Report

Real-Time Observation of Bimodal Proton Transfer in Acid-Base Pairs in Water

See allHide authors and affiliations

Science  18 Jul 2003:
Vol. 301, Issue 5631, pp. 349-352
DOI: 10.1126/science.1085762

Abstract

The neutralization reaction between an acid and a base in water, triggered after optical excitation, was studied by femtosecond vibrational spectroscopy. Bimodal dynamics were observed. In hydrogen-bonded acid-base complexes, the proton transfer proceeds extremely fast (within 150 femtoseconds). In encounter pairs formed by diffusion of uncomplexed photoacid and base molecules, the reaction upon contact was an order of magnitude slower, in agreement with earlier reported values. These results call for a refinement of the traditional Eigen-Weller picture of acid-base reactions: A three-stage model is introduced to account for all observed dynamics.

Reactions of Brønsted acids and bases involve proton transfer and are of fundamental importance in chemistry and biology, because these are key processes in phenomena such as the autoionization in water (1), the anomalously high proton mobility in water (von Grotthuss mechanism) (2), acid-base neutralization reactions (3), enzyme catalysis (4), and proton pumps through membrane protein channels (5). In particular, acid-base neutralization reactions in solution have been the subject of a multitude of studies. According to Eigen (6, 7) and Weller (8), general acid-base reactions in solutions are bimolecular in nature and consist of (i) diffusional motion, where the acid and base approach each other to form an encounter pair when the mutual distance equals the reaction contact radius, followed by (ii) intrinsic proton transfer and (iii) subsequent diffusive separation. Theoretical studies have focused on step (ii) and on whether the reaction rate is determined by activated dynamics of the proton over a reaction barrier (9-11) or by proton-tunneling motion through the barrier (12) or whether solvent motions play a dominant role (13).

Experimentally, within the framework of the Eigen-Weller model, the intrinsic proton-transfer rates have been found to lie in the range of (10 ps)1 M1 with a typical reaction contact radius of 6 to 8 Å. However, diffusion rates are typically much slower than the on-contact proton-transfer rates, so direct access to the actual proton-transfer reaction dynamics between freely diffusing acid and base molecules in liquid solution remains problematic. Moreover, the role of the solvent in the acid-base encounter pair formation remains to be solved, in particular with respect to the observed intrinsic acid-base intermolecular proton-transfer rates contrasting strongly with observed dynamics in molecular systems that show intramolecular hydrogen transfer after electronic excitation (with typical time constants on the order of 100 fs) (14).

We have used photoacids (15) to prepare unreactive acid-base pairs in aqueous solution that are linked by a specific intermolecular (weak) hydrogen bond along which the proton-transfer reaction eventually occurs. An optical trigger pulse switches the acidity of the photoacid by several units of pKa (where Ka is the acid dissociation constant), initiating the proton-transfer reaction along the preexisting hydrogen bond (Fig. 1). Until now, proton-transfer dynamics of photoacids have been studied by probing electronic states of the photoacid or its base conjugate through time-resolved fluorescence or optical-pump/optical-probe spectroscopy (16, 17). These methods do not directly detect the arrival of the proton at the accepting site. Moreover, in condensed phase studies, electronic transitions are usually sensitive to intramolecular vibrational redistribution, vibrational cooling (vibrational energy dissipation to the solvent), and solvent reorganization (solvation dynamics) that may mask the ultrafast dynamics of the proton-transfer process (18).

Fig. 1.

Photoacidity of HPTS and deuteron transfer to acetate are switched on by optical excitation. The solid line indicates the general dependence of the percentage yield of proton/deuteron transfer on ΔpK(base-acid) and is on the basis of a compilation of data of several families of related acid-base reactions reviewed by Zundel (30).

We used femtosecond mid-infrared (mid-IR) spectroscopy after the optical trigger pulse to follow specific IR-active vibrational marker modes present at the proton-donor and proton-acceptor sides. The photoacid 8-hydroxy-1,3,6-trisulfonate-pyrene (HPTS) (19-22), when dissolved in deuterated water, changes its pKa after photoexcitation from 6.6 to 0.0 to become a strong acid. In an HPTS-acetate complex, the photoacid will only transfer its deuteron to acetate (with a pKa of 4.8) when excited to the S1 state (Fig. 1 and Scheme 1).

Scheme 1.

The largest changes between the IR spectra of electronic ground-state photoacid and photobase occur in the 1250 to 1600 cm1 frequency range, where vibrational bands of modes with aromatic ring C-O stretching activity are found (23, 24) (Fig. 2). We used the IR-active C=O stretching mode of acetic acid located at 1720 cm1 to monitor when the deuteron arrives at the accepting base over an acetate concentration between 0.25 and 4 M. At the lower concentrations, we observe a competition between deuteron transfer to the solvent and to the base from initially uncomplexed HPTS. At higher concentrations of acetate, we detect dynamics of hydrogen-bonded HPTS-acetate pairs initially prepared in the ground state.

Fig. 2.

(A) Transient spectra of HPTS in D2O (1350 to 1600 cm1) and growing in of acetic acid signal in a 4 M solution of acetate (1650 to 1780 cm1) after femtosecond laser excitation. OD, optical density. (B) Steady-state IR spectra of the photoacid HPTS in D2O (recorded at pD = 5) and its conjugate photobase (recorded at pD = 12) in their electronic ground states. (C) IR spectrum of acetic acid.

