2008:046
M A S T E R ' S T H E S I S
Adsorption-Desorption Kinetics of Dicarboxylic Acids on Synthesized
Iron Oxide Nano- and Mesoporous Particles
Yu Zhang
Luleå University of Technology Master Thesis, Continuation Courses Chemical and Biochemical Engineering Department of Chemical Engineering and Geosciences
Division of Chemistry
2008:046 - ISSN: 1653-0187 - ISRN: LTU-PB-EX--08/046--SE
Abstract
The adsorption-desorption of aliphatic acid on mineral oxide surfaces regulates several environmentally significant chemical reactions. The kinetics of these reactions provides useful information on the evaluation and prediction of the adsoption rate and implies fundamental information on the adsorption mechanism. In the current work, nano and mesoporous iron oxide has been successfully synthesized using the ultrasonic method.
The adsorption of dicarboxylic acid (succinic acid and maleic acid) at the hematite/water interface was studied by attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR).
The kinetics of adsorption was investigated at different initial bulk concentrations and various pH values. The results show that both adsorption and desorption of the two dicarboxylic acids studied are characterized by a fast initial reaction during the first 10 minutes and a much slower reaction eventually reaching a plateau value after several hours. The adsorption rate increases by increasing concentration and at pH 5 the adsorbed amount and the total rate of reaction rate is higher than at pH 3 and pH 8. This indicated that electrostatic interaction plays an important role in the adsorption process.
Clear spectral differences between the two kinds of dicarboxylic acids upon adsorption onto hematite allowed us to discriminate between chemisorbed and physically adsorbed species as well as inner-sphere/outer-sphere complexes. It was shown that succinic acid is chemisorbed at the hematite surface, whilst the -CH=CH- alkenyl segment in the hydrocarbon chain of maleic acid seemed to prohibit its chemisorption on the surface. Furthermore, for succinic acid, both inner-sphere and outer-sphere complexation existed simultaneously at all pH values. The desorption experiments performed turned out to be very helpful in order to distinguish between inner-sphere and outer-sphere complexation. The initially adsorbed molecules contributing to the higher reaction rate is interacting with the hematite surface through hydrogen bonding interaction (outer-sphere complexation). Eventually, at a lower reaction rate, these outer sphere complexes are forming chemically bonded inner-sphere complexes.
Keywords: Iron oxide; Adsorption; Desorption; Kinetics; Mechanism
1. Introduction... 4
1.1. Background... 4
1.2. Scope of the thesis ... 5
2. Experimental Section... 5
2.1 Sample sythesis and general characterization ... 5
2.1.1. Sample preparation ... 5
2.1.2. General characterization ... 6
2.2. ATR-FTIR experiments... 6
2.2.1. Preparation of hematite film on ZnS crystal... 6
2.2.2. Spectra of adsorbed dicarboxylic acid... 6
3. Results and discussion... 7
3.1. General characteristics of the synthesized sample ... 7
3.2. Adsorption kinetics... 9
3.3. Desorption kinetics ... 14
3.4. Surface complexation mechanism ... 19
4 Conclusions... 21
5 Future work... 22
6 ACKNOWLEDGEMENTS... 23
7 References... 24
Adsorption-Desorption Kinetics of Dicarboxylic Acids on Synthesized Iron Oxide Nano- and Mesoporous Particles
1. Introduction 1.1. Background
Iron oxide nanoparticles are extensively used for magnetic data storage, as ferrofluids, in biomedical sorbents and as catalysts, gas sensors, and ion exchangers1-3. In addition, due to the porous structure greatly increaseing the surface area/volume ratio, mesoporous iron oxide films offer high potential as materials for electrodes, optical devices and catalysts4. Accordingly, mesoporous iron oxides have motivated extensive research interest.
Carboxylic acids are an important class of organic ligands in aquifers and soils. The adsorption-desorption of carboxylic and dicarboxylic acids on mineral oxide surfaces regulates several environmentally significant chemical reactions, such as photochemical dissolution of mineral oxides, distribution and transport rate of colloids and pollutants, and competitive adsorption with other chemical components in aqueous solution 5-8.
