Hydrolysis of an organophosphate ester by manganese dioxide

Article abstract of DOI:10.1021/es001309l

Amorphous manganese dioxide facilitates the hydrolysis of p-nitrophenyl phosphate to p-nitrophenol and orthophosphate despite insignificant adsorption of p-nitrophenyl phosphate or p-nitrophenol to the manganese dioxide. At pH 8, the orthophosphate produc

Full text of DOI:10.1021/es001309l

Environ. Sci. Technol. 2001, 35, 713-716
exhibited a very fast rate of hydrolysis without initial
Hydrolysis of an Organophosphate
Ester by Manganese Dioxide
adsorption of significant NPP, this oxide has been investigated
further to determine if the mechanism for the facilitated
hydrolysis reaction with manganese dioxide differs from that
of oxides that adsorb much of the NPP prior to hydrolysis
occurring.
D A R R E N S . B A L D W I N
Murray-Darling Freshwater Research Centre and the
CRC for Freshwater Ecology, P.O. Box 291,
Albury, New South Wales 2640, Australia
The mechanism of the hydrolysis reaction was studied by
first characterizing the surface of the oxide and then
determining the pH dependence of the reaction. The effects
on the rate of hydrolysis of the concentrations of NPP and
oxide and of added phosphate were examined. Manganese
dioxide is a strong oxidizing agent, so the possibility of
reductive dissolution of the oxide by oxidation of the product,
p-nitrophenol, during the hydrolysis reaction was also
explored (5).
J A M E S K . B E A T T I E * A N D
L YN E T T E M . C O L E M A N
School of Chemistry, University of Sydney,
Sydney, New South Wales 2006, Australia
D A V I D R . J O N E S
EWL Sciences, P.O. Box 39443, Winnellie,
Northern Territory 0821, Australia
Experimental Section
Materials. Amorphous manganese(IV) dioxide was made by
slowly adding a KMnO
4
solution (400 mL; 0.8 M) heated to
6
5 °C to a MnSO solution (300 mL; 1.6 M) heated to 90 °C
4
and then heating for a further 20 min at 90 °C. The resulting
brown suspension was washed with about 2 Lof hot distilled
water until the filtrate was no longer pink in color to remove
Amorphous manganese dioxide facilitates the hydrolysis
of p-nitrophenyl phosphate to p-nitrophenol and ortho-
phosphate despite insignificant adsorption of p-nitrophenyl
phosphate or p-nitrophenol to the manganese dioxide. At
pH 8, the orthophosphate product is released into solution;
at pH 4 and pH 6, some remains adsorbed. The rate of
hydrolysis is an order of magnitude more rapid than the
same reaction facilitated by iron oxides. Because manganese
dioxides are ubiquitous components of soils and sediments,
this suggests the possibility of significant abiotic pathways
for the formation of bioavailable orthophosphate from
phosphate ester precursors.
4
excess KMnO . The product was resuspended in cold distilled
water and filtered again to remove excess electrolytes. This
process was repeated until the filtrate had a conductivity of
-
1
0
.3 mS cm . After this washing, the oxide was placed in an
evaporating dish and oven-dried for 1 h at 100 °C.
Characterization Methods. The specific surface area of
the prepared manganese dioxide was determined using a
Quantasorb apparatus. Adsorption isotherm data were
analyzed by the BET surface area method. The samples were
degassed overnight at 150 °C. Nitrogen was used as the carrier
gas, and a nitrogen-helium mix was used for the adsorption-
desorption cycles.
The powder X-ray diffraction pattern of the oxide samples
was recorded with a Siemens Kristoloflex X-ray diffractometer.
The sample was prepared by grinding in an agate mortar
and pestle. The pattern was recorded over 35° at a collection
rate of 1°/ min.
