Spectroscopy of Hydrothermal Reactions. 5. Decarboxylation Kinetics of Malonic Acid and Monosodium Malonate

Article abstract of DOI:10.1021/jp960397m

The kinetics of decarboxylation of 1.07 m malonic acid and monosodium malonate at 120-230 deg C and 275 bar were determined directly in a combined microflow reactor and short-pathlength IR spectroscopy cell.Malonic acid and monosodium malonate decomposed in a single step to CO2 and acetic acid and CO2 and the acetate ion, respectively.The kinetics for malonic acid were modeled as an equilibrium between malonic acid and the monoanion with parallel decarboxylation of both species.The Arrhenius parameters are E_a = 120 +/- 2 kJ/mol and ln A (s^-1) = 29.4 +/- 0.1 for the malonate monoanion and E_a = 126 +/- 2 kJ/mol and ln A (s^-1) = 31.4 +/- 0.1 for malonic acid in a 316 stainless steel flow reactor at 120-210 deg C.Decarboxylation of the monoanion is slower than that of malonic acid.The rates of decarboxylation of malonic acid are slower above 140 deg C in a Pt/Ir alloy flow reactor with diamond windows than in a 316 stainless steel flow reactor with sapphire windows.

Full text of DOI:10.1021/jp960397m

14352
J. Phys. Chem. 1996, 100, 14352-14355
Spectroscopy of Hydrothermal Reactions. 5. Decarboxylation Kinetics of Malonic Acid
and Monosodium Malonate
P. G. Maiella and T. B. Brill*
Department of Chemistry, UniVersity of Delaware, Newark, Delaware 19716
ReceiVed: February 8, 1996; In Final Form: April 29, 1996X
The kinetics of decarboxylation of 1.07 m malonic acid and monosodium malonate at 120-230 °C and 275
bar were determined directly in a combined microflow reactor and short-pathlength IR spectroscopy cell.
Malonic acid and monosodium malonate decomposed in a single step to CO2 and acetic acid and CO2 and
the acetate ion, respectively. The kinetics for malonic acid were modeled as an equilibrium between malonic
acid and the monoanion with parallel decarboxylation of both species. The Arrhenius parameters are Ea )
120 ( 2 kJ/mol and ln A (s-1) ) 29.4 ( 0.1 for the malonate monoanion and Ea ) 126 ( 2 kJ/mol and ln
A (s-1) ) 31.4 ( 0.1 for malonic acid in a 316 stainless steel flow reactor at 120-210 °C. Decarboxylation
of the monoanion is slower than that of malonic acid. The rates of decarboxylation of malonic acid are
slower above 140 °C in a Pt/Ir alloy flow reactor with diamond windows than in a 316 stainless steel flow
reactor with sapphire windows.
Introduction
without leaching the reactor walls. The pathway can be deduced
with relative ease from the IR spectrum.
Reactions of carboxylic acids occur in the hydrothermal
environment in a broad spectrum of applications. For example,
pressure and temperature are among several variables which
can be manipulated to alter the properties of H2O and, thereby,
optimize the yield of desired products. Examples include the
interconversion of lactic and acrylic acid,1,2 which has been
accomplished in sub- and supercritical H2O. The sequential
conversion of citric to itaconic to methylacrylic acids in
subcritical H2O has recently been described by Antal et al.3
Sedimentary brines and ocean-based volcanic vents are
hydrothermal environments where carboxylic acids are among
the important organic compounds formed. Bell and Palmer4
describe the role of mono- and dicarboxylic acids in brines,
while Shock5 has discussed the thermodynamics of organic acids
in brines.
Decarboxylation of malonic acid and monosodium malonate
has been previously studied at lower temperatures, in the batch
mode, and by ex situ analysis of the products. Bernoulli and
Wege7 discussed malonic acid in the 75-110 °C range at
approximately 1 atm. Fairclough8 examined aqueous mono-
sodium malonate at 76-120 °C in a sealed tube. Hall9
conducted studies of both compounds at 80-90 °C at about 1
atm.
In this paper, we have investigated malonic acid and
monosodium malonate directly at 120-210 °C in a flow reactor
which possessed windows suitable for transmission IR spec-
troscopy. The observation of the reactants and products in real
time by infrared spectroscopy enables the kinetics and pathway
to be determined simultaneously.
Carboxylic acids are present in many waste streams owing
to their widespread use in industrial processes and commercial
products (e.g., preparation of polymers, paper products, phar-
maceuticals, dyes, formates, flavoring agents, perfumery esters,
etc.). The destruction of these aqueous wastes by wet air
oxidation6 involves hydrothermal conditions and is one approach
to meeting local discharge standards for wastewater.
In none of the areas of research or application mentioned
above has a real time, direct spectroscopic measurement of the
rate or pathway of reaction of a carboxylic acid been reported
at hydrothermal conditions. This void was one of the motiva-
tions of the research described in this article. The focus here
is on the conversion of aqueous malonic acid and monosodium
Experimental Section
A microflow reactor, which is also a precision spectroscopy
cell, was used to study the reactions described herein. Con-
trolled, constant conditions of temperature (25-450 °C), pres-
sure (1-335 bar), and flow rate (0.03-2.5 mL/min) are
available as chosen. Both a 90/10 Pt/Ir alloy cell with diamond
wafer windows10 and a 316 stainless steel (SS) cell with sapphire
windows11 were used. A detailed description of the flow reactor
(surface-to-volume ratio of 25) and operating procedures is given
elsewhere.10,11 For the studies described herein, kinetics were
determined at flow rates at 2.00-0.05 mL/min, which translates
to residence times in the cell of 1.79-71.4 s. The temperature
range of the study was 120-230 °C, and the pressure was
maintained at a constant value of 275 bar at all times. These
conditions ensured that a single fluid phase is present over the
entire range of studies. Multiple determinations (three-five)
were made at several sets of conditions to determine the
precision of a given rate measurement. Malonic acid solutions
were prepared from 99% malonic acid (Aldrich). The mono-
sodium malonate was prepared by adding 1 equiv of NaOH to
a solution of malonic acid.
malonate to acetic acid and sodium acetate. Malonic acid was
chosen because it is relatively reactive and noncorrosive and
has a straightforward decarboxylation reaction. Thus, the
conversion takes place rapidly at subcritical conditions in H2O
A Nicolet 60SX FTIR spectrometer with a liquid N2 cooled
MCT detector was used for cell spectral measurements. Spectra
were obtained at 4 cm-1 resolution by adding 32 interferograms.
* Author to whom correspondence should be addressed.
X Abstract published in AdVance ACS Abstracts, August 1, 1996.
S0022-3654(96)00397-8 CCC: $12.00 © 1996 American Chemical Society

