close

Вход

Забыли?

вход по аккаунту

?

Microwaves and sorption on oxides: Surface temperature and adsorption selectivity investigation

код для вставкиСкачать
MICROWAVES AND SORPTION ON OXIDES: SURFACE TEMPERATURE AND
ADSORPTION SELECTIVITY INVESTIGATION
A Dissertation Presented
by
STEVEN JAMES VALLEE
Submitted to the Graduate School of the
University of Massachusetts Amherst in partial fulfillment
of the requirements for the degree of
DOCTOR OF PHILOSOPHY
February 2008
Chemical Engineering
3315479
2008
3315479
© Copyright by Steven James Vallee 2008
All Rights Reserved
MICROWAVES AND SORPTION ON OXIDES: SURFACE TEMPERATURE AND
ADSORPTION SELECTIVITY INVESTIGATION
A Dissertation Presented
by
STEVEN JAMES VALLEE
Approved as to style and content by:
______________________________
W. Curtis Conner, Jr., Chair
______________________________
Scott Auerbach, Member
______________________________
K. Sigfrid Yngvesson, Member
_______________________________________
T. J. Mountziaris, Department Head
Chemical Engineering
ACKNOWLEDGEMENTS
Thanks to Prof. Curt Conner, Prof. Robert Laurence, Prof. Sigfrid Yngvesson,
Prof. Scott Auerbach, and Geoff Tompsett for their discussions. Thanks to Kan Fu, Fan
Lu, and Kyu-Ho Lee for providing permittivity measurements. Thanks to Gerald Ling for
having assembled the piping and flow controllers for the multi-component flow
adsorption system. Thanks to Karl Hammond for writing the computer programs that
collect the mass spectrometer data using labview and convert the data into an excelcompatible format. Funding was provided by a NSF NIRT Nanoscale Interdisciplinary
Research Team grant. Also, thank you to my wife Lisa for all her love, support, and
understanding while I have pursued my academic career.
iv
ABSTRACT
MICROWAVES AND SORPTION ON OXIDES: SURFACE TEMPERATURE AND
ADSORPTION SELECTIVITY INVESTIGATION
FEBRUARY 2008
STEVEN JAMES VALLEE, B.S., WORCESTER POLYTECHNIC INSTITUTE
Ph.D., UNIVERSITY OF MASSACHUSETTS AMHERST
Directed by: Professor William Curtis Conner, Jr.
Microwave heating is not the same as conventional heating, and it is believed that
this difference, the “microwave effect,” may be interpreted to be due to selective local
heating. The temperature at the surface where sorption occurs is “effectively” greater
than the measured solid or gas temperature. In these studies, measurements of the amount
adsorbed as functions of the partial pressure of a specific adsorbate in the presence of
microwave irradiation were related to conventional adsorption isotherms. Equating the
adsorbate pressure required to achieve a specific coverage (an isostere) in the presence of
microwave irradiation to the amount adsorbed for a conventional isotherm allowed for an
estimate of the “effective” surface temperature in the presence of microwaves. It was
found that the effective surface temperature increased when using adsorbates having a
significantly higher permittivity or when increasing the microwave power. The
implication of this change in the surface energy for specific species in the presence of
microwaves is discussed.
It was hypothesized that the adsorption selectivity in the presence of microwaves
is primarily dependent on the permittivity of the adsorbates, while selectivity is
dependent on the heat of adsorption under conventional heating. Sorption experiments
were carried out using a flow based dual-component adsorption system measuring
v
changes in the amount adsorbed with conventional heating and using microwave heating
at 2.45 and 5.8 GHz. The adsorption selectivity as a function of microwave frequency
was examined for a case in which the adsorbates have an opposite dependence of
permittivity with frequency (isopropanol had a greater permittivity than acetone at 2.45
GHz, and acetone had a greater permittivity than isopropanol at 5.8 GHz). It was found
that microwave energy could influence sorption differently than conventional heating.
Differences in the adsorption selectivity were not as great as expected based on the bulk
liquid permittivities due to the miscibility of the components. The permittivities of the
adsorbates in the adsorbed phase at low surface coverage may be different than that of
their respective bulk liquids. The smaller than expected change in adsorption selectivity
with microwave frequency might also be attributed to the miscibility of acetone and
isopropanol.
vi
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS................................................................................................iv
ABSTRACT...………………………………………………………….………………….v
LIST OF TABLES…………………………………………………………….......………x
LIST OF FIGURES…………………………………………....…………………..……..xi
LIST OF SYMBOLS…..………………………………….…...……………...……...…xvi
CHAPTER
1
INTRODUCTION…...…………………………………………………………....1
1.1
2
Sorption with In-situ Microwaves…………………………………………….2
BACKGROUND……………………………………………………………….…5
2.1
Microwave Heating and Permittivity………………………………………….5
2.2
Influence of Temperature and Microwaves on Sorption…………………...…6
2.3
Changes in IR Spectra for Oxides and Sorbates after Exposure to
Microwaves……………………………………………………………………9
2.4
Plasma and Microwave Induced Plasma Formation……………………....…12
3
LITERATURE REVIEW………………………..………………………………14
3.1
Advancements in Sorption Analysis………………………………………....14
3.2
Sorption in the Presence of Microwaves………………………………….…16
3.3
Changes in IR Spectra for Oxides with Temperature and Adsorbates………19
3.4
Initial Research…………………………………...........…………….………21
4
EXPERIMENTAL...................................…………..............……...………….…30
4.1
Single Component Adsorption System Apparatus…...……………..…..…...30
vii
4.2
Sample Preparation……………………………………………………..……31
4.3
Sorption Isotherm Measurement………………………………..……………33
4.4
Isosteres in the Presence of Microwaves………………………….…………36
4.5
Multi-component adsorption in the presence of microwaves..........................37
5
RESULTS: SINGLE COMPONENT ADSORPTION IN THE PRESENCE OF
MICROWAVES...……………………………...………………………………..42
5.1
Plasma Formation…………………………………………………………....42
5.2
Cyclohexane (ε” = 0.05) on Aerosil 200 Silica...............................................42
5.3
Dichloromethane (ε” = 0.39) on Aerosil 200 Silica……………...……….…46
5.4
Isopropanol (ε” = 3.19) on Aerosil 200 Silica…………………………….…50
5.5
Methanol (ε” = 13.77) on Methoxylated Aerosil 200 Silica…………………53
5.6
Methanol (ε” = 13.77) on Temperature Treated Methoxylated Aerosil 200
Silica……………………………………………………………………...….57
5.7
IR Spectroscopy of Aerosil 200 Silica and Methoxylated Samples……..…..60
5.8
n-Pentane on Silicalite…………………………………….............................61
5.9
Dichloromethane (ε” = 0.39) on Silicalite……………………………...……64
5.10
Isopropanol (ε” = 3.19) on Silicalite…………………………………………66
5.11
Discussion…………………………………………......……..………………69
5.12
Conclusions.....................................................................................................83
6
RESULTS: COMPETITIVE ADSORPTION......................................................85
6.1
Reactor Loaded with Low Surface Area Glass Beads....................................88
6.2
Methanol and Cyclohexane on Silicalite........................................................91
6.3
Acetone and Isopropanol on Aerosil 200 Silica: Room Temperature
Experiments....................................................................................................95
6.4
Acetone and Isopropanol on Aerosil 200 Silica: Conventional Heating........99
viii
6.5
Acetone and Isopropanol on Aerosil 200 Silica: Microwave Heating at 2.45
GHz and 120 W; and 5.8 GHz and 20 W......................................................100
6.6
Acetone and Isopropanol on Silicalite: Room Temperature Experiments.....104
6.7
Acetone and Isopropanol on Silicalite: Conventional Heating......................108
6.8
Acetone and Isopropanol on Silicalite: Microwave heating at 2.45 GHz and
120W and 5.8 GHz and 20 W........................................................................109
6.9
Methanol and Benzene on Silicalite; Room Temperature Experiments........113
6.10
Methanol and Benzene on Silicalite; Conventional Heating.........................116
6.11
Results Summary and Change in Adsorption Comparison; Accounting for
Changes in Temperature................................................................................119
6.12
Conclusions....................................................................................................126
APPENDICES
A NITROGEN ADSORPTION ISOTHERMS………….................…..............…128
B PERMITTIVITIES OF MATERIALS AS A FUNCTION OF FREQUENCY..131
C LITERATURE VALUES FOR HEATS OF ADSORPTION.............................139
BIBLIOGRAPHY…………………............……………………….......……………….140
ix
LIST OF TABLES
Table
Page
4.2-1
Permittivity of materials at 2.45 GHz……………...…...……..……………..33
4.5-1
Measured permittivities of materials at 2.45 GHz and 5.8 GHz at 22°C........40
5.7-1
Infrared peaks for chemical groups…………………………………………..60
6.2-1
Change in amount adsorbed for methanol and cyclohexane on silicalite........93
6.2-2
Change in amount adsorbed for methanol and cyclohexane on silicalite,
adjusted for changes in temperature................................................................93
6.5-1
Ratio of the amount adsorbed for acetone and isopropanol on Aerosil.........101
6.5-2
Ratio of the amount adsorbed for acetone and isopropanol on Aerosil with the
amount adsorbed during heating adjusted for temperature differences.........101
6.8-1
Ratio of the amount adsorbed upon heating for acetone and isopropanol on
silicalite..........................................................................................................110
6.8-2
Ratio of the amount adsorbed upon heating for acetone and isopropanol on
silicalite, adjusted for temperature differences..............................................110
6.10-1
Change in amount adsorbed for methanol and benzene on silicalite, in
molecules/silicalite unit cell (top) and ratios of the amount adsorbed
(bottom)……………………………………………………………………..117
6.11-1
Changes in amount adsorbed upon and after heating....................................124
6.11-2
Changes in amount adsorbed upon heating, adjusted for temperature..........125
B-1
Permittivity of materials at 2.45 GHz……………..……………...……...…131
C-1
Literature values for heats of adsorption.......................................................139
x
LIST OF FIGURES
Figure
Page
2.3-1
In-situ FTIR of silicalite-1 disc on heating from 25 to 117°C.........................10
3.4.1-1
High-resolution adsorption system…………………………………………..22
3.4.3-1
Setup for studying the effect of microwaves on automotive catalyst……..…24
3.4.3-2
Effect of microwaves on CO lightoff temperature on automotive catalyst….25
3.4.5-1
Experimental setup for competitive adsorption in zeolites…………..………27
3.4.5-2
Competitive adsorption of cyclohexane and methanol on silicalite…………29
4.1-1
Adsorption apparatus………………………………………………………...30
4.5-1
Multi-Component Sorption System.................................................................37
5.2-1
Change in effective surface temperature of cyclohexane on Aerosil 200 silica
as a function of microwave power employed, using interpolation (solid lines)
and the heat of adsorption (dashed lines) to calculate the effective surface
temperature…………..............................................………………….…...…43
5.2-2
Change in effective surface temperature of cyclohexane on Aerosil 200 silica
as a function of microwave power employed…………………………..……46
5.3-1
Adsorption isotherms for dichloromethane on Aerosil 200 silica as a function
of microwave power employed………………………......................………. 47
5.3-2
Change in effective surface temperature of dichloromethane on Aerosil 200
silica as a function of microwave power employed, using interpolation (solid
lines) and the heat of adsorption (dashed lines) to calculate the effective
surface temperature………….................................................……………….48
5.3-3
Isotherms for dichloromethane on Aerosil 200 silica as a function of
microwave power employed with and without gas phase temperature
control…………………………………………………………………….….50
5.4-1
Adsorption isotherms for isopropanol on Aerosil 200 silica as a function of
microwave power employed ………………………………………………...51
5.4-2
Change in effective surface temperature of isopropanol on Aerosil 200 silica
as a function of microwave power employed.……………………………….53
xi
5.5-1
Adsorption isotherms for methanol on methoxylated Aerosil 200 silica as a
function of microwave power employed ……………………………………55
5.5-2
Change in effective surface temperature of methanol on methoxylated Aerosil
200 silica as a function of microwave power employed..…………………....57
5.6-1
Adsorption isotherms for methanol on temperature treated (700 °C)
methoxylated Aerosil 200 silica as a function of microwave power
employed....…………………………………………………………………..58
5.6-2
Change in effective surface temperature of methanol on temperature treated
(700 °C) methoxylated Aerosil 200 silica as a function of microwave power
employed …………………………………………………………………….59
5.7-1
Transmission IR spectroscopy of methoxylated Aerosil 200 silica………….61
5.8-1
Adsorption isotherms for n-pentane on silicalite as a function of microwave
power employed ……………………………………………………………..62
5.8-2
Change in effective surface temperature of n-pentane on silicalite as a
function of microwave power employed …………………………………....64
5.9-1
Adsorption isotherms for dichloromethane on silicalite as a function of
microwave power employed ………………………………………...………65
5.9-2
Change in effective surface temperature of dichloromethane on silicalite as a
function of microwave power employed………………………………….....66
5.10-1
Adsorption isotherms for isopropanol on silicalite as a function of microwave
power employed………………………………………………………...……67
5.10-2
Change in effective surface temperature of isopropanol on silicalite as a
function of microwave power employed ........................................................68
5.11-1
Energy potentials of systems upon exposure to microwaves…………...……69
5.11-2
Change in effective surface temperature versus bulk liquid permittivity for
Aerosil 200 silica (top) and silicalite (bottom)……………..………………..75
5.11-3
Change in effective surface temperature versus coverage (θ) on Aerosil 200
silica (grouped by microwave power); isopropanol: filled diamonds,
dichloromethane: open squares, cyclohexane: filled triangles........................78
5.11-4
Change in effective surface temperature versus coverage (θ) on Aerosil 200
silica (grouped by adsorbate); 30 W: filled circles, 60 W: open squares, 120 W
filled diamonds, 240 W open triangles............................................................79
xii
5.11-5
Change in effective surface temperature versus coverage (θ) on silicalite
(grouped by microwave power); isopropanol: filled diamonds,
dichloromethane: open squares, n-pentane: filled circles .......……………....80
5.11-6
Change in effective surface temperature versus coverage (θ) on silicalite
(grouped by adsorbate); 30 W: filled triangles, 60 W: open squares, 120 W
filled diamonds, 240 W open triangles........……………………...……….....81
6.0-1
Permittivity as a function of frequency for acetone and isopropanol..............87
6.1-1
Helium on glass beads with conventional heating...........................................89
6.1-2
Acetone and Isopropanol on glass beads with conventional heating...............89
6.1-3
Acetone and Isopropanol on glass beads with microwave heating at 2.45GHz
and 120W.........................................................................................................90
6.1-4
Methanol and Benzene on glass beads with conventional heating .................90
6.1-5
Methanol and Benzene on glass beads with microwave heating at 2.45GHz
and 120W.........................................................................................................91
6.2-1
Methanol and cyclohexane on Silicalite with conventional heating................94
6.2-2
Methanol and cyclohexane on Silicalite with microwave heating at 2.45GHz
and120W..........................................................................................................94
6.3-1
Acetone on Aerosil 200 at 24 C.......................................................................97
6.3-2
Isopropanol on Aerosil 200 at 22 C.................................................................98
6.3-3
Change in isopropanol adsorption due to acetone flow on Aerosil 200
Silica................................................................................................................98
6.3-4
Change in acetone adsorption due to isopropanol flow on Aerosil 200
Silica................................................................................................................99
6.4-1
Acetone and Isopropanol on Aerosil 200 Silica with Conventional
Heating...........................................................................................................100
6.5-1
Acetone and Isopropanol on Aerosil 200 with MW heating at 2.45 GHz and
120 W.............................................................................................................103
6.5-2
Acetone and Isopropanol on Aerosil 200 with microwave heating at 5.8 GHz
and 20 W........................................................................................................104
xiii
6.6-1
Acetone on Silicalite at 24 C.........................................................................106
6.6-2
Isopropanol on Silicalite at 24 C....................................................................106
6.6-3
Change in acetone adsorption due to isopropanol flow on Silicalite.............107
6.6-4
Change in isopropanol adsorption due to acetone flow on Silicalite.............107
6.7-1
Acetone and Isopropanol on Silicalite with conventional heating.................109
6.8-1
Acetone and Isopropanol on Silicalite: Microwave heating at 2.45 GHz and
120 W.............................................................................................................112
6.8-2
Acetone and Isopropanol on Silicalite: Microwave heating at 5.8 GHz and 20
W....................................................................................................................113
6.9-1
Methanol on Silicalite at 24 C.......................................................................114
6.9-2
Benzene on Silicalite at 24 C.........................................................................115
6.9-3
Change in methanol adsorption due to benzene flow on silicalite.................115
6.9-4
Change in benzene adsorption due to methanol flow on silicalite.................116
6.10-1
Partial pressures of methanol and benzene on silicalite with conventional
heating............................................................................................................118
6.10-2
Methanol and benzene on silicalite with conventional heating.....................119
A-1
Nitrogen on Aerosil 200 silica……………...…………..……..……………128
A-2
Nitrogen on Silicalite ……...………………..……………..……….………129
A-3
Nitrogen on temperature treated Aerosil 200 silica ……………..........……130
B-1
Permittivity of Aerosil 200 silica versus frequency at 22 C ……..………...132
B-2
Permittivity of silicalite versus frequency at 22 C ………………..……......133
B-3
Permittivity of cyclohexane versus frequency at 22 C …………..…...……134
B-4
Permittivity of dichloromethane versus frequency at 22 C ……...……...…135
B-5
Permittivity of isopropanol versus frequency at 22 C …………......………136
xiv
B-6
Permittivity of methanol versus frequency at 22 C ….………….....………137
B-7
Permittivity as a function of frequency for acetone and isopropanol............138
xv
LIST OF SYMBOLS
∆Hµ
Change in heat of adsorption due to microwaves
∆Hads
Heat of adsorption
∆Tµ
Change in temperature due to microwaves
α(λ)
Absorption coefficient
ε
Permittivity
ε
Emissivity
ε0
Permittivity of free space
ε’ or e’
Real permittivity
ε” or e”
Imaginary permittivity
εr
Relative permittivity
θ
Fractional surface coverage
λ
Wavelength
λD
Debye length
λm
Wavelength at corresponding maximum intensity
σ
Conductivity
σ
Stefan-Boltzmann constant 5.67x10-8 W/m2K4
υm
Frequency at corresponding maximum intensity
ω
Angular frequency
A
Area for heat transfer
c
Speed of light
E
Electric field
Ha
Enthalpy of the adsorbed phase
xvi
Haµ
Enthalpy of the adsorbed phase in the presence of microwaves
Hg
Enthalpy of the gas phase
h
heat transfer coefficient
h
Planck’s constant (6.626x10-34 J/s)
Iλ(t)
Beam intensity at a position through a uniform medium with at a position
(thickness) t and with absorption coefficient α(λ)
Iλ0
Initial beam intensity
Ka
Equilibrium constant for adsorption
Kaµ
Equilibrium constant for adsorption in the presence of microwaves
k
Boltzmann’s constant (1.381x10-23 J/K).