The adsorption of various mono- and dicarboxylic acids from aqueous solutions onto mineral surfaces has been studied using in-situ attenuated total reflectance Fourier transform infrared (ATR-FTIR) methods9-16. These studies have mainly focused on characterizing the type(s) and structure(s) of surface complexes. Inner-sphere complexes (monodentate and bidentate structures ) as well as outer-sphere complexation (hydrogen bonding) have been inferred to explain the adsorption reactions. Other authors have reported that ligands such as dicarboxylic acids can also be simultaneously present as outer-sphere and inner-sphere surface complexes, as indicated in Figure 117-19. However, in the present investigation the kinetics of adsorption/desorption has been addressed in combination with a synthetic mesoporous iron oxide with high surface area facilitating the interpretation of infrared data. The aim was to elucidate the kinetics of interaction and the binding mechanisms of these carboxylic acids at the mesoporous iron-oxide/aqueous interfaces. The interaction is expected to vary with pH in solution, so this variable was also included. According to the literature the following possible surface complexes have already been proposed
C O O
C O C H H ( )
n O
C O O
C O C H H ( )
n O
C O O
C O C
H H ( )n
O
C O
O H O
H O
O O H
H ...
...
Fe
Fe Fe Fe
n
C C
Fe Fe
C O
O H O
H O
n
C C
H H
H2O H2O H2O H2O
Fe HO
O Fe H
C O
O H O
H n O
C
C Fe
Fe Fe
...H O Fe C
O
O H O
H n O
C
C Fe Fe Fe
C O
O H O
H n
C
C Fe
Fe
...H O Fe
...H O Fe O
(a) monodentate (b) bidentate chelating (c) bidentate bridging
(d) outer-sphere hydrogen bonding (e) outer sphere no chemical bonding
(f) bidentate bridging - monodentate (g) bidentate chelating -
(h) twice monodentate H-bonded monodentate H-bonded
Figre1 Basic surface structures for adsorbed dicarboxylic acids17-19
1.2. Scope of the thesis
The scope of this thesis mainly encompasses two parts: 1) the synthesis of nanoporous and mesoporous iron oxide and the characterization of these oxides, 2) the study of the adsorption and desorption kinetics of two dicarboxylic acids (succinic and maleic acid) at the hematite water interface.
In-situ ATR-FTIR spectroscopy was utilized to monitor the adsorption as a function of time. Due to their two pKa values each of the two acids have different dominant species at different pH values.
Accordingly, the kinetics at various pH values had to be studied and related to the dominant species in solution. Furthermore, the dominant species in solution is a valuable knowledge in order to predict a reasonable mechanism for the surface complexes formed at a certain pH.
2. Experimental Section
2.1 Synthesis and general characterization of porous iron-oxide
2.1.1. Sample preparation
In a 100 mL beaker, 12.0 mmol FeCl3·6H2O and 12.0 mmol dodecylamine were mixed forming a dark brown solution under magnetic stirring. Subsequently adding 16 ml NH3·H2O, the solution
remained dark brown with some precipitation formed. After that the system was subjected to an ultrasonic bath for 3h. The precipitate was the washed with distilled water and ethanol several times and dried before chemical characterization.
2.1.2. General characterization
As-synthesized and template-extracted samples were calcined at different temperatures and subsequently characterized by SEM, Diffuse Reflectance Infrared Fourier Transform (DRIFT) spectroscopy, X-ray diffraction (XRD), and BET N2 adsorption – desorption isotherms.
2.2. ATR-FTIR experiments
2.2.1. Preparation of hematite film on ZnS crystal
A stock dispersion of hematite (1.5 g L−1) was prepared by dispersing solid hematite in MilliQ water as medium. The resulting dispersion was stirred during one hour and then allowed to deposit for 24h. 2 ml of this dispersion was placed on top of the ZnS crystal and allowed to dry under vacuum overnight in order to form a dry and stable hematite film. The film was then rinsed with water to eliminate the excess of hematite particles that did not adhere well to the crystal. The dispersion and the hematite film on a ZnSe crystal is shown in Figure 2.
Figure 2. Suspension of a-Fe2O3 and the corresponding film on the crystal
2.2.2. Spectra of adsorbed dicarboxylic acid
Spectra of adsorbing dicarboxylic acid were recorded in-situ using an aqueous solution of the acid continuously flowing through the cell (5 ml/min) and a Bruker IFS 113V spectrometer equipped with DTGS detector (Figure 3). Desorption experiments were carried out directly after the adsorption experiments, using MilliQ water instead of the aqueous dicarboxylic acid flowing through the cell (5
ml/min). A background spectrum (256 scans, 4 cm−1 resolution) was recorded with the oxide layer in contact with 100 mL of MilliQ water at the same pH as the sample solution. The adsorption/desorption reaction was monitored by recording absorbance spectra (256 scans, 4 cm−1 resolution) at each reaction time interval. The pH of the solution was varied by titration with 0.1M NaOH or HCl. Data acquisition as well as data processing and analysis was carried out by means of OPUS software.
tion) was recorded with the oxide layer in contact with 100 mL of MilliQ water at the same pH as the sample solution. The adsorption/desorption reaction was monitored by recording absorbance spectra (256 scans, 4 cm−1 resolution) at each reaction time interval. The pH of the solution was varied by titration with 0.1M NaOH or HCl. Data acquisition as well as data processing and analysis was carried out by means of OPUS software.
