simulate the acquired waveforms. The waveform generated from fragment production (eq 5) is integrated under instrumental conditions to construct the simulated waveform, which is superimposed upon observation. The program consists of five input parameters, A0 and the four different rate constants. Given the complexity of the rate equations describing parallel consecutive reactions, this model is first applied to the Ni+-assisted dissociation of acetone, a system that follows a simple consecutive mechanism of two elementary steps connected through a single intermediate. Thus, to apply the model to the decomposition of Ni+(Ac), one of the two parallel paths must be eliminated. This is accomplished
Ni+-Assisted Decomposition of Acetaldehyde
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1787
programmatically by setting the value of k1 to zero. This effectively reduces the rate equations (eqs 2-5) to those that describe the concentration dependencies in a simple, two-step consecutive mechanism. Application of the program to fragment production from the decomposition of Ni+(Ac) necessitates input of only three parameters (k2, k4, and A0). The value of one of the rate constants is set to that determined experimentally through linear regression analysis. The remaining two parameters are systematically adjusted until the simulation is visually superimposed upon the observed waveform. However, there are no combinations of A0 and the remaining rate constant that simulate the plateau region of the waveforms observed in the lower panels of Figures 3 and 4. These regions are not artifacts of the experimental procedure. The temporal profiles of these features predictably increase with decreasing laser photon energy (compare the lower panels of Figures 3 and 4) and additionally exhibit dependence upon isotopic substitution. Therefore, these intense features must be related to the dynamics of the dissociation process. We attribute these features to a delay in product formation caused by a slow buildup of an intermediate; thus these induction periods are related to the “unobserved” rate constant in the unimolecular decomposition of the cluster ion. The slow buildup of the intermediate implies that both elementary steps (σ-bond activation as well as CH3 isomerization) are kinetically important in the dissociative process. To account for this, fragment production is delayed from time zero (the time that the precursor ions absorb the laser radiation) by the time required to produce the maximum concentration of the intermediate. The induction period is therefore determined by solving the time derivative of eq 4 set equal to zero. The resulting induction time is given by eq 6:
tind )
ln (k′/k4) (k′ - k4)
(6)
which is incorporated into the program. It should be noted that this does not add any additional adjustable parameters to the software. Application of the improved model to the Ni+-assisted dissociation of acetone results in the solid curves through the data points in the lower panels of Figures 3 and 4. Again, the model requires input of the rate-limiting rate constant determined experimentally (here k4, which is the CH3 isomerization rate constant) and the remaining rate constant, k2, and A0 are adjusted until the simulation is superimposed upon the observation. Modification of k2 results in changes to the induction time and the optimized value provides an estimate of the C-C bond activation rate constant in the Ni+-induced dissociation of acetone. These estimates are provided in the lower panels of Figures 3 and 4. It is noteworthy to realize that incorporation of the induction time in such fashion accounts for the experimentally observed phenomena: the value of the methyl isomerization rate constant is less than the C-C bond activation rate constant, as expected from the results of deuterium-labeling experiments;37 the induction period, which increases with decreasing reaction energy, is due to the decreasing activation rate constant; and the activation rate constant should be affected by isotopic substitution. As described, two limiting behaviors have been observed in the Ni+-assisted dissociation of acetaldehyde. The waveforms acquired at energies greater than 17 700 cm-1 have shown biexponential behavior, while those measured below 16 800
Figure 5. (Bottom panel) The dashed curve indicates rapid precursor depletion and the gradual building and depletion of a long-lived intermediate. The solid curve shows fragment ion buildup, a portion of which is delayed until the intermediate has reached the maximum concentration. (Top panel) The solid curve is the integration of fragment production given in the lower panel and is superimposed upon the observed waveform (symbols).