P
Pressure
Pl
Power dissipated
Q
Heat of adsorption
Qµ
Heat of adsorption in the presence of microwaves
Qg
Heat generated by microwaves
R
Gas constant
r
Reflection coefficient
T
Temperature
Teµ
Effective surface temperature in the presence of microwaves
tan δ
Loss tangent
Vm
Monolayer volume
xvii
CHAPTER 1
INTRODUCTION
It has been found that microwave energy uniquely influences sorption on oxides.
Microwave energy induces desorption at temperatures measured in the gas and (bulk)
solid phases that are significantly lower than required for desorption by conventional
heating. It is proposed that the “microwave effect” may be interpreted as due to selective,
local heating [1, 2]. When using oxide sorbents, the temperature at the surface where
sorption occurs is “effectively” greater than the measured solid or gas temperature, since
oxides have a low permittivity and are relatively transparent to microwaves. Different
adsorbates have different capabilities for absorbing microwave energy (permittivities),
resulting in different local temperatures and different sorption selectivities in the presence
of microwaves [1]. This confirms that the effect of microwave heating is local. Energy is
transferred selectively to the surface via the adsorbate. Microwave heating of this surface
region is rapid and the rate of subsequent heat transfer to the solid and gas phases may be
slower. There is a gradient in temperatures: surface > bulk > gas, while the effective
surface temperature has not been quantified until this research.
In this research, measurements of the amounts of adsorption as functions of the
partial pressures of a specific adsorbate in the presence of microwave energy were related
to the conventional adsorption isotherms. By equating the adsorbate pressure required to
achieve a specific coverage (an isostere) in the presence of microwave energy to the
amount adsorbed for a conventional isotherm, an estimate of the “effective” surface
temperatures in the presence of microwave energy were obtained. The resulting
relationships between adsorbents, adsorbates, and other variables were examined.
1
For multi-component adsorption of a binary system in the presence of
microwaves, it is hypothesized that if the oxide adsorbent is largely transparent to
microwaves and the adsorbate pair has one component with a greater permittivity at one
microwave frequency and the other component at a greater permittivity at another
frequency, changing the microwave frequency will influence the selectivity of adsorption.
In this research, sorption experiments were carried out using a flow based dualcomponent adsorption system. Helium streams were passed through bubblers to saturate
them with adsorbate, and were combined with a helium diluent stream to control the
partial pressure below saturation. The flow then passed through a glass reactor bed
packed with oxide adsorbent (silicalite zeolite or Aerosil® 200 Silica), and the effluent
stream from the reactor was analyzed by a mass spectrometer to determine its
composition. For conventional heating experiments, the reactor bed was heated by
heating tape wrapped along the outside of the reactor. For experiments with microwave
heating, the reactor passed through a section of specially constructed waveguide.
Microwaves entered the waveguide from a generator (2.45 or 5.8 GHz) through a coaxto-waveguide transition, made a single pass through the waveguide and the remaining
microwaves were trapped by a water load. From these experiments, the influence of
microwave energy and microwave frequency on sorption selectivity was studied and
compared to conventional heating.
1.1
Sorption with In-situ Microwaves
There is no direct method of probing the surface where microwaves are most
selectively absorbed on oxides. The surfaces of oxides often are terminated in hydroxyl
groups that are more susceptible to absorbing microwave energy than the bulk oxide.
2
Surface temperature measurements in the presence of an adsorbate are particularly
problematic since desorption achieves equilibrium rapidly and is endothermic. It is
documented that systems which comprise adsorbing oxides exposed to microwave
radiation are not isothermal[1]. The surface can absorb energy faster than it can be
exchanged with the underlying solid or gas phases. It is important to characterize the
variations of temperature for all phases of these systems to understand the "microwave
effect". These measurements must be made in-situ, as the variations in temperature due to
exposure to microwave radiation will be transitory. Due to the proposed differences in
temperature of the phases, the sensitivity of the adsorbate, and the rapid cooling due to
desorption, this temperature would be an “effective” surface temperature. The “effective”
surface temperature represents the conditions of physical sorption in the presence of
microwaves.
Measurements of the amounts of adsorption as functions of the partial pressures
of a specific adsorbate in the presence of microwaves can be related to the conventional
adsorption isotherms. By equating the adsorbate pressure required to achieve a specific
coverage (an isostere) in the presence of microwaves to the amount adsorbed for a
conventional isotherm, an estimate of the surface temperature in the presence of
microwaves can be obtained. Due to the proposed differences in temperature of the
phases, the sensitivity to the adsorbate, and the rapid cooling due to desorption, this
temperature would be an “effective" surface temperature. The “effective” surface
temperature represents the conditions of physical sorption in the presence of the
adsorbate.
The oxide adsorbents used were 1) Degussa Aerosil 200 fumed silica (containing
only mesoporosity induced from being pressed in sample preparation), since micropore
3
filling could complicate the analyses, and 2) silicalite zeolite. The adsorbates, 1) npentane, 2) cyclohexane, 3) dichloromethane, 4) isopropanol, and 5) methanol, were
chosen to cover a range of dielectric permittivities, and had appropriate partial pressures
for sorption experiments. Permittivities at 2.45 GHz are shown later in Table 4.2-1, and
as a function of frequency from 0.5-18 GHz in appendix B.
In the case on methanol adsorption on silica in the presence of microwaves, a
surface chemical reaction (methoxylation of the surface) was suspected to take place.
Pre-treating the silica at subsequently higher temperatures could decrease the
concentration of surface hydroxyls, which affected the surface methoxylation.
4
CHAPTER 2
BACKGROUND
2.1
Microwave Heating and Permittivity
Microwaves are alternating current signals of electromagnetic energy with
electrical wavelengths of 1 mm to 1 m. The absorption of microwave energy by a
medium induces some degree of polarization of the medium and conversion of
electromagnetic energy into heat, and is dependent on a property of the medium called its
permittivity, ε, which is divided into real and imaginary parts as described by the
equation
ε = ε’ – j ε”
This is often expressed by the relative to the permittivity, εr, the permittivity of free
space, ε0, the loss tangent, tan δ, the conductivity, σ, and the angular frequency ω by the
equations [3, 4].
ε’ = εr εo
tan δ = (ω ε” + σ) / (ω ε’)
The ability of a molecule to be polarized by an electric field is expressed by the
real part of the permittivity [5]. The imaginary part of the permittivity accounts for loss in
the medium that is converted to heat [6]. Microwaves will have less of an effect on
materials with a lower permittivity. The permittivity is a function of microwave
5
frequency and also temperature.
To study the effects of microwave frequency on adsorption selectivity, the
adsorbate pair of 2-propanol and acetone was used. Shown later in Figure 6.0-1 it can be
seen that the bulk liquid permittivity for isopropanol is greater at 2.45 GHz, and the bulk
liquid permittivity for acetone is greater at 5.8 GHz.
From Poynting’s Theorem, the average dielectric power dissipated in a volume of
a medium is
Pl = (½) ω ε” |E|2
where ω is the angular frequency, |E| is the magnitude of the electric field, and ε” is the
imaginary part of the permittivity of the dielectric material [4].
2.2
Influence of Temperature and Microwaves on Sorption
Gaseous molecules collide with any surfaces to which they are exposed, and they
either reflect from the surface or interact by absorbing or adsorbing. Absorption occurs
when strong interactions change the chemical composition of the solid. Chemical
adsorption occurs when there is a reaction that changes the composition of only the
exposed surface. The most common and weakest type of adsorption is physical
adsorption, and the gas adsorbs onto the surface of the solid without changing the
composition of the solid. The amount of a physical adsorbed species at equilibrium
increases as the temperature decreases; or as the relative pressure (the gas pressure
divided by its vapor pressure) increases. One method of characterization of porous
6
materials is adsorption or desorption (i.e., sorption). The amount of sorption is dependent
on temperature, pressure, surface area, material being adsorbed (adsorbate), and the
adsorbing medium (adsorbent) [7]. Sorption is often expressed as a volume of gas
adsorbed (cm3 at STP) per weight of sample, i.e., cm3/g, or in terms of the relative
surface coverage, θ, the fraction of a monolayer of surface coverage.
As a gas is dosed over a sample, the pressure increases, and then the pressure of
the gas will decrease as some of the gas is adsorbed into the surface, and the pressure will
come to equilibrium. The volume of gas adsorbed as a function of pressure or relative
pressure (pressure divided by its vapor pressure, P\Po) at a constant temperature can be
represented by a sorption isotherm. The majority of isotherms can be classified into one
of the five classes of isotherms originally proposed by Brunauer [7, 8]. Morphological
characteristics such as surface area and pore size distribution can be determined from
physical sorption isotherms. A hysteresis loop occurs when there is a difference between
the adsorption and the desorption isotherms. This occurs for porous oxides at pressures
above a relative pressure of 0.4 for nitrogen at 77K.
The adsorption of species on a surface is exothermic. Consequently, higher
temperatures lead to a smaller amount adsorbed at equilibrium [9]. The temperature
dependence can be related to the heat of adsorption through the Clausius-Clapeyron
equation [10].
(d ln P) / (d (1/T)) = ∆Hads / R
The heat of adsorption (∆Hads) is the amount of heat evolved during adsorption. Physical
7
adsorption heats are typically less than twice that of the heat of vaporization, which is
1.34 kcal/mol for nitrogen at 77K.
Porous materials have the potential to adsorb larger amounts of condensed gas
due to a larger internal surface area and pore volumes. Porous materials are divided into
three categories based on the pore size, microporous (<20 A), mesoporous (20-500A),
and macroporous (>500 A). Zeolites are microporous crystalline solid oxides, and have a
very uniform pore size distribution, and are hence useful for many applications including
catalysis and separations. Many also have a low permittivity. Mesoporous materials
consist of a spectrum of high surface area materials often employed as sorbates or
catalyst supports. In the past decade, some mesoporous materials (MCMs) have been able
to be synthesized having a narrow pore size distribution and other properties similar to
zeolites but with a larger pore size.
Adsorbed species have their own permittivities, and can also modify the dielectric
properties of the surface where they adsorb [2, 11, 12]. In the presence of two or more
sorbates with different permittivities, microwave heating can influence the sorption
properties in a different way than conventional heating [1, 12]. Since the permittivity is a
function of microwave frequency [13-15], it is hypothesized that the frequency also
influences the sorption properties [16]. It has been shown that microwave frequency can
have an effect on catalytic reactions [17].
Microwaves are also used in the synthesis of many ceramics and zeolites in order
to obtain faster crystallization times [18, 19], to lower required heating temperature [18],
and to calcine samples more rapidly [20]. Again, the local temperatures of these
processes are difficult to measure.
8
2.3
Changes in IR Spectra for Oxides and Sorbates after Exposure to
Microwaves
IR spectroscopy is a common method for the characterization of oxide materials.
The bands in IR spectroscopy correspond to the types (degrees of freedom) of movement
of bonds (usually vibrational) in the structure. The changes in the structure of a material
with temperature can be correlated to changes in the bands [21]. Studies of the infrared
spectra transmitted through a sample, reflected from a sample or emitted from a sample
can be directly related to the temperature of the solid (for an isothermal system). The
4
emission is particularly sensitive, as the amount of radiant energy varies as T (for a
blackbody).
In the literature, experiments have been performed using IR spectroscopy to study
the dehydroxylation of silica to observe how the hydroxyl bands change with temperature
[22], and how the bulk of the structure of oxide materials change with temperature [23].
Sorption properties can be studied by IR spectroscopy by observing changes in the bands
with temperature in the presence of a sorbate [21], and by observing the influence of
different adsorbates [24] at a common temperature. Often adsorbed species can be
detected in the infrared spectrum of a solid exposed to the species. The effects of
temperature on IR spectra, the spectra of various types of sorbates, and the conventional
synthesis of mesoporous silicates [25, 26] have been studied using in-situ IR
spectroscopy. The temperature dependence of the IR spectra for silica is shown below in
figure 2.3-1.
9
In-situ FTIR of silicalite-1 disc on heating from 25 to 117°C [27]
8
4
6
Increasing T
0
2
T r a n s m it t a n c e [ % ]
10
Figure 2.3-1
4000
3500
3000
2500
2 0 00
1 5 00
1 0 00
500
W a ve n u m b e r c m - 1
D : \ G e o ff\ s i1 in s it u 1 . 4
D : \ G e o ff\ s i1 in s it u 2 . 0
D : \ G e o ff\ s i1 in s it u 3 . 0
D : \ G e o ff\ s i1 in s it u 4 . 0
D : \ G e o ff\ s i1 in s it u 5 . 0
D : \ G e o ff\ s i1 in s it u 6 . 0
s ilic a l it e
s ilic a l it e
s ilic a l it e
s ilic a l it e
s ilic a l it e
s ilic a l it e
in s i t u
in s i t u
in s i t u
in s i t u
in s i t u
in s i t u
tem
tem
tem
tem
tem
tem
p
p
p
p
p
p
25degC
28degC
3 9 .5 d eg C
61degC
90degC
117degC
pow der K B r
pow der K B r
powder K B r
pow der K B r
pow der K B r
pow der K B r
Temperature °C: 25, 28, 39, 61, 90, 117
•
20 0 2 /1 2 /0 5
20 0 2 /1 2 /0 5
20 0 2 /1 2 /0 5
20 0 2 /1 2 /0 5
20 0 2 /1 2 /0 5
20 0 2 /1 2 /0 5
Absorbance, Reflectance, Transmittance
The absorption of monochromatic light at a wavelength λ can be expressed by
Lambert-Beer’s law [28].
Ι λ (t ) = Ι λ 0 e − α ( λ ) t
Where Iλ(t) is the beam intensity at a position through a uniform medium with at a
position (thickness) t and with absorption coefficient α(λ), and Iλ0 is the initial beam
intensity. Lambert-Beer’s law is valid when α(λ) is constant with respect to t. The
intensity is also reduced by scattering from small particles and is described by theory by
G. Mie.
The fraction of energy absorbed is then
ε (λ ) = (1 − r ) ⋅ (1 − e −α ( λ )t )
Where r is the reflection coefficient.
10
The fraction of energy transmitted is then:
τ(λ) = 1-ε-r
The temperature dependence of the transmitted spectra must arise from the
temperature dependence of the reflection coefficient and the absorption coefficient.
•
Emittance
Emitted blackbody radiation as a function of temperature and wavelength is
described by Planck’s law [28].
I λ (T , λ )dλ =
2πhc 2
λ5
dλ
ch
e λkT − 1
Where h is Planck’s constant (6.626x10-34 J*s), c is the speed of light, and k is
Boltzman’s constant (1.381x10-23 J/K). Assuming ch/λkT is >>1, which is generally a
good assumption, it can be shown below in the Stefan-Boltzman law [28] that radiated
power is
∞
2π 5k 4T 4
F = ∫ I λ dλ =
= σT 4
2 3
15c h
0
Where σ is the Stefan-Boltzman constant (5.67x10-8 W/m2K4). It can be shown
[28] that the intensity has a maximum at a corresponding λm and υm (frequency) and
hν m =
hc
λm
= β kT
Where β = 4.9651 and if λm is in microns, T λm = 2898. At higher temperatures,
the maximum is shifted towards shorter wavelengths.
If the medium is not a blackbody, the emissivity is a function of temperature and
11
the amount of radiation emitted is expressed as [28]
F = ε(T)*σT4
The temperature dependence of emitted spectra will vary with T4, times the
temperature dependence of the emissivity.
2.4
Plasma and Microwave Induced Plasma Formation
Plasma is the most common state of matter in the universe and comprises more
than 99% of all matter in the universe. In a plasma, energy from molecular collisions
overcomes the binding energy of the outermost orbital electrons. A plasma is
macroscopically neutral; the sum of the charge over a length scale greater than the Debye
length λD (the distance of the influence of an individual charged particle in the electric
field) is zero. A plasma also has a number of electrons that are concentrated enough to
exhibit collective particle behavior known as Debye shielding, where particles arrange
themselves to shield any fields created [29].
There are two methods for creating a plasma, photoionization and gas discharge.
In photoionization, molecules absorb photons with greater energy than the ionization
potential of the atom. In gas discharge, an electric field is applied across the ionized gas
accelerating free electrons with enough energy to ionize other atoms by collision.
Plasmas exhibit a number of properties. They are good electrical and thermal
conductors (high electron mobility). They show collective effects due to long range
electromagnetic forces. For weakly ionized species, charge-neutral interactions dominate,
and for strongly ionized species coulomb (charge-charge) interactions dominate. Plasmas
also exhibit a phenomenon known as ambipolar diffusion, where the faster diffusion of
12
electrons creates a field that increases the rate of diffusion for ions, but decreases the rate
of diffusion of electrons so that the rates are about the same. Plasmas also exhibit wave
propagation characterized by a dispersion relation (a function relating polarization to the
wave frequency and wave number), and the wave amplitude is affected by transfer of
energy form the wave to the plasma (or vice versa). Plasmas also emit radiation from
emitting atoms, accelerated charges, and recombination (line spectra).