The pH of the point of zero charge, pHpzc, of the manganese
dioxide was determined by performing acid-base titrations
of the oxide suspended in three different electrolyte con-
centrations of 0.001, 0.01, or 0.1 M KCl. The oxide loading
was 1 g/ L. The titration curves were plotted on the same
graph, and the point of intersection of the three curves was
Introduction
The bioavailability of phosphorus is an important factor in
controlling aquatic plant growth and eutrophication. The
adsorption of phosphate by mineral phases has been
extensively studied. Strong binding occurs with, for example,
iron and aluminum hydrous oxides that are positively charged
within the range of pH values typically found for natural
waters. Manganese oxides, however, are negatively charged
around pH 7 and have much lower affinities for anions such
as phosphate. Nevertheless, phosphate does bind to man-
ganese oxides at neutral pH (1). In seawater, the adsorption
of phosphate by amorphous manganese dioxide is compa-
rable to that by goethite on a mass basis despite the lower
affinity of phosphate for the manganese surface. The higher
surface area and site density of the amorphous manganese
oxide leads to significant adsorption.
taken to be the pHpzc
.
The number of phosphate adsorption sites was deter-
mined by suspending the oxide in 0.01 M KCl prepared with
doubly distilled water (0.1 g; 100 mL) and adjusting the pH
of the suspension to pH 6.00 with dilute HCl or NaOH. A
2 4
solution of Na HPO solution was then added to give final
2
concentrations of 1.6, 16, or 1.6 × 10 µM phosphate. The
suspensions were stirred for 6 h and filtered through 0.4-µm
Nucleopore polycarbonate membrane filters. The phosphate
remaining in solution was determined colorimetrically using
the Murphy and Riley molybdenum blue method (6). Each
experiment was duplicated and corrected for the blank.
Hydrolysis Experim ents. The catalyzed hydrolysis of the
model phosphate ester, p-nitrophenyl phosphate (NPP), was
examined because of the ease with which the concentrations
of both the nitrophenyl phosphate and the hydrolysis
product, p-nitrophenol (NP), can be determined. In acidic
conditions, the absorbance peaks for NP and NPP are
superimposed. However, under basic conditions the NP is
deprotonated to form the nitrophenolate ion, which is bright
We have recently shown that mineral phases also facilitate
the hydrolysis of organophosphate esters, indicating another
role in their influence on phosphorus availability (2). Three
manganese dioxide samples were examined in a survey of
the effects of metal oxides on the hydrolysis of p-nitrophenyl
phosphate (NPP) (3). Two of the oxides, an amorphous
2
manganese dioxide and akhtenskite (ꢀ-MnO ) (4), effected
the hydrolysis of NPP without exhibiting marked adsorption
of the NPP in the initial stages of the reaction. The third form
of manganese dioxide, pyrolusite, adsorbed NPP before
hydrolysis occurred. Since the amorphous manganese dioxide
*
Corresponding author e-mail: j.beattie@chem.usyd.edu.au;
phone: +61 2 9351 3797; fax: +61 2 9351 3329.
1
0.1021/es001309l CCC: $20.00

2001 Am erican Chem ical Society
VOL. 35, NO. 4, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7 1 3
Published on Web 01/12/2001

FIGURE 1. Time dependence of NPP and NP concentrations at pH
at an initial concentration of 62 µM NPP in a suspension of 1 g/L
of amorphous MnO
FIGURE 2. Time dependence of NPP and NP concentrations at pH
4 at an initial concentration of 55 µM NPP in a suspension of 1 g/L
of amorphous MnO2.
6
2
.
yellow and has an absorbance maximum at 400 nm, quite
distinct from NPP, which has an absorbance maximum at
3
10 nm. The molar extinction coefficients at 310 and 400
3
nm, respectively, are as follows: NP, 1.17 × 10 and 1.83 ×
0 M cm ; NPP, 1.00 × 10 and 63 M cm
The hydrolysis reaction was initiated by adding a known
4
-1
-1
4
-1
-1
1
.
concentration of NPP to a suspension of 0.25 g of manganese
dioxide in 250 mL of 0.01 M KCl prepared with Milli-Q water
with the pH adjusted by the addition of HCl or NaOH. The
solid oxide was kept in suspension by magnetic stirring, and
the temperature was maintained at 25.0 ( 0.1 °C in a water
bath. The pH was measured before the addition of the NPP
and subsequently through the reaction time. Aliquots of the
suspension were removed 5 min after initiation and at
subsequent regular intervals. The concentrations of NPP and
NP were measured colorimetrically after the 5-mL aliquot of
the suspension was syringe-filtered through a 25 mm
diameter Advantec glass fiber GA55 filter placed over a 0.4
µm Nucleopore polycarbonate filter.