Pump
Stirrer Solution
Flow cell of stainless steel, 3 ml.
IRE, 50 x 20 x 2 mm
Incident light IR-spectrometer under vacuum
p-pol.
s-pol.
Incident light To detector
IRE = Internal Reflection Element Adsorbate Pump
Stirrer Solution
Flow cell of stainless steel, 3 ml.
IRE, 50 x 20 x 2 mm
Incident light IR-spectrometer under vacuum
p-pol.
s-pol.
Incident light To detector
IRE = Internal Reflection Element Adsorbate Pump
Stirrer Solution
Flow cell of stainless steel, 3 ml.
IRE, 50 x 20 x 2 mm
Incident light IR-spectrometer under vacuum Pump
Stirrer Solution
Flow cell of stainless steel, 3 ml.
IRE, 50 x 20 x 2 mm
Incident light
Flow cell of stainless steel, 3 ml.
IRE, 50 x 20 x 2 mm
Incident light IR-spectrometer under vacuum
p-pol.
s-pol.
p-pol.
s-pol.
Incident light To detector
IRE = Internal Reflection Element Adsorbate
Figure 3 . Schematic picture of the experimental set up20
Figure 3 . Schematic picture of the experimental set up20
3. Results and discussion 3. Results and discussion
3.1. General characteristics of the synthesized sample 3.1. General characteristics of the synthesized sample
X-ray diffraction indicated that calcined sample was a well-crystallized hematite, according to the 104,110,113,024,116,214, and 012 reflections, Very weak and not well defined lines also appeared
X-ray diffraction indicated that calcined sample was a well-crystallized hematite, according to the 104,110,113,024,116,214, and 012 reflections, Very weak and not well defined lines also appeared before calcination, which indicated weak crystallinity (Figure 4).
before calcination, which indicated weak crystallinity (Figure 4).
10 20 30 40 50 60 70 80 90 0
50 100 150 200 250 300
Intensity (300)(214)(116)
(024)
(113)(110)
(104)
012
A B
2θ / °
Figure 4. XRD pattern of sample at different temperature
Figure 5 . SEM image of the hematite sample
Figure 5 shows the SEM picture of calcined iron oxide, which indicated that the particle size was
~40 nm after calcination. DIRFT spectra were also typical of hematite after calcination, showing characteristic absorption bands at around 480 and 555 cm−1 (Figure 6).
4 0 0 0 3 6 0 0 3 2 0 0 2 8 0 0 2 4 0 0 2 0 0 0 1 6 0 0 1 2 0 0 8 0 0 4 0 0
Transmittance[%]
W a v e n u m b e r [ c m - 1 ] A
B
Figure 6 . DIRFT spectrum of sample
Nitrogen adsorption-desorption isotherms for the samples and corresponding pore size distribution are shown in (Figure 7). The isotherms can be classified as being of type V with hysteresis, which is characteristic of mesoporous materials. The sample show a narrow Barret–Joyner-Halenda (BJH) 21-22 pore size distribution centered at 5.5 nm.
0 50 100 150 200 250 300 350
0. 00E+00 2. 00E- 01 4. 00E- 01 6. 00E- 01 8. 00E- 01 1. 00E+00 1. 20E+00
0. 00E+00 5. 00E- 01 1. 00E+00 1. 50E+00 2. 00E+00 2. 50E+00 3. 00E+00 3. 50E+00
0 5 10 15 20 25 30 35 40
Figure 7 . N2 adsorption/desorption isotherms and pore size distribution
3.2. Adsorption kinetics
Figure 8 shows the ATR-FTIR spectra of aqueous succinic acid (50 mM) as function of pH. The band at (1718) cm-1 is commonly assigned to a carbonyl stretching vibration VC=O23 which disappeared gradually as a result of the deprotonation of succinic acid as pH increased. The intensity of two strong bands at 1553 cm-1 and 1394 cm-1 increased as pH increased and could therefore be assigned to the asymmetric (VasCOO-) and symmetric (VsCOO-) vibrations, respectively. This change is in accordance with the pKa values for the dicarboxylic acids, as shown in Figure 9. The dominant species at pH 3, is protonated succinc acid (H2Suc). Hydrogen deprotonated succinct (HSuc-) is dominant at pH 5, and succinate (Suc2-) is dominant in the basic reagion ( pH 8).