Plasma formation induced by microwaves takes place through the gas discharge
method. At microwave frequencies the electrons move only a short distance before
changing directions. When the ionizing source is turned off, recombination occurs. For a
plasma to form in the presence of microwaves, the criteria that have to be met are that the
size of the system must be much greater than the Debye length (L >> λD), and the average
distance between electrons must be very small compared to λD, (ne = λD-3) [29, 30]. In
laboratory experiments, plasmas have been produced at pressures from 1 mPa to 1atm,
from system sizes of 0.5 mm to 15 cm, and frequencies of 0.5-10 GHz [30]. As related to
these studies, plasmas could form due to microwave exposure for high permittivity
adsorbates at low relative pressures. If a plasma is formed, it results in a sequence of
reactions that will change the system and obscure the sorption measurements being
conducted.
13
CHAPTER 3
LITERATURE REVIEW
3.1
Advancements in Sorption Analysis
A review of recent developments in adsorption analysis will be discussed in this
section. The basic fundamentals of physical adsorption have been well studied and can be
found in many books, such as Gregg and Sing [7] and Rouquerol, Rouquerol, and Sing
[8].
The effects responsible for the characteristics of porosimetry and nitrogen
sorption data of porous materials are: 1) the geometrical shape of the individual pores, 2)
the relation between voids and throats, and 3) the cooperative percolation effect of the
porous network. While the first effect is well known, Zgrablich et al. [31] developed a
model to take into account the other two effects and experimental data was analyzed.
Muller and Conner [32] used X-ray powder diffraction and FTIR to study the
structural changes in sorbent and zeolitic sorbate due to sorption. ZSM-5 can exist in two
symmetries, monoclinic and orthorhombic, and changes in symmetry can be induced at
elevated temperatures or by adsorption of certain void-filling spaces. Results confirmed
that some sorbate molecules can induce this transition, and the lattice of ZSM-5 should
not be regarded as a rigid crystalline structure.
Maglara et al. [33] developed an adsorption system and procedure for the
determination of adsorption isotherms in zeolites for the characterization of micropores.
Nitrogen and Argon isotherms were obtained over a variety of zeolites at 77K and 87K.
Kaminsky et al.[34] assessed the ability of mean-field-based theories to extract
pore size information from adsorption isotherms of microporous materials. It was found
14
that several key underlying assumptions in the method limit its quality, and that simple
mean-field approaches calculate polydisperse pore size distributions for systems which
are in fact monodisperse.
Groen et al. [35] discussed common pitfalls and limitations in adsorption data
analyses for micro and mesoporous materials. Phenomena such as the tensile strength
effect, adsorbate phase transitions, and monolayer formation from combined micro and
mesoporous materials frequently lead to contributions to the isotherm which are often not
taken into account in models for pore size determination.
Halasz et al. [36] compared the adsorption of polar and non-polar adsorbates on
various Y zeolites. Results showed that type I isotherms typical for micropore adsorption
can turn into type II or type III isotherms depending on both the hydrophobicity of the
sorbent and the polarity of the adsorbate.
Conner et al. [37] examined the stability of adsorption isotherms of mesoporous
materials within the hysteresis area. It was found that sound vibrations (up to 233 Hz) had
little effect, while thermal shock gave shift in pressure within the hysteresis area. The
hysteresis within the adsorption isotherm was determined to be a meta-stable state.
Avery and Ramsay [38] studied the effect of compaction on the specific surface
area and pore structure by measuring adsorption isotherms of nitrogen at 77K. Results
were in accord with the structural model for the compacts consisting of regularly packed
spherical particles with a coordination number controlled by the extent of compression.
The modified Antoine and virial adsorption isotherms are more usefully applied
to predict the isosteric heat of adsorption because they are based solely on pressureexplicit correlations in terms of temperature and loading.
15
Al-Muhtaseb and Ritter [39] improved this method by introducing empirical temperature
dependencies to its fitted parameters.
3.2
Sorption in the Presence of Microwaves
Kobayashi et al. [12] studied the adsorption of a mixture of CFC-113 and water
on NaY zeolite in the presence of microwaves. It was concluded that the desorption of
water became greater with increasing microwave intensity because of its large
permittivity. There was an increase in vacant adsorption sites for adsorbing CFC-113 due
to the desorption of water. The increase in adsorption of CFC-113 due to the increase of
vacant adsorption sites was greater than the decrease in the amount of adsorption of CFC113 due to an increase in temperature. This led to a more selective adsorption of CFC113 in the presence of microwaves.
Guermeur et al. [11] used microwave measurements to study the dielectric
properties of silica powder containing surface silanols groups. The surface polarizability
was recorded during the adsorption of nitrogen at low temperatures. Different silanols
species were related to the different adsorption sites. Characteristic changes were evident
when the sites were filled with an adsorbed molecule. The dependence of the surface
polarizability with temperature was found to be p ≈ (To – T) / T, where To ≈ 100K in the
range 63 < T <100 K.
Sorption of a variety of adsorbates and their mixtures on zeolites under irradiation
by a microwave field was studied by Lopez et al. [16]. The study showed that microwave
irradiation affects sorption more for polar molecules than non-polar molecules, and this
behavior could be used to selectively influence sorption.
16
Roussy et al. [40] experimentally studied the dehydration of 13X zeolite
interacting with microwaves. They concluded that the water molecules are directly
desorbed by the electromagnetic field. The rate of desorption was proportional to the
quantity of water which was locally present, and the square of the applied field strength.
The model used for microwave dehydration did not introduce the heating of the zeolite.
Heggs et al. [41] performed a mathematical analysis of the microwave heating of
a loaded adsorbent pellet and compared it to conventional heating. The model constructed
was general and did not require any particular composition of sorbent or sorbate. For the
constructed model; 1) the desorption rate was dependent on the shape of the isotherm, 2)
microwave heating occurred internally within the pellet, 3) heat and mass fluxes were
concurrent, and 4) temperature controlled the desorption rate with the largest effect at the
center of the pellet. The temperature dominated the desorption process at the start of
heating, and after thermal equilibration, the pressure dominated the desorption process.
Ayappa [42] wrote a review article on modeling heat transfer during microwave
heating, and also Ayappa et al. [43] performed analysis of temperature profiles of
materials with temperature dependent properties exposed to microwaves. The method
illustrated that the temperature dependence of dielectric properties affected the
temperature profile, as compared to a constant property model.
Blanco et al. [44] studied “athermal” effects of microwaves from non-equilibrium
molecular dynamics simulations. Since the microwave period was long compared to
typical thermalization time scales (the time required to achieve thermal equilibrium),
energy distributions in microwave zeolite-guest systems were determined from
microscopic simulations. It was found that microwave heating of binary mixtures in
17
zeolites produced energy distributions that were different from conventional heating.
These theoretical studies were motivated by the experimental studies done by Turner et
al. [1].
Microwaves were used to catalytically activate methane for oligomerization to
higher hydrocarbons. Conde et al. [17] studied the effect of the use of a helium diluent,
microwave power, and frequency of microwave radiation on the field distribution pattern
and product distribution. Changes in selectivities were observed with changes in the use
of a helium diluent and microwave frequency. It was noted that different heating patterns
might occur at different microwave frequencies, which might then influence the
microwave enhancement.
Tang et al. [45] studied the purification of auto exhaust over a noble metal
catalyst using both conventional and microwave heating. The effects of microwave
heating on the lighting-off temperature for CO and operation window were observed. It
was found that microwave heating caused the lighting-off temperature to decrease, and
the activity of the catalyst increased causing the operation window to be enlarged.
Kuroda et al. [46] found that for copper-ion-exchanged ZSM-5 zeolites used to
remove NOx from exhaust streams, the difference in NOx decomposition properties arose
from different occupancies of the copper ions at several sites in the zeolite, which
depended on the ion-exchange method used. They have developed a method using
microwaves that gives more efficient decomposition of NOx.
Nagaraj et al. [47] used a microwave field to enhance phase demixing rates of
two-phase aqueous systems. The microwave-enhanced process decreased the demixing
time 2-4 fold of a polyethylene glycol/potassium phosphate system and 1.5-6.5 fold for a
18
polyethylene glycol/maltodextrin system. The enhanced demixing rate was explained by
the dipole rotation, electrophoretic migration of free salts, multiple reflections at the
interfaces, droplet-droplet collision, and the reduced viscosity of the continuous phase
that occurred during the application of a microwave field.
Ania et al. [48] compared the thermal regeneration of activated carbons using a
conventional electric furnace, and a single mode microwave device. Phenol was used as
the adsorbate and the adsorptive capacity after regeneration cycles of the adsorbents were
evaluated from breakthrough curves. It was found that the porous structure was preserved
better and the adsorptive capacities were higher when using microwaves for regeneration
as compared to conventional heating.
Carott et al. [49] studied the thermal treatment of activated carbon fibers using
microwaves. It was shown that microwaves treatment affected the porosity causing a
reduction in micropore volumes and size, and the microwave treatment modified the
surface chemistry of the activated carbon fibers.
Carbonaceous materials were produced from coffee grounds by microwave
treatment by Hirata et al. [50]. This material was applied to the removal of basic dyes in
wastewater by adsorption, and it was found that the volume adsorbed per gram of sorbate
was greater than for activated carbon materials.
3.3
Changes in IR Spectra for Oxides with Temperature and Adsorbates
Several review articles have been published for the study of surfaces with IR
spectroscopy [51-54]. Some additional studies in this area are given below.
Friedman et al. [21] have designed an infrared cell to investigate materials from
19
196-600 °C, and from vacuum to supra-atmospheric pressures. This was used for
temperature programmed desorption studies of CO on an alumina catalyst. The spectra of
the alumina at various temperatures were also examined. At high temperatures, the
dehydration of the catalyst could be detected from the IR spectrum.
Miecznikowski et al. [23] studied ZSM-5 zeolites with different Si/Al ratios at
room, liquid nitrogen, and liquid helium temperatures by using IR and Raman
spectroscopy, and compared the data to the spectra of silicalite-1.
Diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS) was used
by White et al. [22] to study the temperature dependent dehydroxylation of amorphous
silica. The study of siloxane bridges created by thermally induced hydroxyl condensation
was emphasized along with monitoring the hydroxyl groups.
The interactions of weakly interacting probe molecules with zeolites was studied
by Knozinger et al. [24] by observing the changes of the FTIR spectrum. H- and alkali
cation-exchanged zeolites were used as typical Bronsted and Lewis acids and bases.
Quantum chemical calculations also provided information about the probe molecule
interactions.
Calabro et al. [26] used in-situ ATR and FTIR to monitor the synthesis of M41Stype mesoporous silicate, which enabled the observation of simultaneous changes in both
the organic and inorganic phases of the reaction mixtures. Similarly, Holmes et al. [25]
used ATR and FTIR to monitor the synthesis of MCM-41 in-situ, and were able to
develop a reaction sequence and mechanism of formation.
Samples of Cu-ZSM-5 zeolite having various Cu ratios prepared by a microwave
ion-exchange method were studied by ex situ XPS and FTIR by Chen et al. [55]. The
20
nature of the copper sites were examined by CO adsorption and monitored by using
FTIR.
Pelmenschikov et al. [56] combined IR and ab-initio experiments to study
methanol adsorption on silicalite (an MFI zeolite), and silica. It was found that two
chemisorbed species are present, and the reversible transformation of the two
chemisorbed species to each other is more pronounced on silicalite.
Wovshko et al. [57] studied the dehydroxylation of SiO2, adsorption of methanol
on dehydroxylated silica, and thermal decomposition of methanol on silica using infrared
transmission spectroscopy. It was found that adsorption of methanol on dehydroxylated
silica occurs by first reacting with Si-O-Si sites, followed by exchange with Si-OH
groups at elevated temperatures leading to Si-O-CH3 surface species. CH318OH
adsorption studies demonstrated that adsorption on Si-O-Si sites occurs by cleavage of
the methanol OH bond forming Si-OH and Si-18O-CH3.
3.4
Initial Research
3.4.1
Adsorption Apparatus and High Resolution Adsorption Isotherms
U.S. Patent number 5,637,810 [58] describes an adsorption system built to
measure equilibrium adsorption isotherms at low pressures. This system is ideal for
measurement of equilibrium adsorption isotherms since it incorporates larger diameter
tubing and an improved vacuum system to achieve lower pressures. The time allowed for
equilibrium can be monitored and varied for a series of measurements. This adsorption
system has been used to measure adsorption of nitrogen and argon at liquid nitrogen and
argon temperatures on zeolites and MCM materials [33]. A diagram of the system is
21
shown in Figure 3.4.1-1.
Figure 3.4.1-1 High-resolution adsorption system [56]
22
3.4.2
Measurement of Permittivities and their Frequency Dependence
Kyu-Ho Lee [13] and Fan Lu [59] used a network spectrum analyzer to measure
the dielectric permittivity of zeolite powders and common adsorbates as a function of
frequency from 0.5-18 GHz, and are given in Appendix C. Absorption of microwave
energy is a strong function of frequency and varies by orders of magnitude over the
microwave range of frequencies [4].
3.4.3
The Effect of Microwave Energy on Three-way Automotive Catalysts Poisoned
by SO2
The influence of microwave energy on three-way automotive exhaust catalysis in
the presence and absence of a poison (SO2) was studied by Turner et al. [2]. A bed of
catalyst was placed inside a microwave waveguide. The feed gas compositions were
controlled using mass flow controllers, and the reactants and products were monitored
with an online mass spectrometer. The microwave power used was 300 Watts.
23
Figure 3.4.3-1 Setup for studying the effect of microwaves on automotive catalyst [12]
Legend
MFC - Mass Flow Controller
- Valve
Mass
Spectrometer
Microwave
Catalyst
Waveguide
Helium
MFC 1
He/C3 H8 /CO
MFC 2
He/O2
MFC 3
He/SO2
MFC 4
The catalyst used was a commercial three-way catalyst comprising of metals (Pt,
Pd, Rh) on an alumina support. It was found that microwaves shifted the CO lightoff
temperature (the temperature where 50% of final conversion is achieved) of the bulk
catalyst lower, compared to the lightoff temperature using only conventional heating.
Microwaves shifted the CO lightoff temperature of the bulk catalyst poisoned by SO2
lower than the lightoff temperature of the unpoisoned catalyst using only conventional
heating. For a catalyst poisoned by SO2, microwave energy increased the overall
conversion of CO from 85% at 513K using conventional heating, to 100% at 500K using
microwave energy. It was concluded that the surface was being selectively heated, thus
the reaction was initiated at a lower gas phase temperature. Further, the SO2 was
24
selectively desorbed and its ability to inhibit the CO oxidation was reversed. It is
important that the reaction or desorption occurs prior to thermal equilibration; otherwise,
there will be little difference for microwave induced sorption/reaction compared to
conventional heating. Using a higher microwave power might cause an even more
pronounced shift in lightoff temperature.
Figure 3.4.3-2 Effect of microwaves on CO lightoff temperature on automotive catalyst
[12]
Effect of MW Energy on CO Lightoff Temperature in
TWC with and without SO2 Present
CO Lightoff
Temperature (C)
0 .45
485K
0 .35
476K
0 .30
Convent ional Heat
Con vent ional Heat w/ M W
Convent ional Heat w/ M W and S O2
Con vn et ional Heat w/ S O2
0 .40
0 .25
482K
m/e 44 Sign al
494K
0 .20
0 .15
0 .10
453
46 3
473
483
49 3
503
513
Bu lk C atalyst T em perat ure (K)
33
3.4.4
Studies on Desorption Under the Use of Microwave Energy in a 1 Meter Column
Desorption of methanol from DAY zeolite was studied under the use of
microwaves in a 1m x 76.2mm cylindrical column by Meier [60]. Temperature probes at
five axial and three radial positions monitored the temperature profile in time. The
25
reflected power from the water load at the end of the column varied during desorption
and caused a distortion of the axial temperature profile.
The dry zeolite bed was mostly transparent to microwave energy, but the
absorption of microwave energy was high when a methanol adsorbate was present. As
desorption proceeded, more of the microwave energy was able to proceed further down
the column before being absorbed. Higher flow rates of the carrier gas carried the
methanol out of the column and increased desorption rates. The power delivered to the
column influenced the rate of desorption. This experiment confirmed that desorption
takes longer if a low power was used. Thus, more heat was transferred from the zeolite
surface to the bulk by conduction. Since that energy did not contribute to desorption, a
decrease in efficiency was observed.
A change in sorption between the first methanol saturation experiment and
subsequent saturation experiments indicated a chemical reaction during the irradiation
with microwaves. It was found that the choice of adsorbate and adsorbent was
unfortunate: a fraction of the methanol reacted on the zeolite under microwave irradiation
to form dimethylether. Also, the surface properties of the zeolite changed during the
sorption of methanol, most likely due to the formation of a partially methoxylated surface
during microwave irradiation. This change was reversible by the adsorption of water in
the column and exposure to microwave energy.
A closer look at energy efficiency was not possible because the dry-ice trap used
to condense the methanol from the gas leaving the column was insufficient. Still, these
results indicated an easy removal of VOCs from a saturated adsorbent with the use of
microwaves.
26
3.4.5
Microwave Radiation’s Influence on Sorption and Competitive Sorption in
Zeolites
Turner [1] used sorption of cyclohexane and methanol on high-silica zeolites to
study the influence of microwaves on adsorption selectivity. As in the previous
experiment, a bed of catalyst was placed inside a microwave waveguide, and the feed gas
compositions were specified using mass flow controllers. The concentrations of the
sorbates were monitored with an online mass spectrometer.
Figure 3.4.5-1 Experimental setup for competitive adsorption in zeolites [1]
Microwave
Zeolite Bed
Waveguide
Mass
Spec.