cm-1 appear single-exponential. A purpose of this discussion is to demonstrate that the kinetic model proposed can describe both limiting behaviors. Applying the model to the fragment production waveforms observed in the top panels of Figures 3 and 4 relieves the restriction that k1 ) 0, and now the formation of two intermediates is possible. On the basis of the experience gained while studying the Ni+-assisted dissociation dynamics of acetone36 and deuterium-labeled acetone,37 it is assumed that the rate constants can be ordered as k3 (H-isomerization) > k′ () k1 + k2, C-C activation + C-H activation) > k4 (CH3 isomerization). This assumption is based solely on the kinetic isotope effect measured for the Ni+(Ac) system.37 Furthermore, an induction period (analogous to that described above) may be present if two kinetically important, consecutive steps occur along a dissociative coordinate. Therefore, methyl isomerization following C-H activation may result in the delayed production of Ni+CO since both bond activation and alkyl migration are likely slow steps. This has the effect of delaying product formation from one leg of the mechanism presented in Scheme 1. Application of the program to the fragment production waveforms resulting from the unimolecular dissociation of Ni+(Aald) results in the solid curves in the top panels of Figures 3 and 4. To simulate the waveform in the top panel of Figure 3, the methyl isomerization rate constant (k4) is fixed at the value determined through linear regression analysis [k(E ) 18 200 cm-1) ) 11.9 × 104 s-1]. The value of k3 (H-isomerization) is assumed to be large and kinetically unimportant. The values of k′ and A0 are varied until the simulated waveform is superimposed upon the observation. In similar fashion to the Ni+(Ac) studies, it is essential to include an induction time (determined through eq 6) in order to simulate the biexponential behavior of the waveform. The optimal value of k′ is 41 × 104 s-1 (as indicated in Figure 3). Although the values of k1 and k2 can be varied (under the constraint that their sum ) 41 × 104 s-1), the simulation is optimized when k1 ∼ k2. The bottom panel of Figure 5 shows the contribution to the formation of fragment Ni+CO (solid curve) from each dissociative path of Scheme 1 at a reaction energy ) 18 200 cm-1. The dashed curves indicate precursor depletion as well as the build-up and subsequent depletion of the long-lived C-H activated intermediate. The formation of product following methyl isomerization is delayed until the intermediate has reached maximum concentration values. The simulated contour in the top panel of Figure 5 results from integration of the solid curve in the lower panel under
1788
J. Phys. Chem. A, Vol. 114, No. 4, 2010
instrumental conditions. The induction time, which delays production of Ni+CO, results in the biexponential character of the measured waveform. Application of the program to simulate the fragment production waveform in the top panel of Figure 4 is achieved by setting the value of k4 to zero and the value of k′ to that acquired through linear analysis of the intensity of the experimental waveform (k′ ) 15.5 × 104 s-1). Setting the value k4 to zero indicates that the energy supplied is insufficient for methyl isomerization. This effectively eliminates any biexponential character in the simulated waveform through elimination of the induction period (eq 6) and removes any fragment production resulting from this dissociative coordinate. Again, all waveforms acquired at photon energies less than 16 800 cm-1 exhibit the same singleexponential behavior. We conclude that the value of k′, which accounts