The fate of phosphate released by the hydrolysis was
investigated by experiments at pH 4.0 and pH 8.0 with an
initial nitrophenyl phosphate concentration of 35 µM. The
hydrolysis reactions were monitored as described above by
measuring the production of nitrophenol. Additional 5-mL
aliquots were removed for analysis of phosphate with an
autoanalyzer using the colorimetric method developed by
Murphy and Riley (6).
FIGURE 3. Time dependence of NPP and NP concentrations at pH
8 at an initial concentration of 32 µM NPP in a suspension of 1 g/L
2
of amorphous MnO .
Figure 3. The sum of the concentrations of NPP and NP
measured in solution was used to test if either adsorbs on
the manganese dioxide surface.
Figures 1 and 3 show that the sum of the concentrations
of NP and NPP in solution at pH 6 and pH 8 equals the initial
NPP concentration over the entire reaction period, which
indicates that neither NP nor NPP adsorb significantly to the
surface of the oxide at these pH values. However, it is evident
from Figure 2 that at pH 4 there is initially some adsorption
of the NPP to the manganese dioxide. The sum of NPP and
NP increases with time as all of the NPP is hydrolyzed and
released as NP, which does not adsorb significantly to the
surface of manganese oxides (11).
Manganese dioxides are powerful oxidizing agents that
can oxidize substituted phenols (5). The possible dissolution
of manganese by reaction (reductive dissolution) with
nitrophenol produced in the hydrolysis reaction was therefore
investigated. The concentration of dissolved manganese in
the filtrate from a sample of manganese dioxide before and
after the hydrolysis of NPP was determined by graphite
furnace AAS. The small increase in Mn(II) found in solution
after the hydrolysis reaction had gone to completion was not
greater than the blank Mn(II) levels, which indicates that a
reductive dissolution mechanism in the reaction of NP and
manganese dioxide does not occur. This is consistent with
the literature description of the oxidation of substituted
phenols by manganese dioxide, which showed that p-
nitrophenol was unreactive in the presence of manganese
Results
Characterization of the Oxide. The oxide was characterized
by X-ray powder diffraction, BET surface area measurements,
and pHpzc measurements. The oxide was found to be mostly
amorphous material with a weak X-ray diffraction pattern
similar to a hydrated R-manganese dioxide (7); the specific
2
-1
surface area was 135 m
g
. The pHpzc was approximately
2
.3, which is lower than the pHpzc for R-manganese dioxide,
pH 4.6 (8). Several other manganese dioxides have been
identified with reported pHpzc values of 1.5 and 3.0 for
birnessite (9), 1.5 and 2.0 for cryptomelane (9), and 1.5 (8)
and 2.25 (10) for δ-manganese dioxide. The pHpzc for this
oxide is therefore consistent with the published range of
values for preparations of amorphous manganese dioxide.
Products of Reaction. The facilitated hydrolysis of NPP
by manganese dioxide (1 g/ L, equivalent to 11.5 mM MnO
2
)
was followed to completion at pH 4 and at pH 6 and to about
4
5
0% completion at pH 8, with initial NPP concentrations of
5, 62, and 32 µM, respectively. The changes in the con-
centrations of NPP and NP as a function of time are shown
for pH 6 in Figure 1, for pH 4 in Figure 2, and for pH 8 in
7
1 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 4, 2001

FIGURE 5. Dependence of the initial rates of hydrolysis of 30 µM
NPP on oxide loading (g/L) at pH 4 (upper line) and pH 8 (lower line).
FIGURE 4. pH dependence and NPP concentration dependence of
the initial rates of hydrolysis.
dioxide (5). This is also consistent with the observed 1:1
stoichiometry between NPP and NP shown in Figure 1.