MFC 1
MFC 2
MeOH
Saturator
Helium
Legend
MFC Mass Flow Cont
- Valve
Diluent
MFC 3
Cyclohex.
Saturator
27
The zeolite and cyclohexane had a low permittivity, while methanol had a high
permittivity. The adsorbate with the higher permittivity (methanol) was desorbed
selectively under microwave radiation. The adsorbed species with the higher permittivity
was heated selectively since the rate of microwave energy absorption was greater than the
rate of heat transfer from the surface. Hence, there were different “effective”
temperatures in the adsorbed phase on the surface, and in the bulk zeolite when under the
influence of microwaves. Thus, desorption was found to occur without the necessity of
heating the system to the same temperature required if thermal equilibrium had been
achieved.
Energy demands for desorption by microwave radiation are less than by
conventional means, because only the adsorbed species need to be heated as opposed to
heating the entire bulk by conventional means. Microwave energy is transferred more
rapidly, can be selectively absorbed by the adsorbed phase, and can be controlled if the
permittivities of the adsorbates differ.
28
Figure 3.4.5-2 Competitive adsorption of cyclohexane and methanol on silicalite [1]
Flow
Competitive Adsorption of Cyclohexane
and Methanol on Silicalite
60
80W
120W
0.012
80W
40W
40
0W
20
0
0.010
40W
Methanol
0.008
0.006
-20
0.004
-40
-60
-80
0.002
Cyclohexane
Concentration of Adsorbate
in Effluent (mol/L)
Bed Temperature (C)
80
Z
e
o B
l e
i d
t
e
480 600 720 840 960 1080 1200 1320 1440
Time (min)
24
29
CHAPTER 4
EXPERIMENTAL
4.1
Single Component Adsorption System Apparatus
An apparatus has been constructed for measurements of adsorption isotherms and
isosteres in the presence of microwaves and is shown in figure 4.1-1. It is similar to the
high resolution adsorption system in U.S. patent number 5,637,810 [58] as shown in
section 3.4.1.
Figure 4.1-1
Adsorption apparatus
Sorption System
Pressure
Transducer
0-1 torr
Vacuum
P
Pressure
Transducer He
1-1000 torr
P
Ballast
P
Dosing Valves
Sample valve
Pressure
Sample
Transducer
1-1000 torr
Microwave Oven
30
Liquid
Gas
The design for this system has been modified to include only metal seals in the
valves. This prevents problems of seal degradation for many elastomers employed as
valve gaskets that can be caused by some organic species such as alcohols, etc. Also, the
sample can be placed in the center of a specially constructed microwave oven that uses
microwave chokes to prevent microwave radiation leakage from the oven. This system is
employed to measure in-situ adsorption in the presence of microwaves. The system
includes: 1) large diameter tubing for faster outgassing, 2) a turbo pump backed by a
rough pump to achieve low pressures, 3) multiple pressure transducers on the manifold to
accurately measure pressure over a large range, 4) pneumatically controlled valves, 5)
manifold ports for use of vapor or gas adsorbates, and 6) air cooling over the outside of
the sample to help maintain a constant bulk gas phase temperature in the presence of
microwaves, or for room temperature experiments. An oil bath around the sample was
used in experiments measuring conventional isotherms above room temperature.
4.2
Sample Preparation
The silica used in the experiments was Aerosil® 200 fumed silica from
Degussa®, and was calcined at 385°C. To prevent elutriation of the sample, the silica
was pressed to 5000 psi, and broken up into small chunks (about 1-2 mm in size). One of
the silica samples was additionally heat treated by being ramped up to 700°C at 1°C/min
and held at 700°C for 4 hours in order to influence the surface hydroxyl concentration.
Samples were generally outgassed overnight at a temperature above 120°C. The silicalite
used was from Union Carbide, lot #961884061002-S, and was calcined at 385°C.
Silicalite samples were outgassed overnight at a temperature above 220°C before each
sorption experiment.
31
Nitrogen isotherms at 77 K were obtained to characterize the samples. They are
shown in appendix B. The surface area of the Aerosil 200 silica was calculated using a
multipoint BET method [7] to be 183 m2/g. There was a small amount of hysteresis
observed in the silica isotherms due to the pressing of the silica during sample
preparation. This is in agreement with the literature [38]. The surface area of the Aerosil
200 silica sample that was temperature treated up to 700°C was calculated to be 197 m2/g,
and still maintained the same outward appearance (no discoloration). The nitrogen
isotherms on silicalite showed the typical isotherms for that material and were in
agreement with the literature. The surface area of the silicalite was calculated using a
multipoint BET method to be 369 m2/g; although, the BET Surface areas for zeolites are
not meaningful since the pores are filled at pressures below which the BET theory is
valid.
The adsorbates, 1) n-pentane, 2) cyclohexane, 3) dichloromethane, 4) isopropanol,
and 5) methanol, were chosen to cover a range of permittivities, and had appropriate
partial pressures for sorption experiments.
Permittivities were measured with a Hewlett Packard 8510 Network Analyzer,
using a ¾” diameter probe connected to the instrument by a shielded coaxial cable at
room temperature. Liquid permittivities were measured by immersing the probe tip in the
liquid far away from the container walls. Powders were measured by hand packing the
powder to about 1” thickness and placing the probe on top of the packed powder.
Permittivities were measured at from 0.5-18GHz and at 22C, and are shown at 2.45GHz
in Table 4.2-1 [59, 61].
32
Table 4.2-1
Permittivity of materials at 2.45 GHz [61]
Microwave power
Aerosil 200 silica
silicalite
n-pentane
cyclohexane
dichloromethane
isopropanol
methanol
*values are +/- 0.1
relative permittivity (2.45 GHz at 22C)
*(e')
*(e")
1.43
-0.08
2.41
0.01
1.80
NA
2.02
0.05
9.08
0.39
18.30
3.19
23.04
13.77
Methoxylation was suspected to take place during the adsorption experiments
where methanol was adsorbed on the sample in the presence of microwaves. To examine
this, samples were prepared for IR spectroscopy. A silica sample was taken and mixed in
a 1:10 ratio with KBr powder and pressed at 5,000 psi into a thin wafer. The silica
samples used were Aerosil 200, temperature treated Aerosil 200, methoxylated Aerosil
200, and methoxylated temperature treated Aerosil 200. The wafers were then placed in a
sample holder with the IR beam passing through the center of the wafer, and
transmittance of the IR beam was analyzed. The sample holder of the wafer was ramped
up to 110° in nitrogen.
4.3
Sorption Isotherm Measurement
Helium expansion was used for the calculation of the volumes in the system since
it is far above its critical point, does not adsorb on the sample, and reached thermal
equilibrium quickly. A vessel of a known volume connected to the manifold by a valve
was employed to calibrate the manifold volume, sub-volumes, and sample dead space.
The volume of a glass bulb with a valve attached was determined by filling the bulb with
liquid mercury and calculating the volume using the accurately known density of the
33
liquid mercury. The volume of the manifold was then determined by connecting the bulb
to the manifold. The glass bulb was filled with helium to a known pressure. The manifold
was evacuated and the valve on the glass bulb was opened, allowing the glass bulb and
manifold to come to a new pressure. The manifold volume was then calculated using the
Ideal Gas Law. This measurement was repeated a number of times (8-12), and an average
was taken. Each volume or sub-volume of the manifold is then determined in a similar
manner using the Ideal Gas Law or by difference.
The sample dead space is calculated in a similar manner, evacuating the sample
volume, filling the manifold with helium at a known pressure, and opening the valve to
the sample and measuring the resulting pressure. Subsequent calculations of the dead
space can be performed using the Ideal Gas Law by closing the valve to the sample area
at a known pressure, adding (or removing) helium to the manifold to a known pressure,
and then opening the valve to the sample and measuring the resultant pressure.
For example: The manifold volume is 82 [cm3 at stp]. The sample cell is
evacutated to 0 [torr]. 500 torr of helium is dosed to the manifold. The valve to the
sample is then opened, and the resulting pressure is 400 torr. The Ideal Gas Law states
that:
P1V1
PV
= 2 2
n1 RT1 n2 RT2
Subscript 1 denotes before the valve is opened and subscript 2 after the valve is opened.
V1 is the manifold volume, P1 is the pressure in the manifold before the valve is opened, V2
is the sample dead space plus the manifold volume, and P2 is the resulting pressure in the
manifold plus the sample dead space area once the valve is opened. Since the sample cell
was under vacuum, n1 = n2, and R and T are constant, so the ideal gas law reduces to:
P1Vman = P2 (Vman + Vsds)
or
34
Vsds = Vman (P1-P2) / P2 = 82 (500-400) / 400 = 20.5 [cc’s at stp]
During sorption isotherms, the amount of gas adsorbed at a particular pressure
was calculated by taking the difference of the amount of gas that would have been in the
gas phase if no sorption was taking place (calculated using the Ideal Gas Law), and the
measured amount of gas in the gas phase at equilibrium after sorption. The sample was
initially outgassed, and the sorbate was introduced to the manifold to a known pressure
through the dosing valves. Then the sample valve was opened and the system was
allowed to come to equilibrium and the resulting pressure was measured. Additional
adsorption data points were obtained by closing the valve to the sample area and
incrementally dosing a higher pressure (greater than the equilibrium pressure of the
previous step) to the manifold. It is not necessary to evacuate the sample volume before
each adsorption step (since the amount in the gas phase and the amount adsorbed are
known). Desorption data points were obtained by closing the valve to the sample volume
and incrementally dosing a lower pressure (lower than the equilibrium pressure of the
previous step) from the manifold. A sorption isotherm is obtained by plotting data of the
amount adsorbed versus the equilibrium pressure at a constant temperature.
35
4.4
Isosteres in the Presence of Microwaves
The sample on the adsorption system lies inside of the cavity of a specially
modified industrial microwave oven from CEM corporation, model MDS-81. The
microwave oven was a pulsed system with a duty cycle (pulse duration) of 1 second, and
had chokes (grounded metal tubes of precise length and diameter) to prevent the leakage
of microwaves into the environment.
An adsorption isotherm at room temperature was measured up to a particular
pressure. The adsorption procedure was then modified by turning on the microwave oven
at a low power level and the system was allowed to come to a steady state where a new
pressure and hence a new amount adsorbed was measured. In the case of species with a
high permittivity, the minimum pressure in the presence of microwaves was chosen to be
above that at which a plasma would form. A stream of air at room temperature was flown
over the length of the outside of the glass sample tube to maintain a constant gas phase
temperature. The microwave power was increased and the system was allowed to come to
a new steady state (with a new amount adsorbed), and then the microwave power was
increased to a higher power level again, and the system was allowed to come to a new
steady state and resulting pressure. The microwaves were then turned off, and the system
reached equilibrium, matching the original equilibrium data point. Another adsorption
point is obtained (without the microwave on). This entire process is repeated for a
number of data points and allows for the measurement of isosteres in the presence of
microwaves for the different levels of microwave power used.
Measurements of the amounts of adsorption in the presence of microwaves were
compared to the conventional adsorption isotherms.
36
By equating the adsorbate pressure required to achieve a specific coverage (an isostere)
in the presence of microwaves to the amount adsorbed for a conventional isotherm, an
estimate of the surface temperature in the presence of microwaves was obtained. This is
called the “effective surface temperature”.
4.5
Multi-Component Adsorption in the presence of microwaves
•
Apparatus
An apparatus has been constructed for measurements of the amount adsorbed on a
packed bed for two adsorbates, which can be heated conventionally or placed through a
waveguide to be heated with microwaves at 2.45 GHz or 5.8 GHz. It is shown in figure
4.5-1.
Figure 4.5-1 – Multi-Component Sorption System
37
A flow of helium from a regulator feeds a stream that is divided into three
streams. A flow controller is attached to control the flow of each of the three streams.
One stream has a flow controller that is a Tygan General FC-260 0-100 sccm, and that
stream goes through a bubbler that may be filled with any adsorbate with an appropriate
vapor pressure. The second stream has a flow controller that is a Tygan General FC-260
0-200 sccm, and that stream also goes through a bubbler that may be filled with any
adsorbate with an appropriate vapor pressure. The third stream has a flow controller that
is an Edwards 825 series B 0-500 sccm, and this stream is used as the helium diluent
stream. The three streams then combined to a single stream that can be fed to or bypass
the reactor. The valves on the system also allow the helium diluent stream to pass through
the reactor, while the other two adsorbate containing streams bypass the reactor.
The reactor is made of glass tubing that is 15 mm in diameter and 55 cm long.
Temperature is measured by two fiber optic temperature probes (Neoptix model T1), one
entering through a septum on fittings attached to each end of the reactor. One is placed to
measure the temperature in the center of the adsorbent bed, and in the other to measure
the temperature of the effluent gas phase from the reactor.
The process of loading the adsorbent into the reactor involves first filling the bed
half way (about 3/4” from the frit). A small diameter glass tube is then placed going
down the center of the glass tube to hold the fiber optic temperature probe in place in the
center of the bed. The bed is then filled the rest of the way with adsorbent. On top of the
adsorbent is a small amount of glass wool to hold the adsorbent in place. Then, 3 mm
diameter glass beads are placed on top of the bed to distribute the flow. Another small
amount of glass wool is placed above the glass beads to hold them in place. The reactor is
38
then put on-line in the system, connected by o-ring clamp fittings. The reactor may be
placed going through a specially constructed section of waveguide, or wrapped with
heating tape.
The gas phase effluent from the reactor is sampled using a two stage pumping
system. Some of the gas phase exiting the reactor passes through a needle valve to a
section of tubing connected to a rough pump (Pfeiffer model Duo 2.5A), then through
another needle valve to the section to which the turbo pump (Alcatel model cfv 100) and
another rough pump (Alcatel model M2004A) are attached, as well as the mass
spectrometer (Extorr model XT200M).
Some experiments were carried out with the effluent gas from the reactor being
condensed in a liquid nitrogen trap. The trapped liquid was then analyzed using by
GC/MS (gas chromatograph (HP 5890) / mass spectrometry (HP 5989A)) in order to
determine if there were any reactions taking place.
•
Adsorbents and Adsorbates
The silica used in the experiments was Aerosil® 200 fumed silica from
Degussa®, and was calcined at 385°C. To prevent plugging of the glass frit in the reactor,
the silica was pressed to 5000 psi, and broken up into small chunks (about 1-2 mm in
size). Samples were pre-treated with a flow of helium and heated to a temperature above
100°C to remove any water adsorbed on the sample. The silicalite zeolite (Si/Al > 1000)
used was from Union Carbide, lot #961884061002-S, and was calcined at 720°C.
39
The adsorbates, 1) acetone, 2) isopropanol, 3) methanol, 4) cyclohexane, and 5)
benzene were chosen due to their permittivities at the microwave frequencies of interest,
to compare to previous work, and had appropriate partial pressures for sorption
experiments.
Permittivities were measured with a Hewlett Packard 8510 Network Analyzer,
using a ¾” diameter probe connected to the instrument by a shielded coaxial cable at
room temperature. Liquid permittivities were measured by immersing the probe tip in the
liquid far away from the container walls. Powders were measured by hand packing the
powder to about 1” thickness and placing the probe on top of the packed powder.
Permittivities were measured at from 0.5-18GHz and at 22C, and are shown at 2.45GHz
in Table 1 [59, 61]. A summary of the heats of adsorption of the materials use on Aerosil
200 and silicalite are given in appendix C.
Table 4.5-1 – Measured permittivities of materials at 2.45 GHz and 5.8 GHz at 22°C [59,
61]
material
Aerosil 200 silica
silicalite
acetone
isopropanol
methanol
cyclohexane
benzene
*values are +/- 0.1
relative permittivity, 2.45
*ε'
*ε"
1.43
-0.08
2.41
0.01
21.94
1.01
18.30
3.19
23.04
13.77
2.02
0.05
NA
NA
relative perimittivity 5.8 GHz
*ε'
*ε"
1.48
-0.02
2.38
0.13
21.63
3.12
3.83
1.82
12.05
12.19
2.33
0.13
NA
NA
40
•
Procedure
Single component isotherms were obtained at room temperature using the
volumetric system for each adsorbent/adsorbate pair. Using the resulting isotherms and
calculating the partial pressure present in the flow adsorption system allowed the
quantification of the amount adsorbed for a single component while using the flow
adsorption system. To quantify the amount adsorbed when more than one component was
present, experiments were carried by flowing one adsorbate, allow it to come to steady
state, and then add the second adsorbate and measure the change in adsorption of the first
component due to the second. This was done by integrating the changes in the mass
spectrometer signal. This allows the amount adsorbed as well to be calculated in the flow
adsorption system when two adsorbates are present. Experiments were then carried out in
the flow adsorption system with two components adsorbing, and then heating the system
with conventional heating or microwave at 2.45 or 5.8 GHz and measuring the resulting
adsorption behavior by integrating the changes to the mass spectrometer signal for each
component.
41
CHAPTER 5
RESULTS: SINGLE COMPONENT ADSORPTION IN THE PRESENCE OF
MICROWAVES
5.1
Plasma Formation
The initial attempts of obtaining isosteres in the presence of microwaves while
using isopropanol and methanol resulted in the formation of a plasma. The adsorbate was
dosed to the system to a relatively low pressure. When the sample was exposed to
microwaves, after a short time the sample tube began to flash a purple glow when the
microwave was on (during the power portion of the duty cycle). The pressure drastically
increased and the sample temperature increased to the point that in one case the glass
sample tube began to melt.
To prevent plasma formation in future experiments, the isosteres were obtained
only at higher pressures for adsorbates with a high permittivity. Since data could not be
obtained at low pressures in these cases, the effective surface temperature was
determined by interpolation between conventional isotherms; using the heat of adsorption
to estimate the effective surface temperature was not accurate since it should only be
applicable to the Henry’s Law region of the isotherm (at low pressures).
