Using FT-IR spectroscopy to model reaction kinetics of a solvent-free Wittig Reaction
Briggs, Donovan R.
The rate of the solvent-free Wittig reaction of 4-bromobenzaldehyde with (carbethoxymethylene)triphenylphosphorane to form (E)3-(4-bromohpenyl)acrylate is determined by fourier transform infrare3d spectroscopy. The rate of product formation is monitored by tracking the increase of the absorbance intensity at 1705 cm(-1) and the rate of reactant decay is monitored by tracking the decline of the absorbance intensity at 1604 cm(-1). The reaction is peformed under nitrogen to prevent ambient water vapor from being absorbed by the reaction mixture. The rate is fit to models of solid-solid systems in order to determine if the reaction is a solid-solid reaction or merely a solvent-free reaction.
Reaction Kinetics of a Solvent Free Wittig Reaction using FT-IR Spectroscopy Donovan Briggs and Stephen A. Angel, PhD, Department of Chemistry, Washburn University, Topeka, KS, 66621 Abstract The rate of the solvent-free Wittig reaction of 4-bromobenzaldehyde with (carbethoxymethylene)triphenylphosphorane to form (E)-ethyl 3-(4- bromophenyl) acrylate is determined by Fourier Transform Infrared Spectroscopy. The rate of product formation is monitored by tracking the increase of the absorbance intensity at 1164 cm-1 and 1705 cm-1. The rate of reactant decay is monitored by tracking the decline of the absorbance intensity at 1604 cm-1. The reaction is performed under nitrogen to prevent ambient water vapor from being absorbed by the reaction mixture. The rate is fit to models of solid-solid systems in order to determine if the reaction is a solid-solid reaction or merely a solvent-free reaction. Introduction As waste from chemical syntheses may be harmful to the environment, reducing or eliminating solvents in organic reactions has been identified as a beneficial goal of “green” chemistry. It has been shown that most reactions lacking a solvent proceed according to a solvent-free model, in which the neat reactants form a melt prior to reacting. This solvent-free model is in contrast to a solid-solid reaction, in which the neat solid reactants interact to form a solid product without the intervention of a liquid or vapor phase.1 More knowledge of solvent-free reaction models might allow for an extension of the available solvent-free reactions. Recently, some Wittig reactions have been shown to proceed without the requirement of a solvent.2 Discussion The data of product formation and reactant decay were fit to solution phase models, Eqn. 1 and 2, which suggest a solvent-free rather than a solid-solid reaction. The models suggest that the ratio of reactants is one (A/Y=1); however, the differing initial absorbencies show that the initial relative amount of reactants varied with each run. This may suggest the solid-solid kinetic model in which only the reactants in close proximity and with proper orientation react. The significantly different (in the open environment) and greater rate constant of product formation at 1705 cm-1 may reflect less spectral overlap from aldehyde decay than present at1164 cm-1. Likewise, ylide decay at 1604 cm-1 may appear slower due to product formation spectral overlap. A base line absorption at t=infinity was required in the model for reactant decay, suggesting a rate constant, too small to detect within 300 minutes, for non-adjacent reactants. There does not appear to be a statistically significant difference between dry and open conditions. This could be attributed to the reactants not being completely dry, or again suggests the solid-solid kinetic model in which the reactants in close proximity and with the proper orientation react. Experimental A Thermo Scientific Nicolet iS10 Fourier Transform Infrared spectrometer with Smart Orbit attachment was used to collect all spectral data. For all trials, approximately 5 x 10-4 mol of 4-bromobenzaldehyde was added to a stainless steel vial containing 5 x 10-4 mol (carbethoxymethylene)triphenylphosphorane. The