- 0. 01 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06
1200 1400
1600 1800
Wavenumber ( cm- )
Absorbance
PH=3 H2Suc PH=5 HSuc- PH=8 Suc2-
1718 Vc=o
1553 Vascoo-
1295δ CH2 1416δ CH2
1394 Vscoo-
Figure 8 . Spectra of aqueous succinic acid species (50 mM)
2 4 6 8 10 12
0. 0 0. 2 0. 4 0. 6 0. 8 1. 0
Fraction
pH H6C4O4
[H4C4O42−]TOT = 10.00 mM
H4C4O42−
H5C4O4−
Figure 9 Distribution of succinic acid as a function of pH (pK1= 4.2 pK2 = 5.64)
- 0. 01 0 0. 01 0. 02 0. 03 0. 04 0. 05
1350 1450
1550 1650
Wavenumber ( cm- 1)
Absorbance
0 uM 5 uM 10 uM 20 uM 50 uM 100 uM 500 uM 1000 uM 2000 uM
(a)
0 5 0 0 1 0 0 0 1 5 0 0 2 0 0 0
0 .0 0 .5 1 .0 1 .5 2 .0
Integrated Absorption
C o n c e n tra tio n (u M )
(b)
Figure 10. Adsorption spectra as a function of concentration and the corresponding isotherm using the integrated intensities are shown.
Figure 10a shows the spectra of succinic acid adsorbed on hematite with increasing succinic acid concentration. Before each increase in concentration, we allowed equilibration for 90 min, in order get spectra representing equilibrium conditions. Figure 10b shows the corresponding isotherm, which clearly shows that the data closely fit the Langmuir type of adsorption. The Langmuir equation is expressed here as Γ=ΓmaxC/(1+KC)
where K = Langmuir´s equilibrium constant or distribution coefficient, c = aqueous acid concentration, Γ = amount adsorbed at a certain concentration representated by the integrated intensity, and Γmax = maximum amount adsorbed as represented by the integrated intensity. The evaluation of the Langmuir isotherm resulted in K=0.258 and Γmax=0.52.
Figure 11a shows the effects of initial bulk concentrations on the kinetics and the corresponding kinetic models at pH 5. The integrated absorbance at 1441cm-1 was assumed to be proportional to the amount adsorbed and was plotted versus time. All curves have similar characteristics, showing a very fast initial process (0-10 min) followed by a slower process (10-60 min). The adsorption rate increases as the initial concentration increased, so the lower concentration needs longer time to reach equilibrium. A pseudo-first order rate of reaction was assumed and as shown in Figure 11b,c,d , a fairly good fit to a linear curve is obtained at the lowest concentration, which means that the rate is dependent on the available surface sites and initial concentration.
0 0. 2 0. 4 0. 6 0. 8 1 1. 2 1. 4 1. 6 1. 8 2
0 20 40 60 80 100 120 140
Ti me( mi n)
Integrated Absorbance
10 uM 100 uM 500 uM
(a)
0 20 4 0 6 0 80 1 00
0 2 4 6 8
ln(Amax/(Amax-A))
time(min) Y=-0.39115+0.08638X 10 uM
(b)
0 2 0 4 0 6 0 8 0 1 0 0 0
1 2 3 4 5 6
time(min)
ln(Amax/(Amax-A))
Y =0.53037+0.05512X 100 uM
(c)
0 20 40 60 80 100 120
0 1 2 3 4 5 6
ln(Amax/(Amax-A))
time(min) Y=0.87114+0.04112X
500uM
(d)
Figure 11 Adsorption kinetics at different initial bulk concentration and Plot of
ln[Amax/(Amax-At)] versus time,according to a pseudo first order rate of reaction (a) integrated
values (b) 10 uM (c) 100 uM (d) 500 uM
However, at higher concentrations it becomes clear that the kinetics is more complex and that two rate constants are necessary to explain the adsorption behavior. The initial part of the reaction seems to be represented by one constant and the latter part by another rate constant.
Figure 12 shows how the zeta potential changed as a function of adsorbed succinic acid concentration. The surface charge changed from positive to negative as the concentration increased.
As the amount of adsorbed molecules at the surface increased, it is expected to reduced the initially positive surface charge, which in turn should affect the kinetics.