The fate of the phosphate produced by the hydrolysis
reaction was determined by measuring the concentration of
phosphate in solution during the hydrolysis reactions at pH
4
and pH 8, with an initial NPP concentration of 35 µM. At
pH 4, the concentration of phosphate in solution increased
to a maximum of 7 µM, approximately 20% of the total
phosphate produced by the hydrolysis of NPP. After the
hydrolysis reaction was completed, the final solution con-
centration of phosphate was only 3 µM as the phosphate
released into solution by the faster hydrolysis reaction was
slowly re-adsorbed. At pH 8, however, the concentration of
phosphorus in solution was equal to the concentration of
NPP hydrolyzed, indicating that no significant amount of
the 35 µM phosphate was adsorbed to the surface at this pH.
The adsorption of phosphate to manganese dioxide is known
to decrease with increasing pH (12). Balistrieri and Choa
also showed that about 75% of 4.8 µM phosphate was
adsorbed on a 100 mg/ L suspension of amorphous man-
ganese dioxide at pH 4, but only about 10% (0.5 µM) was
adsorbed at pH 8 (13).
Rates of Hydrolysis. The rates of hydrolysis of NPP by
manganese dioxide (11.5 mM equivalent) were determined
between pH 4 and pH 8 with initial NPP concentrations in
the range of 5-65 µM (Figure 4). The initial rates were
calculated from a least-squares fit of the concentration of
NP with time in the first minutes of the reaction during which
time the rate was linear. The initial rates of reaction decreased
with increasing pH. Below pH 6, the rate increased with
increasing initial NPP concentration until a saturation limit
was reached above an initial NPP concentration of about 32
µM. The increase in rate is less than first order with respect
to NPP concentration, so that the first-order rate constants
decrease with the increase in NPP concentration. At pH 6
and above, the dependence of the initial rates on increasing
initial NPP concentration decreases still further until at pH
FIGURE 6. Inhibition by 36 µM phosphate of the hydrolysis of 36
µM NPP.
32 µM at the lower oxide level. In the biphasic reaction, the
faster rate is ascribed to the NPP adsorbed to the surface in
the first fast step of the reaction. The second, slower reaction
is that of the excess NPP that must subsequently be adsorbed
onto a surface that is increasingly negatively charged by the
presence of the phosphate product adsorbed at pH 6. The
existence of this biphasic behavior in the presence of excess
NPP implies that NPP binds more strongly to the surface
than does the product orthophosphate.
Oxide Loading Effect. The dependence of the initial rates
of reaction on the concentration of manganese dioxide was
determined by measuring the rates of the hydrolysis at pH
4 and pH 8 with an initial NPP concentration of 30 µM and
oxide concentrations between 0.4 and 2.2 g/ L (Figure 5). The
initial rates of hydrolysis are linearly dependent on the oxide
concentration with slopes of the log-log plots of 1.10 ( 0.09
at pH 4 and 1.06 ( 0.09 at pH 8, which indicate that the
hydrolysis reaction is first order with respect to oxide
concentration.
8
there is little dependence on the NPP concentration. This
Product Inhibition. Inhibition of the reaction by phos-
phate was examined by measuring rate of reaction after
reflects the decreasing adsorption of phosphate and NPP as
the pH increases.
addition of equimolar concentrations (36 µM) of NaHPO
and NPP to suspensions of manganese dioxide at pH 4, pH
, and pH 8. Figure 6 shows that there is significant product
4
The hydrolysis reactions were first order with respect to
the rate of NP formation except at pH 6 for an initial NPP
concentration of 64 µM, for which biphasic behavior was
observed. The biphasic behavior disappears if the experiment
is conducted with the same initial NPP concentration but
with twice the oxide loading. This is consistent with the
saturation limit at pH 6 of 32 µM described above and suggests
that the biphasic behavior occurs because the surface sites
of the oxide are saturated by the substrate concentration of
6
inhibition at pH 4 and pH 6.