5.2
Cyclohexane (ε” = 0.05) on Aerosil 200 Silica
Conventional isotherms for cyclohexane on silica at 22, 32, and 53°C, and
isosteres in the presence of microwaves at a power of 60, 120, and 240 Watts were
obtained and are shown in Figure 5.2-1.
42
Figure 5.2-1 Adsorption isotherms for cyclohexane on Aerosil 200 silica as a function
of microwave power employed
Cyclohexane on Aerosil 200 Silica
1.2
22 C
12
60 W 120 W
1.0
32 C
10
0.8
8
240 W
6
0.6
0.4
4
53 C
2
0.2
0
0.0
100
0
20
40
60
80
Pressure (Torr)
Coverage (θ)
Volume Adsorbed (cc/g)
14
The isotherms were linear in nature, so the interaction of the first layer of the
adsorbate to the adsorbent was similar to the interactions of the subsequent layers of
adsorbate with itself. Since the slope of the isotherm was not steep in the low pressure
region, the pressure at which a specific amount was adsorbed could be determined
accurately. The heat of adsorption could be estimated reasonably well. This was done by
using the Clausius-Clapeyron equation [10].
43
(d ln P)|θ / (d (1/T)) = ∆Hads/R
For an ideal gas and with a constant ∆Hads with respect to θ:
ln P = - ∆Hads / RT + constant
From isotherms at several temperatures (plots of the amount adsorbed versus
pressure at a constant temperature), a series of plots of isosteres were generated (ln P
versus 1/T at a constant volume adsorbed). From the above equation, the heat of
adsorption was found since the slope of the isostere is equal to - ∆Hads / R.
Once the isosteres were generated, the effective temperature in the presence of
microwaves (at a particular volume adsorbed) can then be estimated using the slope and
intercept of the isosteres and the above equation and solving for the temperature.
Teff = - ∆Hads / R (ln P – constant)
Or from the plots of the isosteres,
Teff = slope / (ln P – intercept)
The effective surface temperature in the presence of microwaves (as a function of
pressure) was also estimated by interpolating the volume adsorbed with the conventional
isotherms.
44
Both methods of calculating the effective surface temperature gave similar results and are
shown in Figure 5.2-2.
Cyclohexane had a relatively small slope in the plot of the effective surface
temperature versus pressure, due to the very low permittivity of cyclohexane. In this case
the presence of cyclohexane in the gas or adsorbed phase has little effect on the effective
surface temperature.
The effective surface temperature of the adsorbent when no adsorbate is present
can be estimated by extrapolating to zero adsorbate pressure (or zero coverage). The
change in effective surface temperature is not zero, which must be due to the fact that the
adsorbent surface is not completely transparent to microwave energy.
45
Figure 5.2-2 Change in effective surface temperature of cyclohexane on Aerosil 200
silica as a function of microwave power employed, using interpolation (solid lines) and
the heat of adsorption (dashed lines) to calculate the effective surface temperature
Change in Effective Surface
Temperature (C)
Change in Effective Surface Temperature Cyclohexane on Aerosil 200 Silica
30
60 W
(interp)
25
120 W
(interp)
20
240 W
(interp)
15
60 W
(Hads)
10
120 W
(Hads)
5
240 W
(Hads)
0
0
5.3
20
40
60
Pressure (Torr)
80
Dichloromethane (ε” = 0.39) on Aerosil 200 Silica
Conventional isotherms for dichloromethane on silica at 22, 42, 57, and 82°C, and
isosteres in the presence of microwaves at a power of 60, 120, and 240 Watts were
obtained and are shown in Figure 5.3-1. The coverage shown (θ) is the relative surface
coverage (surface coverage relative to one monolayer of surface coverage).
46
Figure 5.3-1 Adsorption isotherms for dichloromethane on Aerosil 200 silica as a
function of microwave power employed
Dichloromethane on Aerosil 200 Silica
1.6
30
60 W
25
1.4
120 W 1.2
20
42 C
15
57 C
1.0
0.8
0.6
10
240 W
5
0.4
Coverage (θ)
Volume Adsorbed (cc/g)
22 C
0.2
82 C
0
0
100
200
300
Pressure (Torr)
0.0
400
Similar to cyclohexane on silica, the isotherms were linear in nature, so the
interaction of the first layer of the adsorbate to the adsorbent was similar to the
interactions of the subsequent layers of adsorbate with itself. The heat of adsorption could
be estimated reasonably well and was used to find the effective surface temperature, as
above. The effective surface temperature in the presence of microwaves (as a function of
pressure) was also estimated by interpolating the volume adsorbed with the conventional
isotherms.
47
Both methods of calculating the effective surface temperature gave similar results and are
shown in Figure 5.3-2.
Figure 5.3-2 Change in effective surface temperature of dichloromethane on Aerosil
200 silica as a function of microwave power employed, using interpolation (solid lines)
and the heat of adsorption (dashed lines) to calculate the effective surface temperature
Change in Effective Surface
Temperature (C)
Change in Effective Surface Temperature Dichloromethane on Aerosil 200 Silica
70
60 W
(interp)
60
50
120 W
(interp)
40
240 W
(interp)
30
60 W
Hads
20
120 W
Hads
10
240 W
Hads
0
0
100
200
300
400
Pressure (Torr)
The effective surface temperature increase during dichloromethane adsorption
was greater than that for cyclohexane (for the same microwave power used) due to the
higher permittivity of dichloromethane.
48
It was noticed that the calculated volume adsorbed begins do decrease at high
pressures for the isostere for dichloromethane on silica at 240 W. For conventional
isotherm measurements, a decrease in the amount adsorbed with increasing pressure is
contradictory. We conclude that this is a result of a non-isothermal condition in the gas
phase. The air cooling over the sample at higher pressures is no longer sufficient to keep
the temperature of the gas phase at 22 C. The calculations for the amount adsorbed are
based on an isothermal gas phase.
To confirm this, an experiment was carried out adsorbing dichloromethane at 240
W without air cooling. The calculated amount adsorbed begins to decrease at higher
pressures and even drops to a negative value. A diagram of the results of the experiment
is shown in figure 5.3-3. After observing this behavior, experiments were carried out at a
microwave power where we estimate that the air cooling would be sufficient to achieve a
steady state in the presence of microwaves. This effect also depends on the permittivity
of the adsorbate. A higher permittivity adsorbate necessitates use of a lower maximum
microwave power to maintain an isothermal gas phase.
49
Figure 5.3-3 Isotherms for dichloromethane on Aerosil 200 silica as a function of
microwave power employed with and without gas phase temperature control
Dichloromethane o n Aerosil 200 Silica w/wo Gas Phase Coo ling
Volume Ad sorbed (cc/g)
20
22 C
60 W
15
120 W
10
240 W
5
60 W no
cooling
120 W no
cooling
0
240 W no
cooling
-5
0
50
100 150 200 250 300
Pressure (Torr)
5.4
Isopropanol (ε” = 3.19) on Aerosil 200 Silica
As stated earlier, a plasma was formed during the initial experiments with
isopropanol as the adsorbate. This contaminated the samples. New samples were
prepared and plasma formation was prevented by only exposing the sample to
microwaves at an adsorbate pressure greater than 10 torr. Conventional isotherms for
isopropanol on silica at 22, 42, and 67°C, and isosteres in the presence of microwaves at
a power of 30, 60, and 120 Watts were obtained and are shown in Figure 5.4-1.
50
Figure 5.4-1 Adsorption isotherms for isopropanol on Aerosil 200 silica as a function
of microwave power employed
Isopropanol on Aerosil 200 Silica
20
22 C
18
30 W 1
16
60 W
14
0.8
12
42 C
10
0.6
8
120 W
6
0.4
67 C
4
Coverage (θ)
Volume Adsorbed (cc/g)
1.2
0.2
2
0
0
0
10
20
30
40
Pressure (Torr)
The isotherms for isopropanol on silica were not linear in nature. Unlike the
isotherms for cyclohexane or dichloromethane, there was more adsorbed at lower
pressures, so interaction of the first layer of the adsorbate to the adsorbent was stronger
than the interactions of the subsequent layers of adsorbate with itself. This is probably
due to the hydrogen bonding of the isopropanol.
51
The heat of adsorption could not be estimated well using the Clausius-Clapeyron
equation because data were not obtained at low pressures in the Henry’s Law region (to
prevent plasma formation).
The effective surface temperature in the presence of microwaves (as a function of
pressure) was estimated by interpolating the volume adsorbed with the conventional
isotherms. Data was available over a limited range of pressures because the experiments
had to be started at higher pressures to prevent plasma formation. The effective surface
temperature increase during isopropanol adsorption was greater than that for
dichloromethane (for the same microwave power used) due to the higher permittivity of
isopropanol. A diagram of the effective surface temperature for isopropanol on Aerosil
200 Silica is shown in figure 5.4-2. Isopropanol had a steeper slope than dichloromethane
and cyclohexane in the plot of the effective surface temperature versus pressure, due to
the high permittivity of isopropanol.
52
Figure 5.4-2 Change in effective surface temperature of isopropanol on Aerosil 200
silica as a function of microwave power employed
Isopropanol on Aerosil 200 Silica
45
Change in Effective Surface
Temperature (C)
40
30 W
35
30
25
60 W
20
15
10
120 W
5
0
0
10
20
30
40
Pressure (Torr)
5.5
Methanol (ε” = 13.77) on Methoxylated Aerosil 200 Silica
It was noticed from previous results that the amount adsorbed was not completely
reversible before and after exposure to microwaves, so it was suspected that the methanol
had reacted with the surface. Previous work by Meier [60] denotes that in the presence of
microwaves methanol can replace surface O-H bonds of the sample to methoxylate the
surface and change its adsorption properties.
53
Since the adsorption properties of the surface were changing in the presence of
microwaves and it was suspected that some degree of chemisorption was taking place, the
amount adsorbed and the effective surface temperature cannot be directly compared to
other adsorbates on silica, because the surface properties of the silica are changed by the
adsorption of methanol. Once the methanol has fully reacted with the surface, the
isotherms and isosteres in the presence of microwaves can be obtained, and an effective
temperature in the presence of microwaves can be obtained for methanol on the
methoxylated surface.
Conventional isotherms for methanol on methoxylated silica at 22, 41, and 60°C
and isosteres in the presence of microwaves at a power of 30, 60, and 120 Watts were
obtained and are shown in figure 5.5-1.
54
Figure 5.5-1 Adsorption isotherms for methanol on methoxylated Aerosil 200 silica as
a function of microwave power employed
Methanol on Methoxylated Silica
30
1.2
25
1.0
30 W
41 C
20
0.8
60 W
15
0.6
60 C
10
0.4
120 W
5
0
Coverage (θ)
Volume Adsorbed (cc/g)
22 C
0.2
0.0
0
25
50
75
100
Pressure (Torr)
125
The isotherms for methanol on methoxylated silica were similar in nature to the
isotherms for isopropanol on silica; and the heat of adsorption could not be estimated
well using the Clausius-Clapeyron equation because data were not obtained at low
pressures in the Henry’s Law region (to prevent plasma formation).
The effective surface temperature in the presence of microwaves (as a function of
pressure) was estimated for the methoxylated surface by interpolating the volume
adsorbed with the conventional isotherms. It was noticed that the volume adsorbed begins
55
do decrease at high pressures for the isostere for methanol at 120 W, similar to that of
dichloromethane on silica at 240 W as above. By similar reasoning, it is suspected that
when the volume adsorbed decreases with an increase in pressure, the assumption that the
gas phase is at a constant temperature is no longer valid. The air cooling over the sample
during experiments using high microwave power and higher pressures of a high
permittivity adsorbate is no longer sufficient to keep the temperature of the gas phase at
22 C. This behavior could also be due to the reaction of methanol to form dimethylether.
Meier [60] found that methanol reacts on silicalite under microwave irradiation to form
dimethylether.
The effective surface temperature increase in the presence of microwaves during
methanol adsorption was only slightly greater than that for isopropanol (for the same
microwave power used), because even though methanol has a much higher permittivity,
the methanol reacted with the silica to alter the surface properties. A figure of the
effective surface temperature for methanol on methoxylated Aerosil 200 silica is shown
in figure 5.5-2.
56
Figure 5.5-2 Change in effective surface temperature of methanol on methoxylated
Aerosil 200 silica as a function of microwave power employed
Change in Effective Surface
Temperature (C)
Methanol on Methoxylated Silica
60
50
30 W
40
30
60 W
20
10
120 W
0
0
20
40
60
80
100
Pressure (Torr)
5.6
Methanol (ε” = 13.77) on Temperature Treated Methoxylated Aerosil 200
Silica
In order to investigate the possible methoxylation of the surface, A sample of
Aerosil 200 silica was temperature treated by heating it to 700°C at 1°C/min, and holding
it at 700°C for 4 hours in order to remove a fraction of the surface hydroxyl groups,
reducing the amount of methoxylation of the surface that could occur. A multipoint BET
calculation was performed on isotherms using nitrogen to check for any surface area loss
57
caused by the temperature treatment. It was found that the surface areas were
approximately the same (see section 4.2 for details), and the sample had the same
appearance (no discoloration). Conventional isotherms for methanol on the temperature
treated methoxylated silica at 22, 41, and 58°C and isosteres in the presence of
microwaves at a power of 30, 60, and 120 Watts were obtained and are shown in figure
5.6-1.
Figure 5.6-1 Adsorption isotherms for methanol on temperature treated (700 °C)
methoxylated Aerosil 200 silica as a function of microwave power employed
Methanol on Temperature Treated
Methoxylated Silica (700 C)
22 C
1.4
1.2
50
1.0
40
30 W
0.8
30
60 W
20
0.6
41 C 0.4
58 C
120 W 0.2
10
0
0.0
0
50
100
Pressure (Torr)
58
150
Coverage (θ)
Volume Adsorbed
(cc/g)
60
The shape of the adsorption isotherm for methanol on the temperature treated
sample was similar to that of a sample with standard preparation, however, the surface
properties of the silica were changed and the effective surface temperature was lower for
the temperature treated sample.
A figure of the effective surface temperature for methanol on temperature treated
methoxylated Aerosil 200 silica is shown in figure 5.6-2.
Figure 5.6-2 Change in effective surface temperature of methanol on temperature
treated (700 °C) methoxylated Aerosil 200 silica as a function of microwave power
employed
Methanol on Temperature Treated (700C)
Methoxylated Silica
Change in Effective Surface
Temperature (C)
60
50
30 W
40
30
60 W
20
10
120 W
0
0
20
40
60
80
Pressure (Torr)
59
100 120
5.7
IR Spectroscopy of Aerosil 200 Silica and Methoxylated Samples
In order to confirm that a chemical reaction was taking place with methanol in the
presence of microwaves and to investigate the changes of the silica surface with
temperature treatment, IR spectroscopy was performed to investigate changes on the
surface of the Aerosil 200 silica with 5 different sample treatments; 1) uncalcined
Aersosil 200, 2) Aerosil 200 with standard pretreatment (calcined to 385°C), 3)
methoxylation of the surface of Aerosil 200 when methanol was adsorbed in the presence
of microwaves, 4) temperature treated Aerosil 200 (700°C for 4 hours), to investigate
changes in the hydroxyl content due to temperature treatment, and 5) temperature
treatment followed by methoxylation of the surface. Significant peaks in the spectrums
will be discussed, and Table 5-1 below is given for reference.
Table 5.7-1
Infrared peaks for chemical groups [22, 56, 57, 62]
wavenumbers
3748
3400-3550
2997, 2960, 2857
1865
1630
1470
attributed species
isolated and geminal Si-OH stretch
associated hydroxyls stretch
CH stretches from Si-0-CH3
Si-O stretch
OH bend
CH3 bend
This transmission IR spectroscopy was used because the samples, after adsorption
in the presence of microwaves, could afterward easily be studied with minimal sample
preparation to determine the presence of methoxyl groups. Because we were studying
surface characteristics using a bulk technique, the signal resolution was not high;
however, the methoxyl peaks were still evident. Methanol reacted with the silica in the
60
presence of microwaves to methoxylate the silica surface, and reacted to a lesser extent
on the sample treated to 700°C. It is hypothesized that the temperature treatment reduces
the surface hydroxyl concentration so that there are fewer reactive sites for which
methanol to react with to form a methoxyl group. A figure of the methoxyl peaks is
shown in figure 5.7-1.
Figure 5.7-1
5.8
Transmission IR spectroscopy of methoxylated Aerosil 200 silica
n-Pentane on Silicalite
Conventional isotherms for n-pentane on silica at 22, 35, and 48°C, and isosteres
in the presence of microwaves at a power of 60, 120, and 240 Watts were obtained and
are shown in figure 5.8-1.
61
Figure 5.8-1 Adsorption isotherms for n-pentane on silicalite as a function of
microwave power employed
n-Pentane on Silicalite
1.8
22 C
Volume Adsorbed (cc/g)
40
60 W
120 W
35
240 W
30
35 C
25
1.6
1.4
1.2
1.0
20
0.8
15
0.6
48 C
10
0.4
5
0.2
0
0.0
0
100
200
300
Pressure (Torr)
Coverage (θ)
45
400
The shape of the adsorption isotherms for all of the adsorbates used on silicalite
were steep in the low pressure region, since the adsorbent is microporous. The effective
surface temperature in the presence of microwaves (as a function of pressure) was
estimated by interpolating the volume adsorbed with the conventional isotherms.
Similar to cyclohexane on silica, n-pentane on silicalite showed a very small
increase in the effective surface temperature with microwave exposure due to its low
permittivity. A plot of the effective surface temperature versus pressure is shown in
62
figure 5.8-2.
For each set of adsorbates, the extrapolated y-intercepts of the effective surface
temperature versus pressure (or coverage) are much closer to zero for silicalite than
silica; this indicates that with no adsorbate present, the silica absorbs more microwave
energy than silicalite. This is reasonable since the permittivity of the silica is expected to
be higher than that of silicalite due to the higher concentration of hydroxyl groups on the
surface.