- 20 - 15 - 10 - 5 0 5 10 15 20
- 200 200 600 1000 1400 1800 2200
Concent r at i on( uM)
Zeta(mV)
Figure 12 Zeta potential of hematite as function of adsorbed succinic acid concentration
In order to further investigate the electrostatic influence on the kinetics, experiments were performed at different pH values. Figure13 shows the pH effect on the adsorption kinetics at a constant initial succinic acid concentration (5*10-4M).
0 0. 3 0. 6 0. 9 1. 2 1. 5 1. 8 2. 1
0 20 40 60 80 100 120 140 Ti me( mi n)
Integrated Absorbance
PH=3PH=5 PH=8
Figure 13 Adsorption kinetics of succinic acid on hematite at different PH values
The results show that, at a given time, the adsoption rate and the amount adsorbed is highest at pH 5. At this pH, the hydroxyl group in HSuc- has a strong electrostatic interaction compared with the situation at pH 3 (neutral H2Suc). In addition, at pH 8, the deprotonated and negatively charged surface and negative Suc2- species have difficulty in forming chemical bonds due to the repulsion, which imply that the adsorbed amount decreases, as does the rate of adsorption. From the above results, it is evident that the electrostatic repulsion plays an important role in the adsorption process.
3.3. Desorption kinetics
Figure 14 shows the typical time sequences of infrared spectra for succinic acid adsorbing and desorbing on the hematite surface at pH 5 and its corresponding integrated absorbance at 1441cm-1. In Figure 14a, the intensity of the IR bands increases as the reaction time increases, indicating an increased HSuc- adsorption. In Figure 14b the intensity decreases with reaction time, indicating the HSuc- desorption from the surface by the flow water. It also shows that the desorption need much longer time to reach equilibrium compared with the adsorption in Figure 14c. This may be expected since the rate constant for adsorption is usually much larger than the corresponding constant for desorption. On the other hand, the concentration of species at the iron.oxide surface is much higher than the concentration in solution because of the small volume of species at the surface. The desorption rate is also higher in the beginning and then goes slowly to an equilibrium value. This strongly indicates that physisorbed molecules forming outer sphere complexes through hydrogen bonding desorb first and that molecules forming inner-sphere complexes by forming chemical bonds with the surface are more strongly bonded and therefore leave the surface first. This experimental fact was illustrated by the desorption spectra in Figure 14b. The difference between chemically bonded
(inner-sphere) and hydrogen bonded (outer-sphere) complexes is most clearly seen the behavior of the antisymmetric stretch of the carboxylate group at 1553 cm-1. With increasing time, the intensity of the band at 1553 cm-1 decreases whilst the shoulder on the high frequency side of this band remains upon desorption (shoulder at 1590 cm-1).
PH=5 500 uM adsor pt i on
- 0. 01 0 0. 01 0. 02 0. 03 0. 04
1350 1450
1550 1650
Wavenumber ( cm- 1)
Absorbance 0 mi n
1 mi n 2 mi n 5 mi n 15 mi n 25 mi n 50 mi n 90 mi n
(a)
PH=5, 500 uM desor pt i on
- 0. 01 0 0. 01 0. 02 0. 03 0. 04
1350 1450
1550 1650
Wavenumber ( cm- 1)
Absorbance
0 mi n 5 mi n 15 mi n 30 mi n 60 mi n 120 mi n 240 mi n 420 mi n 540 mi n 660 mi n 750 mi n 780 mi n
(b)
0 0. 005 0. 01 0. 015 0. 02 0. 025 0. 03 0. 035 0. 04
0 100 200 300 400 500 600 700 800 900 1000 Ti me( mi n)
Absorbance
adsor pt i on
desor pt i on
(c)
Figure 14 Time series of IR spectra for adsorption-desorption and its corresponding intergrated values at PH 5; (a) adsorption spectra, (b) desorption spectra, and (c) itegrated values
Similar spectral behavior is also observed at pH 3 and 8. Figure 15 shows desorption spectra and its normalized spectra at pH 3. There is a very obvious difference in the shape of normalized curves as the time increases, suggesting that a new surface species is formed by a chemical reaction. Acoording to Seunghun Kang and Baoshan Xing，the bands at 1553 cm-1 and 1590 cm-1 were assigned to the υascoo-outer sphere and Vascoo- inner sphere complexes, respectively. An interesting result is that the same behavior does not show up for the symmetric COO- vibration at 1394 cm-1. The exact reason for this is not clear at present, but we propose that the surface complexes changed from initially being outer-sphere to inner sphere complexes at a later point of time suggesting that the structure of the surface layer changes with surface coverage. According to Alexander and Erdogan24, there is some association of the aliphatic chains and the contribution from the chain-chain interaction may be large enough to change a surface complex from outer-sphere to inner-sphere. A similar trend observed by Seunghun Kang and Baoshan Xing’ , who claimed that the structure could changed from outer sphere to inner sphere during the drying process. However, these authors did not show corresponding adsorption reactions with maleic acid as adsorbate, which will be discussed in the following part of this exam work.