Discussion
The hydrolysis of NPP is clearly facilitated by manganese
dioxide. The homogeneous rate is negligible by comparison
(14). This is consistent with the first-order dependence on
the mass of added oxide (Figure 5). At pH 8, this process can
VOL. 35, NO. 4, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7 1 5

be described as catalysis, for both the phenol and phosphate
products are released into solution and not adsorbed.
However, at pH 4 most of the phosphate remains adsorbed
on to the oxide, and added phosphate inhibits the NPP
hydrolysis reaction.
are consistent with a role for surface hydroxyl groups acting
as nucleophiles in the hydrolysis mechanism. This would be
an analogous process to that observed for intramolecular
hydrolysis of phosphate esters in discrete metal complexes
in homogeneous aqueous solution (for example, ref 22).
Manganese dioxide is unusual among the oxides previ-
ously surveyed for catalytic activity because it does not adsorb
significant amounts of NPP prior to the release of NP into
solution. Examples of related heterogeneous catalysis in
which there is no significant adsorption of the substrate to
the catalytic surface are known. These include the hydrolysis
of phenyl picolinate by goethite and titania (15), the oxidation
of As(III) to As(V) by manganese dioxide (16), and the
oxidation of 1,4-dihydroxybenzene by manganese dioxide
p-Nitrophenyl phosphate has been used as a substrate to
measure what is believed to be the activity of phosphatase
enzymes in soils (23, 24). Hence, it is possible to compare
“
enzymatic” activity measured in soil systems with the abiotic
rates measured for manganese dioxide in this work. In one
study with phosphomonoesterase at its optimum pH between
6
h
and 7, approximately 1 µmol of p-nitrophenol was released
(g of soil) at 37 °C. This can be compared with the rate
-1
-1
at pH 6 shown in Figure 1 of about 50 µmol of NP/ g of
amorphous manganese. In another study, the maximum
hydrolysis rate calculated from the integrated Michaelis-
(
17). All of these reactions involve the adsorption of the
substrate to the surface of the oxide. The fact that there is
little net absorption of the primary reactant prior to ap-
pearance of the product implies that either the equilibrium
binding constant is low or that the rate of the primary
adsorption process is slower than that of the subsequent
reaction of the adsorbed species (that is, the adsorption
process is rate limiting). Previous workers have suggested
that anion adsorption onto amorphous manganese dioxide
is weak because of the net negative surface charge above pH
-1
Menten equation was of the order 10-100 µmol h (g of
-1
soil) (23), comparable to the abiotic rates observed here.
While there are many assumptions made in these compari-
sons, it seems clear that the abiotic process is potentially
competitive with the inferred activity of phosphatase en-
zymes. Similar conclusions have been reached in a parallel
study on the hydrolysis of tripolyphosphate also facilitated
by amorphous manganese dioxide (12).
2
(18-20).
The second acid dissociation of NPP has a pKa2 of 5.18.
Acknowledgments
At pH 4, therefore, the monoanion predominates, while at
pH 6 and above the dianion is the dominant species. The
probably accounts for the limited adsorption of the NPP to
the manganese dioxide surface below pH 6 and no measur-
able adsorption above this pH. The pH dependence of the
rate constants for the catalyzed hydrolysis of NPP thus reflects
two factors: differences between the monoanion and dianion
of NPP and the pH dependence of the surface charge of the
manganese oxide. The data do not allow a distinction between
the two effects to be made. What is clear is that the reaction
is fastest at pH 4, where some of the NPP monoanion is
adsorbed and where the surface charge of the oxide is less
negative than at higher pH values.
The financial support of the Australian Research Council
Small Grants Scheme is gratefully acknowledged.
Literature Cited
(
(
1) Yao, W.; Millero, F. J. Environ. Sci. Technol. 1996, 30, 536.
2) Baldwin, D. S.; Beattie, J. K.; Jones, D. R. Water Res. 1996, 30,
1
123.
(
3) Baldwin, D. S.; Beattie, J. K.; Coleman, L. M.; Jones, D. R. Environ.
Sci. Technol. 1995, 29, 1706.
(
(
(
4) Jambor, J. L.; Grew, E. S. Am. Mineral. 1990, 75, 931.