reaction mixture was immediately shaken for 10 seconds in a Spex Sample Prep-80,000- Mixer/Mill and then a scoop of the mixture was transferred to the FT-IR spectrometer. “Open” reactions were performed with the mixture open to the environment while running and “dry” reactions were performed with the mixture inside a glove bag under a nitrogen environment. Spectral data of the selected peaks was recorded and analyzed to determine their respective rate constants using Microsoft Excel (2007). Derivation of the kinetic models used to determine rate constants are shown in Equation 1 and 2. Results Fig.6 and Fig. 7 show representative data and model results for the reaction run in a dry environment. Rate constant determined from Eqn. 1 and 2 averaged over all trials ran for both open and dry environments are reported in Table 1. References 1) Rothenberg, G.; Downie, A.P.; Raston, C.L.; Scott, J.L., J Am. Chem. Soc, 2001, 123, 8701-8708. 2) Balema, V.P.; Wiench, J.W.; Pruski, M.; Pecharsky, V.K, J Am. Chem. Soc, 2002, 124, 6244-6245. Acknowledgements We would like to recognize the department of chemistry at Washburn University for the use of their equipment and research space. Also, I would like to thank Dr. Stephen A. Angel for his guidance through the duration of this research project. Theoretical The rate of the reaction shown in Fig. 1 was fit to models of solid-solid systems in an effort to determine the nature of the reaction. The relative amounts of the 4 components of this reaction were monitored by collecting FT-IR spectra of the reaction mixture. The spectra of the 4 components were overlaid in order to select peaks unique to either one reactant or the desired product. As is shown in Fig. 2 and Fig.3, 1705 cm-1 and 1164 cm-1 were selected to track product formation, and 1604 cm-1 was selected to track ylide decay. In accordance with Beer’s Law, the change in intensity of the selected peaks is assumed to be directly proportional to the amount of reactant remaining or product formed. Figure 1 aldehyde + ylide → product Rate = -dA/dt = -dY/dt = dP/dt = k*A*Y A, Y, P are the amount of aldehyde (A), ylide (Y), and product (P) at time = t k = rate constant; relative A, Y, P determined by FT-IR absorbance Reactant Decay: Rate = -dY/dt = k*A*Y dY/dt = -k*Y*( Y – Y + A ) note: Yreacted = Y – Y Areacted = A – A Yreacted = Areacted dY/ [Y*( Y – Y + A )] = -k* dt separation of variables Solving this differential equation yields the model for ylide decay Equation 1: or Figure 5: Reactant decay at 1604 cm-1 Blue curve t = 0 Red curve t = 219 min Product formation: Rate = dP/dt = k*A2 substitution for A comes from solution to 2nd order decay let C=1/A0 and solve for P P = Pi + [2(C + 2kti)]-1 - [2(C + 2kt)]-1 At t=0, Pt=0= Pi + [2(C + 2kti)]-1 – [2C]-1 At t=∞, Pt=∞ = Pi + [2(C + 2kti)]-1 = ([2C]-1 + Pt=0) Therefore, [2C]-1 = Pt= - Pt=0 => 2C = (Pt= - Pt=0)-1 Equation 2: P = Pt= – (2C + 4kt)-1 Figure 2: Selection of peaks at 1705 cm-1 to track product formation and 1604 cm-1 to track ylide decay. aldehyde ylide product byproduct Figure 3: Selection of peak at 1164 cm-1 to track product formation aldehyde ylide byproduct product Table 1: Average rate constants with standard deviations 1705 cm-1 1164 cm-1 1604 cm-1 Avg. k Std. Dev. Avg. k Std. Dev. Avg. k Std. Dev. Avg. k in open environment 0.0055 s-1 0.0019 s-1 0.0020 s-1 0.0008 s-1 0.0016 M s-1 0.0008 M s-1 Avg. k in dry environment 0.0037 s-1 0.0042 s-1 0.0028 s-1 0.0017 s-1 0.0016 M s-1 0.0013 M s-1 Figure 6: Kinetic Models of Reaction Rate 0.15 0.2 0.25 0.3 0.35 0.4 0 50 100 150 200 250 300 Absorbance at 1164 cm-1 time (min) Actual Abs Abs from Model Figure 6: Representative data and model fit for product formation at 1164 cm-1 Figure 7: Representative data and model fit for reactant decay at 1604 cm-1 0.25 0.3 0.35 0.4 0.45 0.5 0 50 100 150 200 250 300 Absorbance at 1604cm-1 time (min) Actual Abs Abs from model Figure 4: Product formation at 1164 cm-1 Blue curve t = 0 Red curve t = 219 min