Spectroscopy of Hydrothermal Reactions. 5
J. Phys. Chem., Vol. 100, No. 34, 1996 14353
Figure 2. First-order rate plots for decarboxylation of 1.07 m
monosodium malonate.
TABLE 1: Rate Constants (1.07 m Solutions) under 275
bara
HCO2CH2CO2H
NaCO2-
CH2CO2H
SS-sapphire cell
Pt/Ir-diamond cell
Figure 1. Mid-IR spectra of 1.07 m malonic acid at 200 °C and 275
bar at several residence times in the Pt/Ir-diamond flow cell.
temp,
°C
103k1,
103k1,
103k2,
103k1,
103k2,
s-1
s-1
s-1
s-1
s-1
Absorbance areas were determined by curve fitting the base-
line-corrected spectra using Peakfit (Jandel Scientific).
120
140
160
170
180
190
200
210
220
230
0.5
2.6
16
0.515
4.17
22.3
13
27
55 ( 7
114 ( 62
242 ( 76
611
Results and Discussion
74 ( 13
128 ( 1
110 ( 25
243 ( 1
The major reaction of dicarboxylic acids is decarboxylation
to produce CO2 and the monocarboxylic acid in which the
backbone has one fewer carbon atoms.4 Figure 1 shows several
of the many IR spectra of 1.07 m (m ) mol kg-1 of solvent)
malonic acid recorded with the Pt/Ir-diamond cell at different
residence times. Because the temperature of this data set is
200 °C, significant conversion to CO2 has taken place, as
evidenced by ν3(CO2) at 2345 cm-1. Also, the overlapping
C-O stretch and COH in-plane bending modes characteristic
70
113
220 ( 62 377 ( 111 147
201
334
573
831
490 ( 86 703 ( 250 209
295
375
a Rates without standard deviations are single measurements.
conditions of this study, the ionization constant of malonic acid
and the rates of decarboxylation of the malonic acid and its
monoanion are unknown. Therefore, the relative roles of these
two species in producing CO2 are not known. Hence, deter-
mination of this rate of decarboxylation of the monosodium
malonate solution at hydrothermal conditions is essential to
determining the kinetics of decarboxylation of malonic acid.
Monosodium Malonate. The malonate monoanion is in
equilibrium with malonic acid and the malonate dianion.
However, on the basis of an isocoulombic form of the
equilibrium constant,15 the amount of diacid and dianion in a
1.07 m monosodium malonate solution is negligible within the
uncertainty of our kinetic measurements (i.e., a % dianion range
of (1.7-1.1) × 10-1 and a % diacid range of (0.48-1.0) ×
10-2 assuming an ionic strength of 1) from 160 to 210 °C.16,17
Therefore, decarboxylation of the aqueous malonate monoanion
is considered to be the single-step, first-order reaction shown
in eq 1. In eq 1 the concentration (R) m of CO2 is a simple
of malonic and acetic acid are observed at 1100-1300 cm-1
.
The CdO stretch at about 1750 cm-1 overlaps the ν2 bending
mode of H2O, which is difficult to subtract completely from
the spectrum under these conditions. All of the absorbances
observed could be attributed to malonic acid, acetic acid, and
CO2, which is valuable information because it eliminates the
existence of other major reaction channels. A total carbon
balance, however, was not determined because of spectral
overlap of the acids and their anions. It was confirmed that
decomposition of the acetic acid and acetate ion products is
not detected by IR spectroscopy at the temperatures and
residence times used here. This is not surprising given the
stability of acetic acid against hydrothermal decomposition.12-14
As a result, the absorbance of ν3(CO2) forms the basis of the
kinetic analysis.
The studies were conducted in both the 316 SS-sapphire
cell and the Pt/Ir-diamond cell. Sapphire permits only the rate
of conversion to CO2 to be discerned. However, the absence
of secondary and side reactions means that knowledge about
the concentration of CO2 enables the decarboxylation of malonic
acid to be specified unambiguously. One problem remains:
malonic acid is in equilibrium with the malonate ion(s). As a
result, two sources of CO2 exist, (i.e., decarboxylation of
malonic acid and the malonate monoanion4,9). Under the
k1
HOOCCH2COO- 98 CO2 + CH3COO-
(1)
function of the concentration (1.07 - R) m of the monoanion.
The area of ν3(CO2) was converted to concentration by using
the external calibration results for aqueous CO2 described earlier
under the same pressure and temperature conditions.11 Selected
concentration-time data are plotted in Figure 2 as a first-order