PH=3 desor pt i on
- 0. 01 0 0. 01 0. 02 0. 03
1350 1450
1550 1650
Wavenumber ( cm- 1)
Absorbance
0 mi n 5 mi n 10 mi n 15 mi n 25 mi n 40 mi n 100 mi n 160 mi n 220 mi n 280 mi n 340 mi j n
1591 1552
Vascoo- i nner spher e Vascoo- out er spher e
(a)
- 0. 1 0. 4 0. 9 1. 4 1. 9
1250 1350
1450 1550
1650
1750 Wavenumber s( cm- )
Normalized Absorbance
5 mi n 340 mi n v
1591 1552
(b)
Figure 15 Time series of IR spectra for desorption and its normalized spectra at PH 3: (a) desorption spectra (b) Normalized spectra
Figures 16a and 16b show infrared spectra from the desorption of maleic acid at pH 8 in comparison with the corresponding spectra for succinic acid. In addition the desorbtion of the two adsorbates at pH3 is compared in Fig. 16c. As evident from Fig. 16c, the rate of desorption is larger for maleic acid as compared with succinic acid. This indicates that succinic acid forms more of inner-sphere complexes, whereas maleic acid forms more of outer-sphere complexes. As shown by the comparison of Figures 16a and 16b, the band shift with time in a) did not show up in b). According to Buckland et al., the -CH=CH- alkenyl segments in the hydrocarbon chains may promote orientations
of the chains which hinder chemisorption. Not only the rate of desorption is larger for maleic acid but so is also the rate of adsorption. This implies that outer-sphere complexes in this case are more rapidly adsorbed but also more rapidly desorbed. In turn this behavior strongly supports the suggestion that succinic acid is also chemisorbed at the iron-oxide surface. Similar results were also obtained at pH 3 and 5, although they are not shown here.
PH=8, succi ni c aci d desor pt i on
- 0. 01 - 0. 006 - 0. 002 0. 002 0. 006 0. 01 0. 014 0. 018
1350 1450
1550 1650
Wavenumber ( cm- 1)
Adsorption 0 mi n
5 mi n 60 mi n 140 mi n 200 mi n 380 mi n 500 mi n 560 mi n 660 mi n
(a) PH=8 mal ei c aci d desor pt i on
- 0. 01 0 0. 01 0. 02 0. 03
1250 1350
1450 1550
1650 1750
Wavenumber ( cm- )
Absorbance
0 mi n 5 mi n 10 mi n 15 mi n 30 mi n 60 mi n 90 mi n 150 mi n 210 mi n 270 mi n 300 mi n
(b)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 100 200 300 400 500 600 700 Time(min)
Integrated Absorbance
Mal ei c aci d Succi ni c aci d
Adsor pt i on Desor pt i on
(c)
Figure 16 Time series of IR spectra for two kinds of acid desorption and its corresponding integrated values at PH 3: (a ) Succinic desorption (b) Maeic desorption (c) Integrated values for maleic acid (blue) and succinic acid (red).
3.4. Surface complexation mechanisms
Based on kinetics adsorption-desorption studies of the two kinds of dicarboxylic acids, it was shown that they have different adsorption –desorption behaviour and surface complexation structure depending on the pH range. Possible surface complexation mechanisms have been proposed, as shown in Figure 17. According to Deacon25 et al., the surface structure can be assigned depending the positions of the asymmetric and symmetric carboxylate stretching vibrations, υas(COO-) and υs(COO-), and the separation between these vibrations,∆υas-s. By comparing ∆υas-s=(υas(coo-)-υs(coo-)) measured on metal complexes and ∆υas-s(ionic) of species in aqueous solution, they gave the following correlations:
∆υ as-s >∆υ as-s(ionic) monodentate bidentate
∆υ as-s <∆υ as-s(ionic) bidentate chelating or bridging
∆υ as-s <<∆υ as-s(ionic) bidentate chelating
Our spectral data show that the surface strcture is a complex with bridging bidentate sturctrue, as indicated by the small ∆υ as-s for the adsorbed succinic acid species, which is consistant with Dobson and McQuillan’s results9. The results indicated that the inner sphere and outer sphere complexation simultaneously show up at the three pH values, and the difference is that their proportion is different at different pH value. By comparing their desorption spectra at three pHs, we concluded that outer-sphere adsorption was dominant at high pH and inner sphere adsorption was dominantat at low pH.