63
Figure 5.8-2 Change in effective surface temperature of n-pentane on silicalite as a
function of microwave power employed
n-Pentane on Silicalite
Change in Effective Surface
Temperature (C)
3.0
60 W
2.5
2.0
120 W
1.5
1.0
240 W
0.5
0.0
0
5.9
50 100 150 200 250 300 350
Pressure (Torr)
Dichloromethane (ε” = 0.39) on Silicalite
Conventional isotherms for dichloromethane on silicalite at 22, 38, and 52°C, and
isosteres in the presence of microwaves at a power of 30, 60, and 120 Watts were
obtained and are shown in Figure 5.9-1.
64
Figure 5.9-1 Adsorption isotherms for dichloromethane on silicalite as a function of
microwave power employed
Dichloromethane on Silicalite
70
22 C
30 W
60 W
1.6
1.4
50
38 C
120 W 1.2
40
1.0
52 C
0.8
30
0.6
20
0.4
10
Coverage (θ)
Volume Adsorbed (cc/g)
60
1.8
0.2
0
0
100
200
300
400
0.0
500
Pressure (Torr)
The effective surface temperature in the presence of microwaves (as a function of
pressure) was estimated by interpolating the volume adsorbed with the conventional
isotherms. Similar to the results on the silica, dichloromethane had an intermediate slope
in the plot of the effective surface temperature versus pressure. A plot of the effective
surface temperature versus pressure is shown in figure 5.9-2.
65
Figure 5.9-2 Change in effective surface temperature of dichloromethane on silicalite
as a function of microwave power employed
Change in Effective Surface
Temperature (C)
Change in Effective Surface TemperatureDichloromethane on Silicalite
30
25
30 W
20
60 W
15
10
120 W
5
0
0
100
200
300
400
Pressure (Torr)
5.10
Isopropanol (ε” = 3.19) on Silicalite
Conventional isotherms for isopropanol on silicalite at 22, 53, 71, and 83°C, and
isosteres in the presence of microwaves at a power of 30, 60, and 120 Watts were
obtained and are shown in Figure 5.10-1.
66
Figure 5.10-1 Adsorption isotherms for isopropanol on silicalite as a function of
microwave power employed
Isopropanol on Silicalite
22 C
Volume Adsorbed (cc/g)
40
30 W
1.4
53 C
35
60 W
30
120 W
1.2
1.0
0.8
25
71 C
20
0.6
88 C
15
0.4
10
Coverage (θ)
45
0.2
5
0
0.0
0
10
20
30
Pressure (Torr)
40
The effective surface temperature in the presence of microwaves (as a function of
pressure) was estimated by interpolating the volume adsorbed with the conventional
isotherms. It was noticed that that the volume adsorbed begins do decrease for the
isosteres of isopropanol on silicalite at 60W and 120W, similar to the experiment for
dichloromethane on Aerosil 200 silica at 240 W. It is suspected that when the volume
adsorbed decreases with an increase in pressure that the assumption that the gas phase is
at a constant temperature is no longer valid. The air cooling over the sample at higher
67
pressures is no longer sufficient to keep the temperature of the gas phase at 22 C. The
trend of the decrease in the volume adsorbed versus pressure begins at a lower power
with isopropanol than the other adsorbates because it has a higher permittivity; and it is
also begins at a lower power on silicalite than on silica because the adsorbates adsorb at a
lower pressure on silicalite due to the microporous network structure. A plot of the
effective surface temperature versus pressure is shown in figure 5.10-2.
Figure 5.10-2 Change in effective surface temperature of isopropanol on silicalite as a
function of microwave power employed
Change in Effective Surface
Temperature (C)
Change in Effective Surface TemperatureIsopropanol on Silicalite
50
45
30 W
40
35
30
60 W
25
20
15
10
120 W
5
0
0
10
20
Pressure (Torr)
68
30
5.11
Discussion
The effect of microwaves on the adsorption equilibrium relationship (and hence
the amount adsorbed) can be visualized in two equivalent ways. The microwaves can
increase the energy level within its energy potential well, or the microwaves can raise the
potential energy of the well itself, shifting the entire well up in energy. This is depicted in
Figure 5.11-1.
Figure 5.11-1
Energy potentials of systems upon exposure to microwaves
D
∆(∆Η)
∆Teµ
69
Similarly, it is shown mathematically that the change in the equilibrium constant
due to the presence of microwaves can be attributed to a change in the enthalpy of the
adsorbed phase (or the heat of adsorption) due to microwaves, or to a change in the
effective temperature due to microwaves.
The equilibrium constant for adsorption can be expressed as
Ka = e
Q
RT
=e
− ( Ha −H g )
RT
Where Ka = the equilibrium constant for adsorption, Q is the heat of adsorption, Ha is the
enthalpy of the adsorbed phase, and Hg is the enthalpy of the gas phase. The equilibrium
constant for adsorption in the presence of microwaves can be expressed as
Q
K aµ = e
µ
RT
=e
− ( H aµ − H g )
RT
=e
Q
RTeµ
Where Kaµ is the equilibrium constant for adsorption in the presence of microwaves, Qµ is
the heat of adsorption in the presence of microwaves, Haµ is the enthalpy of the adsorbed
phase in the presence of microwaves, and Teµ is the effective temperature in the presence
of microwaves. If the equilibrium constant in the presence of microwaves is divided by
the equilibrium constant,
70
K aµ
Ka
=e
− ( H aµ − H a )
RT
=e
Q 1
(
R Teµ
− T1 )
so
− ( H aµ − H a )
T
= Q( T1eµ − T1 )
or
∆H µ
Q
=
∆Tµ
Teµ
Where ∆Hµ is the change in the heat of adsorption due to the presence of microwaves
(-(Haµ-Ha)), and ∆Tµ is the change in temperature due to the presence of microwaves (TeµT). The change in temperature due to microwaves can be expressed as
∆Tµ =
∆H µ Teµ
Q
and the effective temperature can be expressed as
71
Teµ =
T
1−
∆H µ
Q
The change in the amount adsorbed due to microwaves (and change in enthalpy of
the adsorbed phase due to microwaves ∆Hµ) is expected to be function of: 1) the
adsorbate used, 2) the adsorbent used, 3) the amount adsorbed (θ), 4) the permittivity of
the adsorbed phase (which is a function of the amount adsorbed for θ < 3 [63]), and 5) the
strength of the electric field. The amount adsorbed (θ) is itself a function of temperature,
the adsorbate used, the adsorbent used, and the pressure of the adsorbate in the gas phase.
It is expected that ∆Hµ is proportional to ε” and proportional to |E|2, due to Poynting’s
Theorem, the average dielectric power dissipated in a volume of a medium [4].
Pl = (½) ω ε” |E|2
As a first assumption, if microwave power is only dissipated in the adsorbed
phase, the amount of heat generated at steady state will be
Qg = (½) ω ε” |E|2 * Vm * θ
Where Qg is the heat generated by microwaves, Vm is the monolayer volume, and θ is the
fractional surface coverage.
As a first approximation, the permittivity of the adsorbed phase is assumed to be
the same as that of the liquid phase of the adsorbate, and is generally valid for θ > 3 [63].
72
However, very few measurements of the permittivity of adsorbed phases and of 2D
surfaces are available in the literature. As the amount adsorbed decreases, the permittivity
of the adsorbed phase is expected to be lower than that of the bulk liquid phase, to a value
of the permittivity of the surface as theta approaches zero. Thorp [64] studied water and
benzene on alumina and suggested that unless the adsorbent surface polarized the
monolayer to any great extent, the value of the permittivity would be expected to be equal
to that of the solid. Thiebaut et al. [65] suggested that for small values of θ, ε’ increases
linearly with θ, and ε” increases by a polynomial relationship to the second power for
water on X13 zeolite. It would still be expected that for two adsorbates of different
permittivities adsorbed to the same value of theta, their permittivities would be
proportional to their bulk liquid permittivities. Permittivity is also a function of
temperature, and as a first approximation can be treated as being proportional to the bulk
liquid permittivity at room temperature. Although the permittivity of bulk liquids changes
with temperature, Thiebaut et al. [65] found that the permittivity of moistened zeolite
showed little variation with temperature compared to the expected corresponding
variation for bulk liquid water.
Figure 5.11-2 shows the change in effective temperature versus the imaginary part
of the bulk liquid permittivity of the adsorbate used. It shows a linear trend for each
adsorbent because the change in effective surface temperature is proportional to the bulk
liquid permittivities, but the slopes and intercepts depend on the microwave power used
and the surface coverage of the adsorbate. Also, the y-intercepts are much closer to zero
for silicalite than silica; this indicates that with no adsorbate present, the silica absorbs
more microwave energy than silicalite. This is reasonable since the permittivity of the
73
silica is expected to be higher than that of silicalite due to the higher concentration of
hydroxyl groups on the surface of silica [51]. The experiments with methanol on
methoxylated Aerosil 200 silica were not included because chemisorption takes place
with that particular adsorbate/adsorbent pair and alters the surface of the adsorbent, also
changing the interactions for physical adsorption and preventing a direct comparison.
74
Figure 5.11-2
Change in effective surface temperature versus bulk liquid permittivity
for Aerosil 200 silica (top) and silicalite (bottom)
Change in Effective
Surface Temperature (C)
Aerosil 200 Silica - Change in Effective Surface
Tem perature vs. Bulk Liquid Perm ittivity
40
60W
theta
0.85
30
120W
theta
0.68
20
10
240W
theta
0.3
0
0
1
2
3
Perm ittivity (e")
4
Silicalite - Change in Effective Surface
Temperature vs. Bulk Liquid Permittivity
Change in Effective Surface
Temperature (C)
40
30W
theta
1.2
30
60W
theta
1.15
20
120W
theta
1.1
10
0
0
1
2
3
Permittivity (e")
75
4
The power delivered to the system by the microwave oven is approximately
proportional to the square of the average electric field, so the amount of heat generated
should be approximately proportional to the power delivered. This will not be exact
because the electric field within the adsorbate will not necessarily be proportional to the
average electric field delivered in the oven, because the sample will distort the electric
field in the oven to varying amounts depending on the permittivity of the adsorbate, the
amount adsorbed, and the geometry of the sample. The increase in effective temperature
with microwave power shown in figures 5.11-4 and 5.11-6 shows this general trend.
At steady state, the amount of heat generated by the adsorbed layer in the
presence of microwaves must be transferred through the glass sample tube. The cooling
air over the outside of the glass sample tube was adequate to transfer the heat in most
cases, except when a high microwave power and high permittivity adsorbate were used.
As a first approximation, the heat transfer can be described as being transferred by
convection from the adsorbed phase to the gas phase using a heat transfer coefficient,
Qg = h A ( Teµ – T )
where h is the heat transfer coefficient, and A is the surface area through which heat is
transferred. T is the temperature of the gas phase and is assumed to be the same
temperature as the cooling air on the outside of the glass sample tube. It is also assumed
that the “effective” Teµ for adsorption will also be the same temperature for heat transfer
purposes. The area of heat transfer from the adsorbate to the gas phase is the sample
surface area times theta for θ < 1 and the surface area for θ > 1, however, some of the
76
heat conducted into the solid phase may be convected to the gas phase when θ < 1. For
heat transfer by convection from and adsorbed layer to the gas phase, it would be
expected that h decreases as θ increases.
For adsorption in zeolites, it might also be expected that the network effects of a
zeolite such as silicalite would have an effect on the permittivity of the adsorbed phase,
changes with the area for heat transfer between the adsorbed phase and the gas phase, and
possibly the heat transfer coefficient.
The effective surface temperature can be expressed as a function of the fractional
surface coverage (θ) of the adsorbent by the adsorbate, instead of as a function of
pressure. Expressing the effective surface temperature as a function of surface coverage
provides a better comparison between adsorbates than expressing the effective surface
temperature as a function of pressure or relative pressure, because adsorption isotherms
reflect the nature of the interactions between adsorbate molecules and the adsorbing
surface, and layers of adsorbate molecules with themselves. As the fractional surface
coverage increases, the adsorbed phase will behave more like a liquid phase and be able
to absorb more microwave energy, since liquid phases generally have a much greater
permittivity than gasses [4].
Plots of the change in effective surface temperature versus the fractional surface
coverage were made. The surface area of the adsorbent was found from a multipoint BET
calculation using nitrogen. The monolayer volume for the adsorbate of interest (volume
of adsorbate to achieve coverage of one monolayer) was then found by accounting for the
difference in the surface area factor between the adsorbate of interest and nitrogen. The
volume adsorbed at a particular pressure and calculated effective surface temperature in
77
the presence of microwaves was then divided by the monolayer volume to get the
fractional surface coverage. Plots for Aerosil 200 silica are shown in figure 5.11-3
(grouped by microwave power) and figure 5.11-4 (grouped by adsorbate). Plots for
silicalite are shown in figure 5.11-5 (grouped by microwave power) and figure 5.11-6
(grouped by adsorbate).
Figure 5.11-3
Change in effective surface temperature versus coverage (θ) on Aerosil
200 silica (grouped by microwave power); isopropanol: filled diamonds,
dichloromethane: open squares, cyclohexane: filled triangles
120 Watts Microwave Power
60
50
40
30
20
10
0
60
Change in
Effective Surface
Temperature (C)
Change in Effective
Surface
Temperature (C)
30 Watts Microwave Power
50
40
30
20
10
0
0.0
0.2
0.4
0.6
0.8
1.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
1.2
Coverage (Theta)
Coverage (Theta)
Coverage (Theta)
78
240 Watts Microwave Power
60
Change in
Effective Surface
Temperature (C)
Change in
Effective Surface
Temperature (C)
60 Watts Microwave Power
60
50
40
30
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
50
40
30
20
10
0
0.0
0.2 0.4 0.6 0.8 1.0
Coverage (Theta)
1.2
Figure 5.11-4
Change in effective surface temperature versus coverage (θ) on Aerosil
200 silica (grouped by adsorbate); 30 W: filled circles, 60 W: open squares, 120 W filled
diamonds, 240 W open triangles
Cyclohexane on Silica
Isopropanol on Silica
60
50
Change in
Effective Surface
Temperature (C)
Change in Effective
Surface
Temperature (C)
60
40
30
20
10
0
50
40
30
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Coverage (Theta)
Change in Effective
Surface
Temperature (C)
Dichloromethane on Silica
60
50
40
30
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Coverage (Theta)
79
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Coverage (Theta)
Figure 5.11-5
Change in effective surface temperature versus coverage (θ) on
silicalite (grouped by microwave power); isopropanol: filled diamonds, dichloromethane:
open squares, n-pentane: filled circles
120 Watts Microwave Power
30 Watts Microwave Power
50
Change in Effective
Surface
Temperature (C)
Change in
Effective Surface
Temperature (C)
50
40
30
20
10
0
1.0
1.1
1.2
1.3
1.4
1.5
40
30
20
10
0
1.0 1.1 1.2 1.3 1.4 1.5 1.6
1.6
Coverage (Theta)
Coverage (Theta)
240 Watts Microwave Power
50
Change in
Effective Surface
Temperature (C)
Change in
Effective Surface
Temperature (C)
60 Watts Microwave Power
50
40
30
20
10
40
30
20
10
0
0
1.0 1.1 1.2 1.3 1.4 1.5 1.6
1.0 1.1 1.2 1.3 1.4 1.5 1.6
Coverage (Theta)
Coverage (Theta)
80
n-pentane on Silicalite
50
Isopropanol on Silicalite
50
Change in
Effective Surface
Temperature (C)
Change in Effective
Surface
Temperature (C)
Figure 5.11-6
Change in effective surface temperature versus coverage (θ) on
silicalite (grouped by adsorbate); 30 W: filled triangles, 60 W: open squares, 120 W filled
diamonds, 240 W open triangles
40
30
20
10
0
30
20
10
0
1.0 1.1 1.2 1.3 1.4 1.5 1.6
Coverage (Theta)
Dichloromethane on Silicalite
50
Change in
Effective Surface
Temperature (C)
40
40
30
20
10
0
1.0 1.1 1.2 1.3 1.4 1.5 1.6
Coverage (Theta)
81
1.0 1.1 1.2 1.3 1.4 1.5 1.6
Coverage (Theta)
Extrapolating the data to where no adsorbate is present (theta = 0) yields an
effective surface temperature that is greater than room temperature, and increases with
increasing microwave power. The assumption that the adsorbent is transparent to
microwave energy is good for low microwave power, because the permittivity of the
adsorbent is very low. Since it is not zero, there is a small increase in the effective surface
temperature when no adsorbate is present.
The effective surface temperature versus theta increases at a higher rate for
adsorbates with a higher permittivity, because as theta increases the permittivity of the
adsorbed phase becomes closer to the bulk liquid permittivity, and increases the ability of
the adsorbate to absorb microwave energy.
A steep slope (almost 90° angle) of effective surface temperature versus theta
corresponds to an isostere (volume adsorbed versus pressure) that would be near
horizontal. A slope of effective surface temperature versus theta greater than a 90°
indicates that the amount adsorbed is decreasing with increasing theta (which
corresponds to increasing pressure), and that the assumptions used to measure the amount
adsorbed are breaking down as described in in the results section with dichloromethane
on Aerosil 200 silica at 240W; it is suspected that when the volume adsorbed decreases
with an increase in pressure that the assumption that the gas phase is at a uniform
temperature is no longer valid. The air cooling over the sample at higher pressures is no
longer sufficient to keep the temperature of the gas phase at 22 C. The amount adsorbed
is measured by the difference of the total amount of adsorbate in the system and the
amount in the gas phase. Since the Ideal Gas Law is used to calculate the amount of
adsorbate in the system and the gas phase, a change the temperature of a given volume of
82
adsorbate will change the measured pressure, which leads to error in the calculated
volume adsorbed.