for the observed fragment buildup during the final 5 µs (t ) 0 to -5 µs) in the top panel of Figure 3, is responsible for the entire waveform in the top panel of Figure 4. The value of k′ has decreased from ∼40 × 104 s-1 to 15.5 × 104 s-1 as a result of a reduction in the internal energy of the precursor complex ion by 1400 cm-1. Finally, to further our conclusion, we compare our results to recent DFT calculations of the Ni+ + acetaldehyde reaction coordinate.35 The results of these calculations indicate that the geometry of the encounter complex locates the Ni+ cation as either cis or trans to the methyl group, and these isomers are separated by only 280 cm-1. Two reaction coordinates have been calculated: one that initiates with C-C bond activation and the other with C-H activation. The authors indicate that the energy barrier for C-H activation is 11.2 kcal/mol (∼3900 cm-1) less than that for C-C bond activation. The activated intermediates progress to products via an aldehyde H-shift or CH3-shift isomerization prior to dissociation. The transition-state energies for each isomerization represent maxima along each calculated reaction coordinate. For the C-H activation/methyl isomerization reaction coordinate, the DFT calculated activation energy is 19 700 cm-1 above the lowest energy encounter complex isomer. The C-C activation/hydrogen isomerization reaction coordinate represents a lower energy path with activation energy calculated at 18 600 cm-1. Our experimental results are in qualitative agreement with the theoretical predictions. We find that the energy ordering of the reaction coordinates is consistent with measurement. The C-H activation/methyl isomerization is the most energetically demanding reaction coordinate; however, it lies 17 200 ( 400 cm-1 above the ground state (∼2500 cm-1 less than the theoretical prediction). Additionally, Ni+CO products are detected at energies as low as 15 600 cm-1 (Table 1). This value thus provides an upper limit to the energy of activation for the C-C activation/hydrogen isomerization reaction coordinate. This upper limit is ∼3000 cm-1 less than that predicted by DFT theory.35 Conclusions Rate constants for the low-energy, Ni+-assisted dissociation of acetaldehyde have been measured for the first time. A discontinuity in the measured rate-limiting rate constant is observed by monitoring the reaction kinetics at continually smaller ion internal energies. This discontinuity occurs at 17 200 ( 400 cm-1 and marks the closure of the dissociative coordinate that initiates with C-H σ-bond activation. Therefore, the sum of the energy requirements for the Ni+-assisted C-H bond
Dee et al. activation and methyl isomerization is 17 200 ( 400 cm-1. The remaining dissociative pathway, Ni+ activation of the C-C σ-bond followed by H-migration to form Ni+CO + CH4, remains open at energies as low as 15 600 cm-1. These results qualitatively agree with the energetic ordering of each dissociative coordinate as recently calculated.35 However, it appears that DFT overestimates the activation energy barrier along each reaction coordinate. Acknowledgment. We gratefully acknowledge research support from the ACS Petroleum Research Fund (44393-G6). Additionally, funds from the Baylor University Research Committee and the Vice Provost for Research supported this study. References and Notes (1) Allison, J. In Progress in Inorganic Chemistry; Lippard, S. J., Ed.; Wiley: New York, 1986; Vol. 34y. (2) Buckner, S. W.; Freiser, B. S. In Gas Phase Inorganic Chemistry; Russell, D. H., Ed.; Plenum Press: New York, 1989. (3) Beauchamp, J. L.; Van Koppen, P. A. M. NATO ASI Ser., Ser. C. 1992, 367, 287. (4) Hettich, R. L.; Freiser, B. S. Gas-Phase Photodissociation of Transition Metal Ion Complexes and Clusters. In Fourier Transform Mass Spectrometry; Buchanan, M. V., Ed.; ACS Symposium Series 359; American Chemical Society: Washington, DC, 1987. (5) Jacobson, D. B.; Freiser, B. S. J. Am. Chem. Soc. 1983, 105, 5197. (6) Burnier, R. C.; Byrd, G. D.; Freiser, B. S. Anal. Chem. 1980, 52, 1641. (7) Schwarz, H. Int. J. Mass Spectrom. 