O H2
_ _OH
_OH
(a) succinic acid adsorbed on protonated hematite at pH<pHisosteric
Fe +
OH OH +
2
Fe OH
OH +
C O
O H O
H O
C
C Fe
Fe Fe
...H O Fe
C O
O H O
H
C
C Fe
Fe
...H O Fe
...H O Fe O
2 2
C (CH2)
2 C
O
O O
O_ _
C (CH2)
2 C
O O
O_
OH
Fe OH
+
Fe OH
OH + C C
O
O O
O_ _
OH+ 2
C C C
O O
O_
OH
O H2
C O
O H O
H
O C C
HH2O
H2OHO Fe
2
C O
O H O
H
O
O O H
H ...
...
C C
Fe
2 Fe
H-bonded
H O
... Fe
+
(b) succinic acid adsorbed on hydroxy hematite at pH>pHisosteric
(c) maleic acid adsorbed on protonated hematite at pH<pHisosteric
(d) maleic acid adsorbed on hydroxy hematite at pH>pHisosteric
H CH
CH CH
Figure 17 A possible surface complexation mechanism for two dicarboxylic acid :(a) and (b) Succinc acid adsorbed on hematite via a chemisorption reaction (c) and (d) maleic acid adsorbed on hematite via physisorption reaction
At pH 3 and 5, the hematite surface is protonated, giving rise to a net positive charge on the surface. Under these conditions, the different dominate species of succinic acid have different electrostatic intereaction, which determines that the water abstraction or hydroxyl abstraction reactions are taking place. Results from maleic acid adsorption experiments verified our conclusions.
At higher pH 8, because of the negatively charged surface, there is no net positive surface charge that can be used to electrostatically adsorb the carboxylate, suggesting that contributions from hydrogen bonded species are possible16, as shown in Figure 17. In addition, the adsorbed succinic acid molecules are not desorbed completely even after more than ten hours in sharp contrast with maleic acid, suggesting that the adsorption is irreversible for the initially adsorbed succinic acid molecules and that the water abstraction reaction is favored at low pH.
4 Conclusions
We have, for the first time shown the adsorption-desorption kinetics of dicarboxylic acid at the hematite/water interface using the in situ ATR-FTIR technique. Both adsorption and desorption occurs in two different processes: a fast one that is completed in less than 10 minutes, and a slow one that takes several hours. The adsorption rate increases by increasing the concentration, and at pH 5 the adsorbed amount and rate of adsorption is higher than at pH 3 and 8. This indicated that the electrostatic reaction plays an important role on the adsorption process.
All these desorption experiments performed, allowed us to distinguish between irreversibly adsorbed structures of the initial adsorbed molecules (inner-sphere complexation) and the reversible structures of later adsobed ones (outer-sphere complexation through hydrogen bonding). By comparing with the maleic acid, we could definitely conclude that the adsorption type for succinic acid belongs to chemisorption. The results of this study clearly show that the inner-sphere and outer- sphere complexation existed simultaneously at all pHs for the succinic acid adsorption on the hematite surface. ATR-FTIR has shown to be an excellent technique to monitor the adsorption kinetics of succinic acid and maleic acid on hematite.
These results from this exam work show that ATR in combination with in-situ experiments is a very promising approach in order to gain insight into surface complexation mechanisms. The temperature effect on the kinetics of adsorption and the orientation of species adsorbed on hematite surfaces is underway utilizing the in-situ ATR-FTIR method.
5 Future work
It seems that adsorption measurements at different temperatures together with related calorimetric data can provide useful information on the dominant factors in the adsorption process. The number of published thermodynamic data for adsorption of non-electrolytes from dilute solution is rather small, and free energy, enthalpy, and entropy data have been calculated from adsorption isotherms over a limited range of temperatures only. So, a systemic study the effect of temperature on adsorption–desorption behavior is therefore valuable.
The -CH=CH- alkenyl segments in the hydrocarbon chains may effect the orientation of adsorbed molecules on the surface. Accordingly, the structure and orientation of dicarboxylic acids at the hematite/water interface should be further studied using polarized ATR-FTIR spectroscopy.