5.12
Conclusions
It has been shown that microwave heating is not the same as conventional heating,
and it is believed that this difference, the “microwave effect,” may be interpreted to be
due to selective, local heating. The temperature at the surface where sorption occurs is
“effectively” greater than the solid or gas temperature. In these studies, measurements of
the amounts of adsorption as functions of the partial pressures of a specific adsorbate in
the presence of microwave irradiation were related to the conventional adsorption
isotherms. Equating the adsorbate pressure required to achieve a specific coverage (an
isostere) in the presence of microwave irradiation to the amount adsorbed for a
conventional isotherm allowed for an estimate of the “effective” surface temperature in
the presence of microwaves. It was found that:
1.
The effective surface temperature of the adsorbed phase during adsorption in the
presence of microwaves can be found by relating the conventional adsorption
isotherms to isosteres obtained in the presence of microwaves.
2.
The assumptions used for the calculation of the amount adsorbed began to
deteriorate when using high microwave power and high permittivity adsorbates,
as the cooling of the sample was no longer sufficient to maintain the constant
temperature of the gas phase in the presence of microwaves.
83
3.
The effective surface temperature increased and was proportional to the bulk
liquid permittivities for adsorbates having a higher permittivity, for a particular
adsorbent, surface coverage (theta), and microwave power used.
4.
The effective surface temperature increased with increasing the microwave
power, approximately to the microwave power, for a particular adsorbent,
adsorbate, and relative pressure used.
5.
The adsorbates isopropanol and methanol formed plasma when a microwave field
was applied (30 W or greater) when they were at low pressures (below 10 torr).
6.
Methanol chemically reacted with the Aerosil 200 silica surface in the presence of
microwaves, altering the surface characteristics for adsorption. It was found that
methanol reacted to form a methoxyl group which replaced a hydroxide group on
the surface.
84
CHAPTER 6
RESULTS: COMPETITIVE ADSORPTION
In the case of competitive adsorption it is useful to think about what might be
expected to happen to changes in the surface coverage of the components due to a change
in temperature. Using a simple model such as that developed by Langmuir [7] the surface
coverage of each component (assuming a total relative surface coverage < 1 and that the
heats of adsorption of the components are not a function of surface coverage) can be
experessed as:
θA =
bA PA
1 + bA PA + bB PB
and
θB =
bB PB
1 + bB PB + bA PA
Where θA and θB are the surface coverage of their respective components, PA and
PB are the pressure, and bA and bB are:
b=
a1κ ( − ∆Hads RT )
e
z m v1
Where α1 condenstion coefficient; the fraction of incident molecules which
actually condense on the surface, κ is the constant from kinetic theory of gasses,
1
2
κ = L ( MRT )1 2
85
zm is number of sites per unit area, and v1 is the frequency of oscillation of the molecule
in a direction normal to the surface.
In a model such as the one presented here, if the pre-exponential factors of b are
held constant while the system temperature is increased, the component with a greater
heat of adsorption will desorb and the one with a smaller heat of adsorption will adsorb;
however, the pre-exponential factor of b for each species is generally not constant with
respect to temperature, so the selectivity of adsorption for a multi-component adsorption
system cannot be determined solely from the heat of adsorption of the components.
Experiments by Turner [1] were performed studying microwave radiation’s
influence on competitive sorption in zeolites (cyclohexane and methanol on silicalite).
The zeolite and cyclohexane had a low permittivity, while methanol had a high
permittivity. It was found that the adsorbate with the higher permittivity (methanol) was
desorbed selectively under microwave radiation. The adsorbed species with the higher
permittivity (methanol) was heated selectively since the rate of microwave energy
absorption was greater than the rate of heat transfer from the surface. Hence, there were
different “effective” temperatures in the adsorbed phase on the surface, and in the bulk
zeolite when under the influence of microwaves.
The frequency dependence of permittivity is not the same for all adsorbates. For
example, figure 6.0-1 shows that ε” for acetone and isopropanol have an opposite
dependence of ε” with microwave frequency.
86
Figure 6.0-1
Permittivity as a function of frequency for acetone and isopropanol
Permittivity vs. Frequency
9
Acetone
Isopropanol
8
Permittivity (e")
7
6
5
4
3
2
1
0
0
2
4
6
8
10
12
14
16
18
Frequency (GHz)
For the case of competitive adsorption with isopropanol and acetone in the
presence of microwaves, it is hypothesized that at low microwave frequencies
isopropanol will absorb more microwave energy and therefore be desorbed to a greater
extent from the adsorbent, shifting the adsorption selectivity toward acetone. Conversely,
at higher frequencies above 4.5 GHz, acetone will absorb more microwave energy and
therefore be desorbed to a greater extent from the adsorbent, shifting the adsorption
selectivity toward isopropanol.
87
6.1
Reactor Loaded with Low Surface Area Glass Beads
Experiments were carried out with low surface area glass beads (3 mm diameter)
loaded in the place of the adsorbent bed in the reactor. These were conducted to make
sure that the changes in partial pressure measured by the mass spectrometer were due to
changes in adsorption, and no changes took place due to the heating of the vapor phase.
Experiments were carried out with only helium flowing through the reactor with
conventional heating (figure 6.1-1), acetone and isopropanol with conventional heating
(figure 6.1-2), acetone and isopropanol with microwave heating at 2.45 GHz and 120 W
(figure 6.1-3), methanol and benzene with conventional heating (figure 6.1-4), and
methanol and benzene with microwave heating at 2.45 GHz and 120 W (figure 6.1-5). In
all cases, the results showed that the partial pressures did not change upon heating the
reactor, since there was very little adsorption taking place on the low surface area glass
beads in the reactor.
88
Figure 6.1-1
Helium on glass beads with conventional heating
He with 3mm glass beads
Conventional Heating
100
1.2E-05
1.0E-05
80
8.0E-06
70
60
6.0E-06
50
4.0E-06
Temperature (C)
partial pressure (Torr)
He
90
Bed
Temp
40
2.0E-06
30
0.0E+00
1175
Effluent
Temp
20
1195
1215
1235
1255
1275
1295
time (minutes)
Figure 6.1-2
Acetone and isopropanol on glass beads with conventional heating
2.0E-06
110
1.8E-06
100
1.6E-06
90
1.4E-06
80
1.2E-06
70
1.0E-06
`
60
8.0E-07
50
6.0E-07
4.0E-07
40
2.0E-07
30
0.0E+00
750
20
770
790
810
830
time (minutes)
89
850
870
Acetone
temperature (C)
partial pressure (Torr)
Acetone and Isopropanol on glass beads
conventional heating
Isopropanol
Bed Temp
Effluent
Temp
Figure 6.1-3 Acetone and isopropanol on glass beads with microwave heating at
2.45GHz and 120W
Acetone and Isopropanol on glass beads
Microwave heating at 2.45 GHz and 120 W
3.5E-06
38
He
36
3.0E-06
32
Acetone
2.0E-06
30
1.5E-06
28
26
Temperature (C)
Partial Pressure (Torr)
34
2.5E-06
Isopropanol
Bed Temp
1.0E-06
24
5.0E-07
22
0.0E+00
1125
1150
1175
1200
Efflent
Temp
20
1250
1225
time (minutes)
Figure 6.1-4
Methanol and benzene on glass beads with conventional heating
3.5E-06
100
3.0E-06
90
He
methanol
80
2.5E-06
70
2.0E-06
60
1.5E-06
50
1.0E-06
40
5.0E-07
30
0.0E+00
1000
20
1020
1040
1060
1080
time (minutes)
90
1100
1120
1140
Temperature (C)
partial pressure (Torr)
methanol and benzene on glass beads
conventional heating
benzene
Bed
Temp
Effluent
Temp
Figure 6.1-5
and 120W
Methanol and benzene on glass beads with microwave heating at 2.45GHz
Methanol and benzene on 3mm glass beads
Microwave heating at 2.45 GHz and 120 W
40
3.0E-06
He
38
36
34
2.0E-06
32
1.5E-06
30
28
1.0E-06
26
benzene
Bed
Temp
24
5.0E-07
22
0.0E+00
790
methanol
Temperature (C)
partial pressure (Torr)
2.5E-06
Effluent
Temp
20
840
890
940
time (mninutes)
6.2
Methanol and Cyclohexane on Silicalite
To test the validity of the results, experiments were performed that were
comparable to those done by Turner[1]. In addition to observing the trends of the
adsorption of methanol and cyclohexane in the presence of microwaves, the change in
amount adsorbed was quantified. The adsorption of methanol and cyclohexane on
silicalite with conventional heating were also studied.
However, due to the very slow timescale for adsorption of cyclohexane on
silicalite, a single component isotherm for the adsorption of cyclohexane on silicalite was
not attainable due to the timescale in which cyclohexane adsorbs within silicalite.
Therefore, the actual surface coverage could not be calculated; but the change in amount
91
adsorbed could still be found by integrating the mass spectrometer signal as described in
the experimental section.
The changes in amount adsorbed due to conventional heating (figure 6.2-1) and
microwave heating at 2.45 GHz and 120 W (figure 6.2-2) are shown. The first series for
each component is the change in amount adsorbed upon heating, and the second is the
change in the amount adsorbed upon cooling. The results for methanol and cyclohexane
on silicalite show a change in molecules adsorbed/silicalite unit cell, since the surface
coverage could not be calculated for competitive adsorption of cyclohexane in these
studies.
For the conventional heating experiments, a rheostat and was used to control the
heating, and for the experiments using microwave heating, the heating was controlled by
using a fixed microwave power. This leads to a difference in temperature between
experiments. In comparing the changes in amount adsorbed between the experiment with
microwave heating and the one with conventional heating, the amount adsorbed must be
adjusted for the differences in the change in temperature in the experiments. It was
assumed that the heat of adsorption was constant with respect to surface coverage
(similar to a Langmuir-type isotherm), only physical adsorption was taking place, and
that the change in the amount adsorbed due to a change in temperature is proportional to
A*exp(-∆Hads/RT); where it is assumed that the pre-exponential factor A is not a
function of temperature. The change in amount adsorbed due to heating was adjusted to
what would be expected at 112.6 °C, since this was the highest temperature measured
during any of the experiments. It should be noted that the measured bed temperature
might not be the same as the effective surface temperature during sorption.
92
Tables with the changes in amount adsorbed both before and after adjusting for
the temperature are shown below. In both methods of heating, methanol desorbs and
cyclohexane adsorbs, although with conventional heating the methanol is desorbed to a
lesser extent.
Table 6.2-1
Change in amount adsorbed for methanol and cyclohexane on silicalite
temperature
Silicalite
Silicalite
Conventional
107.8
2.45 GHz
59.5
change in amount adsorbed w/ heating (molecules/silicalite unit cell)
methanol
cyclohexane
-8.46
2.79
-6.89
1.38
Table 6.2-2 Change in amount adsorbed for methanol and cyclohexane on silicalite,
adjusted for changes in temperature
Silicalite
Silicalite
Conventional
2.45 GHz
change in amount adsorbed w/ heating (molecules/silicalite unit cell)
methanol
cyclohexane
-8.84
2.91
-15.12
3.03
93
Figure 6.2-1
Methanol and cyclohexane on Silicalite with conventional heating
Methanol and Cyclohexane on Silicalite
Conventional Heating
120
8
100
4
80
2
0
800
-2
850
900
100060
950
40
-4
Bed Temperature (C)
change in amount adsorbed
(molecules / silicalite unit cell)
methanol 1
6
cyclohexane 1
methanol 2
cyclohexane 2
-6
20
-8
bed
temperature
0
-10
time (minutes)
Figure 6.2-2 Methanol and cyclohexane on Silicalite with microwave heating at
2.45GHz and120W
8
65
6
60
55
4
50
2
45
0
870
-2
methanol 1
890
910
930
950
970
990
1010
40
35
-4
Bed Temperature (C)
change in amount adsorbed
(molecules / unit cell)
Methanol and Cyclohexane on Silicalite
Microwave heating at 2.45 GHz and 120 W
cyclohexane 1
methanol 2
cyclohexane 2
30
-6
25
-8
20
time (minutes)
94
bed
temperature
6.3
Acetone and Isopropanol on Aerosil 200 Silica: Room Temperature
Experiments
The bulk of the experimental work was involving the adsorbate pair of acetone
and isopropanol. This was because it was hypothesized that since the bulk permittivity of
isopropanol is greater than acetone at 2.45 GHz, and the bulk permittivity of acetone is
greater than isopropanol at 5.8 GHz, there should be a change in adsorption selectivity
due to heating with the microwaves at different frequencies. The adsorption of acetone
and isopropanol was studied on both Aerosil 200 silica and silicalite.
In order to quantify the amount adsorbed, conventional single component
isotherms were measured on the volumetric adsorption system and are shown in figures
6.3-1 and 6.3-2. From the partial pressures known to exist in the flow adsorption system,
the volume adsorbed for a single component can then be calculated. The second
component is then added to the flow system, and the change in amount adsorbed of the
first component due to the second can be measured by integrating the mass spectrometer
signal. These experiments are then repeated for the second component. This allows the
calculation of the amount adsorbed when both species are present at the flow conditions
and room temperature. Changes in the amount adsorbed measured by integrating the
signal from the mass spectrometer, combined with the previous information allow the
calculation of the amount adsorbed for each of the components in the two component
system upon heating. The experiments performed to quantify the amount adsorbed at
room temperature are shown in figure 6.3-3 and 6.3-4.
The acetone had only a small effect on the surface coverage of isopropanol, and
there was almost no change in the bed temperature when the acetone was introduced.
95
Conversely, the isopropanol had a larger effect on the surface coverage of acetone and a
temperature change was observed when the isopropanol was introduced.
Although
the
literature values of the heats of adsorption vary for acetone and isopropanol on Aerosil
and there is some overlap in the literature values as shown in Appendix C. From the
above experiment it is suspected that isopropanol has a higher heat of adsorption on
Aerosil due to a greater amount of hydrogen bonding of the isopropanol to surface
hydroxyl groups.
One concern from these experiments is that the integrated area of signal from the
mass spectrometer for the changes in adsorption of one species due to the second species
being introduced was not exactly the same as the opposite change when the second
species was removed. The experimental error in this may occur due to the difficulties of
trying to maintain the same partial pressure of flow of the first component while
instantaneously adding the second component in the stream flowing to the reactor. When
the second component is introduced, a valve is manipulated to switch it from a stream bypassing the reactor to one that combines with the flow of the first component and the
helium diluent. At the same time the helium diluent flow must be reduced by the same
amount that the second component is flowing to maintain the same partial pressure of the
first component being introduced to the reactor by manipulating a potentiometer for the
flow controller of the helium diluent. Error may occur due to differences in the second
component flow and the amount that the helium diluent is changed, the time it takes to
adjust the helium diluent, and from having a small “dead volume” of the system when the
valve was switched introducing the second adsorbate that may not be the same
concentrations as the second component flow, due to the way the system plumbing was
96
constructed.
Figure 6.3-1
Acetone on Aerosil 200 at 24 C
Acetone on Silicalite, 24C
60
Volume Adsorbed (cc STP)
50
40
30
20
10
0
0
0.2
0.4
0.6
Relative Pressure (P/P0)
97
0.8
1
Figure 6.3-2
Isopropanol on Aerosil 200 at 22 C
Isopropanol on Aerosil 200 Silica
20
Volume Adsorbed (cc/g)
18
16
14
12
10
8
6
4
2
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Relative Pressure (P/Po)
Figure 6.3-3
Silica
Change in isopropanol adsorption due to acetone flow on Aerosil 200
Isopropanol adsorption infuenced by acetone on Aerosil 200 Silica
36
1.82
34
amount adsorbed
(molecules/nm^2)
32
1.78
30
28
1.76
26
1.74
24
1.72
1.70
850
22
870
890
910
930
time (minutes)
98
950
20
970
Isopropanol 1
Bed Temperature (C)
1.80
Isopropanol 2
Bed
Temperature
Figure 6.3-4
Silica
Change in acetone adsorption due to isopropanol flow on Aerosil 200
Acetone adsorption influenced by isopropanol on Aerosil 200 Silica
30
2.3
29
2.2
acetone 1
2.1
27
26
2.0
25
1.9
24
23
1.8
Bed Temperature (C)
amount adsorbed
(molecules/nm^2)
28
acetone 2
22
1.7
1.6
890
21
910
930
950
970
990
1010
Bed Temp
20
1030
time (minutes)
6.4
Acetone and Isopropanol on Aerosil 200 Silica: Conventional Heating
Changes in the amount adsorbed due to conventional heating are shown in figure
6.4-1. As the reactor is heated, the isopropanol desorbs. The acetone also shows a small
amount of desorption initially, but once more of the isopropanol desorbs, there is more
surface available and the acetone adsorbs.
99
Figure 6.4-1
Acetone and Isopropanol on Aerosil 200 Silica with Conventional Heating
Acetone and Isopropanol on Aerosil 200 Silica
Conventional Heating
2.0
100
acetone 1
1.9
90
1.8
70
1.6
60
1.5
1.4
50
1.3
40
Bed Temperature (C)
amount adsorbed
(molecules/nm^2)
80
1.7
isopropanol 1
acetone 2
isopropanol 2
1.2
30
1.1
1.0
850
900
950
1000
1050
20
1100
Bed Temp
time (minutes)
6.5
Acetone and Isopropanol on Aerosil 200 Silica: Microwave Heating at 2.45
GHz and 120 W; and 5.8 GHz and 20 W
The adsorptions of acetone and isopropanol on Aerosil 200 with microwave
heating at 2.45 GHz and 120 W (figure 6.5-1) and 5.8 GHz and 20 W (figure 6.5-2) are
shown. In comparing the change in surface coverage for these experiments, however, the
temperatures at which the reactor was heated to during each experiment were different.
This must be taken into account and was done in a manner similar to that as described for
methanol and cyclohexane; the change in the amount adsorbed due to a change in
temperature is proportional to e(-∆Hads/RT).