5) Stone, A. Environ. Sci. Technol. 1987, 21, 979.
6) Murphy, J.; Riley, J. P. Anal. Chim. Acta 1962, 27, 31.
The rate constants at pH values less than 6 decrease
significantly with increasing initial NPP concentration. At
pH 8, however, the rate constants are almost independent
of the NPP concentration. This behavior is consistent with
stronger adsorption of the monoanion than of the dianion.
At low pH, the surface of the oxide becomes saturated at low
NPP concentrations. As the concentration of NPP at the
surface increases, the net negative surface charge increases
making the adsorption of more NPP more difficult.
The inhibition by phosphate displays a similar pattern
with inhibition at pH 4 and no effect at pH 8. At the lower
pH, phosphate from the hydrolysis reaction remains adsorbed
to the surface. In addition, added phosphate also binds and
therefore blocks active surface sites. This effect should lead
to product inhibition and deviation from first-order kinetics
as phosphate is produced at pH 4, but the kinetics data were
not precise enough to observe this effect. The pKa2 of
orthophosphate is 7.2. At pH 8, all the phosphate produced
by the hydrolysis reaction is released into solution, reflecting
the weak interaction between the surface and the phosphate
dianion, and added phosphate does not inhibit the hydrolysis
reaction.
(7) Rossouw, M. H.; Liles, D. C.; Thackeray, M. M. Mater. Res. Bull.
992, 27, 221.
1
(8) Healy, T. W.; Herring, A. P.; Fuerstenau, D. W. J. Colloid Interface
Sci. 1966, 21, 435.
(
9) McKenzie, R. M. Aust. J. Soil Res. 1981, 19, 41.
(10) Murray, J. W. J. Colloid Interface Sci. 1974, 46, 357.
(11) Stone, A. T.; Morgan, J. J. Environ. Sci. Technol. 1984, 18, 450.
(12) Inman, M. P.; Beattie, J. K.; Jones, D. R.; Baldwin, D. S. Water
Res., in press.
(13) Balistrieri, L. S.; Chao, T. T. Geochim. Cosmochim. Acta 1990,
5
4, 739.
(14) Cox, J. R.; Ramsey, O. B. Chem. Rev. 1964, 64, 317.
(15) Torrents, A.; Stone, A. T. Environ. Sci. Technol. 1990, 25, 143.
(16) Scott, M. J.; Morgan, J. J. Environ. Sci. Technol. 1995, 29, 1898.
(
(
(
17) McBride, M. B. Clays Clay Miner. 1989, 37, 479.
18) McKenzie, R. M. Geochim. Cosmochim. Acta 1979.
19) Kawashima, M.; Tainaka, Y.; Hori, T.; Koyama, M.; Takamatsu,
T. Water Res. 1986, 20, 471.
(
20) Takematsu, T.; Kawashima, M.; Koyama, M. Water Res. 1985,
1
9, 1029.
(
(
21) Coleman, L. M. Ph.D. Thesis, University of Sydney, 1996.
The hydrolysis of NPP by amorphous manganese dioxide
is an order of magnitude more rapid, per gram of oxide, than
the same reaction facilitated by hydrous ferric oxide (2). This
difference is probably due to the increased acidity of the
surface hydroxyl groups of the manganese(IV) surface, with
pHpzc of 2.3, as compared with hydrous iron(III) oxide with
pHpzc of 8.2. Hence, the rate of reaction decreases from pH
22) Jones, D. R.; Lindoy, L. F.; Sargeson, A. M. J. Am. Chem. Soc.
1
983, 105, 7327.
(
(
23) Gerritse, R. G.; van Dijk, H. Soil Biol. Biochem. 1978, 10, 545.
24) Eivazi, F.; Tabatabai, M. A. Soil Biol. Biochem. 1977, 9, 167.
Received for review May 29, 2000. Revised manuscript re-
ceived November 9, 2000. Accepted November 14, 2000.
4
to pH 8 for the manganese system, but for the iron system
it reaches a maximum at pH 6.9 (21). The observed differences
ES001309L
7
1 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 4, 2001