2004, 237, 75. (8) Dunbar, R. C. In Gas Phase Inorganic Chemistry; Russell, D. H., Ed.; Plenum Press: New York, 1989. (9) Weisshaar, J. C. In Gas-Phase Metal Reactions; Fontijn, A., Ed.; Elsevier: Amsterdam, 1992. (10) Operti, L.; Rabezzana, R. Mass Spectrom. ReV. 2006, 25, 483. (11) Freiser, B. S. J. Mass Spectrom. 1996, 31, 703. (12) Arsitov, N.; Armentrout, P. B. J. Am. Chem. Soc. 1986, 108, 1806. (13) Holthausen, M. C.; Fiedler, A.; Schwarz, H. J. Phys. Chem. 1996, 100, 6236. (14) Tolbert, M. A.; Beauchamp, J. L. J. Phys. Chem. 1986, 90, 5015. (15) Schulze, C.; Weiske, T.; Schwarz, H. Organometallics 1988, 7, 898. (16) Mo, O.; Yanez, M.; Salpin, J.-Y.; Tortajada, J. Mass Spectrom. ReV. 2007, 26, 474. (17) Sablier, M.; Mestdagh, L.; Poisson, L.; Leymarie, N.; Rolando, C. J. Am. Soc. Mass Spectrom. 1997, 8, 587. (18) Haynes, C. L.; Chen, Y. M.; Armentrout, P. B. J. Phys. Chem. 1995, 88, 9910. (19) Liu, F.; Zhang, X. G.; Armentrout, P. B. Phys. Chem. Chem. Phys. 2005, 7, 1054. (20) Halle, L. F.; Armentrout, P. B.; Beauchamp, J. L. Organometallics 1984, 3, 1694. (21) Hanratty, M. A.; Beauchamp, J. L.; Illies, A. J.; van Koppen, P. A. M.; Bowers, M. T. J. Am. Chem. Soc. 1988, 110, 1. (22) Van Koppen, P. A. M.; Bowers, M. T.; Fisher, E. R.; Armentrout, P. B. J. Am. Chem. Soc. 1994, 116, 3780. (23) Van Koppen, P. A. M.; Brodbelt Lustig, J.; Bowers, M. T.; Dearden, D. V.; Beauchamp, J. L.; Fisher, E. R.; Armentrout, P. B. J. Am. Chem. Soc. 1990, 112, 5663. (24) Van Koppen, P. A. M.; Brodbelt Lustig, J.; Bowers, M. T.; Dearden, D. V.; Beauchamp, J. L.; Fisher, E. R.; Armentrout, P. B. J. Am. Chem. Soc. 1991, 113, 2359. (25) Allison, J.; Freas, R. B.; Ridge, D. P. J. Am. Chem. Soc. 1979, 101, 1332. (26) Burnier, R. C.; Byrd, G. D.; Freiser, B. S. J. Am. Chem. Soc. 1981, 103, 4360. (27) Surjasasmita, P. I.; Freiser, B. S. J. Am. Soc. Mass Spectrom. 1993, 4, 135. (28) Carpenter, C. J.; van Koppen, P. A. M.; Bowers, M. T. J. Am. Chem. Soc. 1995, 117, 10976. (29) Yi, S. S.; Reichart, E. L.; Weisshaar, J. C. Int. J. Mass Spectrom. 1999, 185-187, 837. (30) Halle, L. F.; Crowe, W. E.; Armentrout, P. B.; Beauchamp, J. L. Organometallics 1984, 3, 1694. (31) Schroder, D.; Schwarz, H. J. Am. Chem. Soc. 1990, 112, 5947.
Ni+-Assisted Decomposition of Acetaldehyde (32) Zhao, L.; Zhang, R.; Guo, W.; Lu, X. Chem. Phys. Lett. 2006, 431, 56. (33) Chen, X.; Guo, W.; Zhao, L.; Fu, Q. Chem. Phys. Lett. 2006, 432, 27. (34) Zhao, L.; Guo, W.; Zhang, R.; Wu, S.; Lu, X. Chem. Phys. Chem. 2006, 7, 1345. (35) Chen, X.; Guo, W.; Zhao, L.; Fu, Q.; Ma, Y. J. Phys. Chem. A 2007, 111, 3566. (36) Castleberry, V. A.; Dee, S. J.; Villarroel, O. J.; Laboren, I. E.; Frey, S. E.; Bellert, D. J. J. Phys. Chem. A 2009, 113, 10417.
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1789 (37) Dee, S. J.; Castleberry, V. A.; Villarroel, O. J.; Laboren, I. E.; Frey, S. E.; Ashley, D.; Bellert, D. J. J. Phys. Chem. A [Online early access] 2009. (38) Laboren, I. E.; Villarroel, O. J.; Dee, S. J.; Bellert, D. J. J. Phys. Chem. A 2009 (submitted for publication). (39) Handbook of Chemistry and Physics, 85th ed.; CRC Press: Boca Raton, FL, 2004-2005. (40) Wiley, W. C.; McLaren, I. H. ReV. Sci. Instrum. 1955, 26, 1150.
JP910396T