We allocate a transient band at 1486 cm1, which appears upon electronic excitation of HPTS within time resolution, to be indicative of the photoacid in the S1 state, and a band at 1503 cm1, which grows in on a time scale of several hundred picoseconds in the case of HPTS in D2O (no acetate added), to be a marker for the deuteron-transferred HPTS (photobase) in the S1 state (Fig. 2) (25). We measure the signals at parallel and perpendicular polarizations to correct for rotational anisotropy decay (time constant is 150 ps, assuming spherical rotational diffusion). The deuteron transfer occurs with a 250-ps time constant, in accordance with previous reports (19).

The transient C=O marker band provides a direct measure of the fraction of acetic acid generated by deuteron transfer, because no other species (HPTS photoacid and photobase or acetate) contributes in this frequency range. For an acetate concentration of 4 M, an initial substantially large contribution to the transient C=O band is generated within our time resolution (150 fs) that does not alter its magnitude and shape significantly within the first 1 to 2 ps. This initial contribution reveals through its dynamics a different microscopic origin than the additional C=O signal that grows in with a time constant depending on acetate concentration.

We ascribe this initial component of the transient C=O stretching band signal to the fraction of HPTS molecules that form a hydrogen-bonded complex with acetate when HPTS is in the ground electronic state. This fraction transfers its deuteron, after electronic excitation, to the associated acetate within our time resolution. When comparing the magnitude of the initial component relative to the total C=O signal that is generated [relative fractions are 0.54 for 4 M, 0.39 for 2 M, 0.24 for 1 M, and 0.11 for 0.5 M acetate; no detectable fraction was observed for 0.25 M acetate (Fig. 3)], we derive a complexation constant for HPTS and acetate of 0.28 (26). This bimolecular reaction rate is at least one order of magnitude larger than the rate previously reported (21) and likely originates from a HPTS-acetate “tight” complex observed at high acetate concentrations, made up by a direct hydrogen bond between the photoacid (A) and the base (B) of the type AH+···B, with water solvent molecules present only in the outer sphere of the complex. Direct hydrogen-bonded complexes are typical of relatively nonpolar environments not rich with water molecules. Aqueous solutions of acetate salts in excess of 1 M concentration appear to resemble water-deficient environments as far as their solvation abilities are concerned. At lower concentrations of acetate salts, aqueous solutions retain their bulk properties and fully solvate the acid and base molecules.

Fig. 3.

(A) Transient experimental band intensities of acetic acid taken over an extended period of time as function of acetate concentration, together with calculated traces (olive, 0.5 M; blue, 1.0 M; cyan, 2.0 M; purple, 4.0 M). arb. units, arbitrary units. (B) Comparison between the rise of the photobase, corrected for rotational diffusion and electronic excited state decay (red dots), and that of acetic acid (blue dots, experimental; blue line, calculated) for 1 M acetate concentration. (C) Same as (B) but for a lower acetate concentration (0.25 M).

We ascribe the concentration-dependent dynamics of the C=O signal to the fraction of HPTS molecules that initially are uncomplexed when in the ground state and that upon electronic excitation to the S1 state will either, depending on the acetate concentration, transfer the deuteron to a nearby acetate molecule that has diffused to form an encounter complex with HPTS or to the solvent D2O, after which the deuteron will be ultimately picked up by an acetate ion. Here, we compare measurements of the rise of the HPTS-anion band at 1503 cm1 with the rise of the C=O band of acetic acid at 1720 cm1 at a specific acetate concentration (Fig. 3). After correction for rotational anisotropy effects on the anion band, comparison of the rise times of the HPTS photobase and the acetic acid signals reveals that, for concentrations of 1 M of acetate or higher, direct deuteron transfer to the base in the “loose” and “tight” complexes dominates the reaction dynamics. In contrast, for lower concentrations of acetate, the HPTS-photobase signal rises faster than the acetic acid signal and indicates that deuteron transfer to the solvent occurs before deuteron pickup by the base.

We analyzed the fraction of the diffusion-controlled dynamics with the use of the expression for the time-dependent rate constant for a bimolecular reaction controlled by diffusion as derived by Szabo with the use of the Collins-Kimball radiative boundary conditions at contact (24, 27). The calculated response appears to mimic the observed signals in a consistent way, considering the assumptions made for the input parameters.

The dynamical model emerging from these observations points to a three-stage reaction scheme for acids and bases forming encounter pairs in water-rich environments (Fig. 4). Initially, diffusion allows the acid and base to form a “loose” encounter complex, each retaining its water solvation shell, at contact distance a. The reaction continues to be driven downhill, but now at the expense of the solvation energy of the solvent that needs to at least partially desolvate (or “rearrange”) the acid and base pair for further approach. This step is relatively slow compared to the final reaction stage of proton transfer along the hydrogen bond. The actual proton transfer along the connecting hydrogen bond occurs within a few hundred femtoseconds. The above-outlined mechanism agrees with the observation of Eigen and Weller of abnormally large contact radii measured in many diffusion-controlled acid-base reactions, including the self-neutralization reaction of water (OH- + H+ → H2O), where a large reaction distance of 6 to 8 Å was derived by various groups (3, 6, 28). Interestingly, a recent theoretical study on the HF dissociation-recombination reaction (29) has led to a conclusion in favor of a three-stage proton-transfer reaction mechanism and has found a concerted reaction mechanism upon diffusional encounter to be energetically unfavored.

Fig. 4.

Three-stage reaction mechanism for a general acid-base reaction.

Supporting Online Material

www.sciencemag.org/cgi/content/full/301/5631/349/DC1

Materials and Methods

SOM Text

References and Notes

View Abstract

Navigate This Article