Furthermore, the reaction kinetics of the adsorption of the two carboxylic acids should be modeled and a reasonable mechanism suggested involving the elementary reaction steps. This study could be extended to encompass other adsorbates such as silicates.
6 ACKNOWLEDGEMENTS
First of all I would like to greatly acknowledge Associate Professor Allan Holmgren for helpful guidance and providing ideas during this work. I will also thank you for the fruitful discussions we have had during the work.
I woud also like to thank research Engineer Maine Ranheimer and Dr Mats Lindberg for the laboratory and computer help.
I want to thank my friends and colleagues Payman Roonasi, Xiaofang Yang, Ivan Carabante, Anamaria Vilinska, Tingting Zheng, for their help on the instrument operation.
I also want to thank Prof. Oleg N Antzutkin for his good suggestions.
There are many people at the department of Chemical Engineering and Geosciences I want to thank, so THANK YOU ALL.
Finally, I would like to take the opportunity to send my best gratitude to my supervisor Prof.
Zhongxi Sun, for giving me this opportunity to study here. He always gave me support and guidance during this work through E-mail.
The financial support from the division of Chemistry is gratefully acknowledged.
7 References
[1] Zeng, H.; Li, J.; Liu J. P.; Wang, Z. L.; Sun, S. H. Nature 2002, 420,395.
[2] Doyle, P. S.; Bibette, J.; Bancaud, A.; Viovy, J. L. Science 2002, 295, 2237.
[3] Bandara, J.; Mielczarski, J. A.; Kiwi, J.Appl. Catal. B. 2001, 34, 307.
[4] (a) Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J. M. Nature 2000, 407, 496.
(b) Rolison, D. R. Science 2003, 299, 1698.
(c) Wang, L. Y.; Han, L.; Luo, J.; Zhong, C. I. Abstr. Pap. Am. Chem. Soc. 2004, 228, 473;
[5] Dolamic, I.; Bürgi, T. Journal of Catalysis 2007, 248 268
[6] Johnson, S. B.; Yoon, T. H.; Kocar, B. D.; Brown, G. E. Langmuir 2004,20, 4996.
[7] Angove, M. J.; Fernandes, M. B.; Ikksan, J. J. Colloid Interface Sci. 2002,247, 282.
[8] Axe, K. Vejgården, M. Persson, P. Journal of Colloid and Interface Science 2006, 294,31.
[9] Degenhardt. J, McQuillan A.J., Chem. Phys. Lett. 1999, 311 179.
[10] Duckworth, O.W.; Martin S.T., Geochim. Cosmochim. Acta 2001, 65, 4289.
[11] Johnson, S.B. ; Yoon ,T.H. ; Kocar, B.D.; Brown, G.E. Langmuir 2004, 20, 4996.
[12] Kubicki, J.D.; Schroeter, L.M.; Itoh, M.J. ; Nguyen, B.N.; Apitz ,S.E. Geochim. Cosmochim.
Acta .1999, 63, 2709.
[13] Specht, C.H.; Frimmel, F.H. Phys. Chem. Chem. Phys. 2001, 3, 5444.
[14] Rosenqvist, J.; Axe, K.; Sjoberg, S.; Persson, P. Colloid Surf. A Physicochem. Eng. Asp. 2003, 220, 91.
[15] Johnson, B.B.; Sjoberg, S.; Persson, P. Langmuir 2004, 208, 23.
[16] Hug, S. J.; Bahnemannb, D. Journal of Electron Spectroscopy and Related Phenomena 2006,150,208
[17] Kang, S. H. ; Xing, B.S. Langmuir 2007, 23, 7024
[18] Nordin, J.; Persson, P.; Laiti, E.; Sjoberg, S. Langmuir 1997, 13, 4085
[19] Boily, J. F.; Persson, P.; Sjoberg, S. Geochim. Cosmochim. Acta 2000, 64, 3453.
[20] Larsson, M.L. ; Fredriksson, A. ; Holmgren, A. J. Colloid Interf. Sci. 273 (2004) 345–349.
[21] Barrett, E.P.; Joyner, L.G; Halenda, P.P. J. Am. Chem. Soc., 1951, 73, 373.
[22] Kruk, M.; Jaroniec , M.; Sayari, A.. Langmuir, 1997, 136, 267.
[23] Szaraz, I.; Forsling, W. Langmuir 2001, 17, 3987.
[24] Couzis, A; Gulari, E. Langmuir, 1993, 9, 3414
[25] Deacon, G.B.; Phillips, A.. J.; Coord. Chem.Rev.1980, 33, 227