14354 J. Phys. Chem., Vol. 100, No. 34, 1996
Maiella and Brill
Figure 3. Comparison of the experimental CO2 concentrations (points)
and the optimization of the rates in the model (lines) for eqs 1-3.
rate function at several temperatures. The slope of each line is
the rate (k1) for decarboxylation of the monoanion at that
temperature. Although the difference in the ionic strength was
not taken into account, the values of k1 compiled in Table 1 are
useful to complete the analysis of malonic acid. The uncertainty
given for the values of k is the standard deviation based on
multiple determinations.
Figure 4. Arrhenius plots for the available experimental data for
decarboxylation of monosodium malonate. Where no error bar is
shown, the rate is based on a single measurement.
Malonic Acid. The same measurements of the concentration
of CO2 were made on 1.07 m malonic acid as were made for
monosodium malonate. In principle, according to eq 3, the
kf
HOOCCH2COOH y
\
z HOOCCH2COO- + H+
(2)
kr
k2
HOOCCH2COOH
9
k1
8 CO2 + CH3COOH
(3)
(1)
HOOCCH2COO- 98 CO2 + CH3COO-
concentration of malonic acid can be obtained directly from
the concentration (R) of CO2. However, the equilibrium (eq
2) produces a small amount of the malonate monoanion which
contributes to the concentration of CO2 (eq 1) by the rate
constant k1. Consequently, the best fit of the coupled differential
rate equations for eqs 1-3 with the experimental data was
achieved by minimizing the residuals with a multilevel, single-
linkage optimization routine. The optimization procedure took
into account the estimation of the malonic acid concentration
as well as k1 for the malonate monoanion described above.
Figure 3 shows the fit of the experimental concentration data
at 190 °C with those predicted by the model. The resulting
values of k2 are tabulated in Table 1. The values of kf and kr
are not reported because of their random behavior. However,
their ratio was fairly constant and the calculated values of k1
and k2 were rather insensitive to kf and kr. Consistent with
previous reports,4,7-9 the rate of decarboxylation of the malonate
monoanion is observed to be slower than that of malonic acid.
This finding may be explained by the existence of a cyclic
transition state (shown below) for the decarboxylation reaction
(∆Sq ) 1 eu at 200 °C).
Figure 5. Arrhenius plots of the available experimental data for
decarboxylation of malonic acid. Where no error bar is shown, the
rate is based on a single measurement.
to the total carbon balance, it is noteworthy that the final CO2
concentration in Figure 3 (1.05 m) is close to the initial
concentration of malonic acid (1.07 m) as required by the
stoichiometry of eqs 1-3.
Figures 4 and 5 are Arrhenius plots of the rate data in Table
1 for the monosodium malonate and malonic acid, respectively.
Also shown are the Arrhenius data obtained previously7-9 for
these compounds at lower temperature and pressure, in the batch
mode, and by post-reaction analysis. It is not surprising that
differences exist among all of these data. The rates measured
by Hall9 at 80-90 °C can be roughly extrapolated to the
hydrothermal regime of this study. Those of Bernoulli and
Wege7 and Fairclough8 have similar activation energies, but the
A factors are markedly different.
Although there is only anecdotal evidence from this study, it
is interesting to find that the rate of formation of CO2 in the
316 SS-sapphire cell is faster than in the Pt/Ir-diamond cell
above about 140 °C. Figure 6 shows the Arrhenius plots
obtained with these two cells. Wall effects and surface catalysis
in these high surface-to-volume ratio reactors appear to have
This transition state is statistically less favored for the monoan-
ion. As a result, a lower A factor is expected and found (Vide
infra) for the monoanion (∆Sq ) -3 eu at 200 °C). With regard

Spectroscopy of Hydrothermal Reactions. 5
J. Phys. Chem., Vol. 100, No. 34, 1996 14355
References and Notes
(1) Mok, W. S. L.; Antal, M. J., Jr.; Jones, Jr. J. Org. Chem. 1989, 54,
4596.
(2) Lira, C. T.; McCrackin, P. J. Ind. Eng. Chem. Res. 1993, 32, 2608.
(3) Carlsson, M.; Habenicht, C.; Kam, L. C.; Antal, M. J., Jr.; Bian,
N.; Cunningham, R. J.; Jones, M., Jr. Ind. Eng. Chem. Res. 1994, 33, 1989.
(4) Bell, J. L. S.; Palmer, D. A. In Organic Acids in Geochemical
Processes; Pittman, E. D., Lewan, M. B., Eds.; Springer-Verlag: Berlin,
1994; p 226 and references therein.
(5) Shock, E. L. In Organic Acids in Geochemical Processes; Pittman,
E. D., Lewan, M. B., Eds.; Springer-Verlag: Berlin, 1994; p 270 and
references therein.
(6) Mishra, V. S.; Mahajuni, V. V.; Joshi, J. B. Ind. Eng. Chem. Res.
1995, 34, 2.
(7) Bernoulli, A. L.; Wege, W. HelV. Chim. Acta 1919, 2, 511.
(8) Fairclough, R. A. J. Chem. Soc. 1938, 1186.
(9) Hall, G. A., Jr. J. Am. Chem. Soc. 1949, 71, 2691.
(10) Schoppelrei, J. W.; Kieke, M. L.; Wang, X.; Klein, M. T.; Brill, T.
B. J. Phys. Chem. 1996, 100, 14343.
(11) Kieke, M. L.; Schoppelrei, J. W.; Brill, T. B. J. Phys. Chem. 1996,
100, 7455.
(12) Meyer, J. C.; Marrone, P. A.; Tester, J. W. AIChE J. 1995, 41,
Figure 6. Comparison of the Arrhenius parameters for decarboxylation
of malonic acid in the 316 SS-sapphire cell and Pt/Ir-diamond cell.
2108.
(13) Boock, L. T.; Klein, M. T. Ind. Eng. Chem. Res. 1993, 32, 2464.
some influence on the kinetics of hydrothermal decarboxylation
of these dicarboxylic acids. At lower temperatures and pres-
sures, the conditioning of a glass surface was found to affect
the decarboxylation rate of malonic acid.7 Much stronger wall
effects and catalysis appear to exist in monocarboxylic acids.5,18-20
At higher temperature, the rate of decarboxylation of acetic acid
is faster on SS than on gold.21 In contrast, the decomposition
rate of urea, for which the pH is higher than malonic acid, only
minimally depends on whether the material of construction of
the cell is SS or Pt/Ir alloy.10
(14) Foussard, J.; Debellefontaine, H.; Besombes-Vaihe, J. J. EnViron.
Eng. 1989, 115, 367.
(15) Marshall, W. L.; Franck, E. U. J. Phys. Chem. Ref. Data, 1981,
10, 295.
(16) Kettler, R. M.; Wesolowski, D. J.; Palmer, D. A. J. Solution Chem.
1992, 21, 883.
(17) Bell, J. L. S.; Wesolowski, D. J.; Palmer, D. A. J. Solution Chem.
1993, 22, 125.
(18) Falconer, J. L.; Madix, R. J. Surf. Sci. 1974, 46, 473.
(19) McCarty, J.; Falconer, J.; Madix, R. J. J. Catal. 1973, 30, 235.
(20) Maiella, P. G.; Brill, T. B. To be published.
(21) Palmer, D. A.; Drummond, S. E. Geochim. Cosmochim. Acta 1991,
50, 813.
Acknowledgment. We are grateful to the Army Research
Office for financial support of this project on DAAL03-92-G-
0174.
JP960397M