100
Since the change in amount adsorbed depends on how much is initially adsorbed, it is the
ratios of the amount adsorbed at steady state during heating (after adjusting for the
differences in temperature) to that before heating that are compared.
The tables below show the changes in surface coverage due to heating with and
without adjusting for the differences in temperature. Th2/Th1 is the amount adsorbed at
steady state during heating divided by the amount adsorbed at steady state before heating.
Table 6.5-1
Ratio of the amount adsorbed for acetone and isopropanol on Aerosil
Temperature
acetone
isopropanol
Aerosil
Conventional
86.3
Th2/Th1
1.10
0.68
Aerosil
2.45 GHz
59.6
Th2/Th1
1.14
0.85
Aerosil
5.8 GHz
42.0
Th2/Th1
1.17
0.94
Table 6.5-2 Ratio of the amount adsorbed for acetone and isopropanol on Aerosil with
the amount adsorbed during heating adjusted for temperature differences
acetone
isopropanol
Aerosil
Conventional
Th2/Th1
1.14
0.57
Aerosil
2.45 GHz
Th2/Th1
1.30
0.68
Aerosil
5.8 GHz
Th2/Th1
1.73
0.73
The use of microwaves did not change which component desorbs upon heating
compared to conventional heating. A significant portion of the microwave energy may be
absorbed by the adsorbent and then transferred to the adsorbed phase, leading to results
similar to that of conventional heating. The ratio of the bulk permittivity of isopropanol
to acetone is only 3.15 (compared to a factor of 275 for methanol to cyclohexane) at 2.45
GHz.
101
It was expected based on the bulk permittivity frequency dependence that
microwave heating at 5.8 GHz would cause the acetone to desorb and the isopropanol to
adsorb. The ratio of the bulk liquid permittivities for acetone to isopropanol is at 5.8 GHz
is 1.71. At 5.8 GHz the isopropanol was still desorbed and the acetone adsorbed. It is
possible that since the surface coverage is low, the dielectric properties of the adsorbates
are not the same as the respective bulk liquid permittivities, and is expected to be lower
than the measured bulk permittivities. The adsorption selectivity might also depend on
the frequency dependence of the permittivity of the surface, especially for Aerosil 200
due to the surface hydroxyl groups.
Unlike the adsorbate pair of methanol and cyclohexane (which are immiscible),
acetone and isopropanol are completely miscible as bulk liquids. If the adsorbed phase is
also miscible, the adsorbed phase containing both species might behave as if it were a
solution with a single permittivity that is intermediate of both acetone and isopropanol.
The amount of energy absorbed by the adsorbed phase when exposed to microwaves
would still be a function of microwave frequency; however, the microwave frequency
would no longer directly influence the adsorption selectivity if the multi-component
adsorbed phase behaved with a single average permittivity.
Differences in adsorption due to the change in microwave frequency may differ
from those expected if the frequency dependence of the adsorbed phase at low surface
coverage is not the same as the bulk liquids. Based on the measured bulk liquid
permittivities, it was expected that isopropanol would absorb more microwave energy
and desorb more at 2.45 GHz, and acetone would absorb more microwave energy and
desorb more at 5.8 GHz.
102
Figure 6.5-1
and 120 W
2.00
Acetone and Isopropanol on Aerosil 200 with MW heating at 2.45 GHz
Acetone and Isopropanol on Aerosil 200 Silica
Microwave heating at 2.45 GHz and 120 W
65
acetone 1
60
1.90
50
1.80
45
1.70
40
35
1.60
Temperature (C)
amount adsorbed
(molecules/mn^2)
55
isopropanol 1
acetone 2
isopropanol 2
30
1.50
25
1.40
940
960
980
1000
time (minutes)
103
1020
20
1040
Bed
temperature
Figure 6.5-2 Acetone and Isopropanol on Aerosil 200 with microwave heating at 5.8
GHz and 20 W
amount adsorbed
(molecules/nm^2)
2.0
40
1.9
35
1.8
30
1.7
isopropanol 1
acetone 2
isopropanol 2
25
1.6
1.5
830
acetone 1
Bed Temperature (C)
2.1
Acetone and Isopropanol on Aerosil 200 Silica
Microwave heating at 5.8 GHz and 20 W
45
20
850
870
890
910
bed
temperature
930
time (minutes)
6.6
Acetone and Isopropanol on Silicalite: Room Temperature Experiments
In order to quantify the amount adsorbed, conventional single component
isotherms were measured on the volumetric adsorption system and are shown in figures
6.6-1 and 6.6-2. From the partial pressures known to exist in the flow adsorption system,
the volume adsorbed for a single component can then be calculated. The second
component is then added to the flow system, and the change in amount adsorbed of the
first component due to the second can be measured by integrating the mass spectrometer
signal. These experiments are then repeated for the second component. This allows the
calculation of the amount adsorbed when both species are present at the flow conditions
and room temperature. Changes in the amount adsorbed measured by integrating the
104
signal from the mass spectrometer, combined with the previous information allow the
calculation of the amount adsorbed for each of the components in the two component
system upon heating. The experiments performed to quantify the amount adsorbed at
room temperature are shown in figures 6.6-3 and 6.6-4. On silicalite, the acetone had a
much larger effect on the adsorption of isopropanol. This is partially because acetone has
a greater heat of adsorption on silicalite. Silicalite is hydrophobic, in contrast to the
Aerosil, which has many hydroxide groups on its surface.
It is noted that the integrated area of signal from the mass spectrometer for the
changes in adsorption of acetone due to isopropanol being introduced was not exactly the
same as the opposite change when Isopropanol was removed. In addition to the
experimental error due to the experimental difficulties mentioned with acetone and
isopropanol on Aerosil, it is possible that a reaction is taking place with the acetone and
possibly the isopropanol on silicalite. This will be discussed in more detail in the later
sections. The changes in the room temperature of the lab can also be noticed in these
experiments. The room temperature in the lab fluctuates between 25-28°C.
105
Figure 6.6-1
Acetone on Silicalite at 24 C
Acetone on Silicalite, 22C
60
Volume Adsorbed (cc STP)
50
40
30
20
10
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Relative Pressure (P/P0 )
Figure 6.6-2
Isopropanol on Silicalite at 24 C
Isopropanol on Silicalite
40
Volume Adsorbed (cc STP)
35
30
25
20
15
10
5
0
0
0.2
0.4
0.6
Relative Pressure (P/P0)
106
0.8
1
Figure 6.6-3
Change in acetone adsorption due to isopropanol flow on Silicalite
Acetone adsorption influenced by isopropanol on silicalite
34
12.5
acetone 1
amount adsorbed
(molecules/silicalite unit cell)
32
30
11.5
28
26
11.0
24
Bed Temperature (C)
12.0
acetone 2
10.5
Bed
Temperature
22
10.0
1070
20
1120
1170
1220
time (minutes)
Figure 6.6-4
Change in isopropanol adsorption due to acetone flow on Silicalite
7.8
34
7.6
32
isopropanol 1
7.4
30
7.2
28
7.0
26
6.8
24
6.6
22
6.4
6.2
800
820
840
860
time (minutes)
107
20
880
Bed Temperature (C)
amount adsorbed
(molecules/silicalite unit cell)
Isopropanol adsorption influenced by acetone on silicalite
bed
temperature
6.7
Acetone and Isopropanol on Silicalite: Conventional Heating
The adsorptions of Acetone and Isopropanol on Silicalite are shown in Figure
6.7-1.
In contrast to the previous case on Aerosil, acetone has a greater heat of
adsorption than isopropanol on silicalite. Also in contrast to the case on Aerosil, at the
starting flow conditions used, acetone has a higher surface coverage than isopropanol.
When the sample is heated conventionally it is again isopropanol that desorbs more upon
conventional heating. As the reactor is heated, both acetone and isopropanol initially
begin to desorb, but as more isopropanol desorbs, acetone re-adsorbs on the surface as the
temperature increases.
As the reactor temperature decreases, the surface coverage for both components
increases. As the reactor heating is turned off and the reactor cools, the acetone and
isopropanol should return to the same surface coverage as the steady state before the
reactor was heated. This is approximately true for isopropanol but not for acetone in this
case. From integrating the changes in the amount of acetone present in the system,
acetone must have a higher surface coverage for the same flow conditions and
temperature after the reactor had been heated.
It is possible that there is some degree of chemisorption or a surface reaction was
taking place. There was no evidence of any chemical reaction taking place in the form of
new molecules being observed with the online mass spectrometer; however, the acetone
and isopropanol both are fragmented in the mass spectrometer and it is the fragmentation
peaks that are tracked to quantify the adsorption, and it is possible that the fragmentation
may mask new products. There were no abnormal temperature effects observed from the
fiber optic temperature probe in the adsorbent bed. In order to examine whether there was
108
any reaction was taking place, the experimental setup was modified to put a liquid
nitrogen trap on the effluent stream to condense it. The condensate was studied by
GCMS, but no components besides the adsorbates were found. It is possible that the
adsorbate molecules that react with the surface are bound there. This could also change
the surface properties over extended re-use of the silicalite; however we did not have
sufficient time to fully examine this. The silicalite adsorbent was periodically re-calcined.
Figure 6.7-1
Acetone and Isopropanol on Silicalite with conventional heating
Acetone and Isopropanol on Silicalite
Conventional Heating
90
10.9
Acetone 1
amount adsorbed
(molecules/silicalite unit cell)
9.9
70
9.4
60
8.9
8.4
50
7.9
40
7.4
Bed Temperature (C)
80
10.4
isopropanol 1
Acetone 2
isopropanol 2
6.9
30
6.4
Bed Temp
5.9
1180
20
1200
1220
1240
1260
1280
1300
time (minutes)
6.8
Acetone and Isopropanol on Silicalite: Microwave heating at 2.45 GHz and
120W and 5.8 GHz and 20 W
The adsorptions of acetone and isopropanol on silicalite with microwave heating
at 2.45 GHz and 120 W (figure 6.8-1) and 5.8 GHz and 20 W (figure 6.8-2) are shown.
109
In comparing the change in surface coverage for these experiments, however, the
temperatures at which the reactor was heated to during each experiment were different.
This must be taken into account and was done in a manner similar to that as described
before; the change in the amount adsorbed due to a change in temperature is proportional
to e(-∆Hads/RT). Since the change in amount adsorbed depends on how much is initially
adsorbed, it is the ratios of the amount adsorbed at steady state during heating (after
adjusting for the differences in temperature) to that before heating that are compared.
The tables below show the changes in surface coverage due to heating with and
without adjusting for the differences in temperature. Th2/Th1 is the amount adsorbed at
steady state during heating divided by the amount adsorbed at steady state before heating.
Table 6.8-1
silicalite
Ratio of the amount adsorbed upon heating for acetone and isopropanol on
Temperature
acetone
isopropanol
Silicalite
Conventional
70.0
Th2/Th1
1.04
0.96
Silicalite
2.45 GHz
61.2
Th2/Th1
0.98
0.91
Silicalite
5.8 GHz
73.2
Th2/Th1
0.92
0.82
Table 6.8-2 Ratio of the amount adsorbed upon heating for acetone and isopropanol on
silicalite, adjusted for temperature differences
acetone
isopropanol
Silicalite
Conventional
Th2/Th1
1.08
0.93
Silicalite
2.45 GHz
Th2/Th1
0.95
0.82
Silicalite
5.8 GHz
Th2/Th1
0.86
0.71
When the reactor is heated using microwave heating, both components initially
desorb, similar to conventional heating, but the acetone does not re-adsorb as more of the
isopropanol desorbs. This is due to the much faster temperature ramp when microwave
110
heating is used.
The use of microwaves did not change which component desorbs upon heating
compared to conventional heating. The acetone did seem to absorb more microwave
energy at 5.8 GHz, but not to as great an extent as expected. The reasons for this are
similar to those on Aerosil; it is possible that since the surface coverage is low, the
dielectric properties of the adsorbates are not the same as the respective bulk liquid
permittivities and is expected to be lower than the measured bulk permittivities, and the
frequency dependence of the permittivity may also be different. From previous work,
with silicalite it is expected that a smaller portion of the microwave energy be adsorbed
by the adsorbent compared to Aerosil, so larger differences between conventional heating
and microwave heating were evident using silicalite.
Also, as stated before, acetone and isopropanol are completely miscible as bulk
liquids. If the adsorbed phase is also miscible, the adsorbed phase containing both species
might behave as if it were a solution with a single permittivity that is intermediate of both
acetone and isopropanol.
As before, as the reactor heating is turned off and the reactor cools, the acetone
and isopropanol should return to the same surface coverage as the steady state before the
reactor was heated if only reversible physical adsorption is taking place. In the case of
microwave heating on silicalite the surface coverage for both species is higher after the
reactor has been heated, indicating that some degree of change has taken place.
It
is
suspected that there is some degree of chemisorption or a surface reaction that was taking
place. The heating method used does seem to have an effect on the change in adsorption
of the adsorbates and may be related to a reaction rate.
111
Figure 6.8-1
and 120 W
Acetone and Isopropanol on Silicalite: Microwave heating at 2.45 GHz
11.0
65
10.5
60
10.0
55
9.5
50
9.0
8.5
45
8.0
40
7.5
35
7.0
30
6.5
isopropanol 1
acetone 2
isopropanol 2
25
6.0
5.5
825
acetone 1
Bed Temperature (C)
amount adsorbed
(molecules/silicalite unit cell)
Acetone and Isopropanol on Silicalite
Microwave heating at 2.45 GHz and 120W
875
925
time (minutes)
112
20
975
Bed Temp
Figure 6.8-2
20 W
Acetone and Isopropanol on Silicalite: Microwave heating at 5.8 GHz and
Acetone and Isopropanol on Silicalite
Microwave heating at 5.8 GHz and 20 W
80
12
acetone 1
70
10
60
9
50
8
40
7
isopropanol 1
acetone 2
isopropanol 2
30
6
5
920
Bed Temperature (C)
amount adsorbed
(molecules/silicalite unit cell)
11
20
940
960
980
1000
1020
bed
temperature
1040
time (minutes)
6.9
Methanol and Benzene on Silicalite; Room Temperature Experiments
Competitive adsorption experiments with methanol and benzene on silicalite were
compared to the previous work, and done to help further simulations by the Auerbach
research group at Umass.
Like the experiment using acetone and isopropanol, in order to quantify the
amount adsorbed for the multicomponent system, conventional single component
isotherms were measured on the volumetric adsorption system and are shown in figures
6.9-1 and 6.9-2. From the partial pressures known to exist in the flow adsorption system,
the volume adsorbed for a single component flowing can then be calculated. The second
component is then added to the flow system, and the change in amount adsorbed of the
first component due to the second can be measured by integrating the mass spectrometer
113
signal. These experiments are then repeated for the second component. This allows the
calculation of the amount adsorbed when both species are present at the flow conditions
and room temperature. The amount adsorbed for each of the components in the two
component system upon heating is calculated from changes in the amount adsorbed
measured by integrating the signal from the mass spectrometer, combined with the
previous information. The experiments performed to quantify the amount adsorbed at
room temperature are shown in figures 6.9-3 and 6.9-4.
Figure 6.9-1
Methanol on Silicalite at 24 C
Methanol on Silicalite
80
Volume Adsorbed (cc STP)
70
60
50
40
30
20
10
0
0
0.1
0.2
0.3
0.4
0.5
Relative Pressure (P/P0)
114
0.6
0.7
0.8
Figure 6.9-2
Benzene on Silicalite at 24 C
Benzene on Silicalite
40
Volume Adsorbed (cc STP)
35
30
25
20
15
10
5
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Relative Pressure (P/P0)
Figure 6.9-3
Change in methanol adsorption due to benzene flow on silicalite
methanol adsorption influenced by benzene on silicalite
14.4
40
38
36
14.0
34
32
13.8
30
13.6
28
26
13.4
24
13.2
22
13.0
760
780
800
820
time (minutes)
115
840
20
860
methanol 1
Bed Temperature (C)
amount adsorbed
(molecules/silicalite unit cell)
14.2
Bed
Temperature
Figure 6.9-4
Change in benzene adsorption due to methanol flow on silicalite
benzene adsorption influenced by methanol on silicalite
36
6.6
benzene 1
6.5
32
30
28
6.4
26
6.3
Bed Temperature (C)
amount adsorbed
(molecules/silicalite unit cell)
34
24
Bed
Temperature (C)
22
6.2
1030
20
1050
1070
1090
1110
time (minutes)
6.10
Methanol and Benzene on Silicalite; Conventional Heating
Experiments were performed with methanol and benzene measuring the changes
in the amount adsorbed due to conventional heating. The changes in the measured partial
pressures and the resulting surface coverage are shown in figures 6.10-1 and 6.10-2.
Since the experiments with the heating of methanol and benzene were only
studied with conventional heating, the amount adsorbed did not have to be adjusted for
any differences in temperature (the adjustments in volume adsorbed for the other
experiments were done to match this temperature).
116
The changes in the amount adsorbed in the table below are expressed in
molecules/silicalite unit cell to compare to the experiments with methanol and
cyclohexane on silicalite (top), and in the ratios of the amount adsorbed during heating to
that before heating (bottom) to compare to the other experiments.
Table 6.10-1 Change in amount adsorbed for methanol and benzene on silicalite, in
molecules/silicalite unit cell (top) and ratios of the amount adsorbed (bottom)
methanol
benzene
Silicalite
Conventional
112.6
change in amount adsorbed w/ heating (molecules/silicalite unit cell)
-3.89
3.04
methanol
benzene
Silicalite
Conventional
Th2/Th1
0.70
1.49
temperature
The results for the adsorption of methanol and benzene with conventional heating
were similar to that of methanol and cyclohexane; however, since the difference in the
heat of adsorption of methanol and benzene is smaller than that of methanol and
cyclohexane, the differences in the amount adsorbed were not as great. Also, as opposed
to methanol and cyclohexane (which are immiscible), methanol and benzene are
miscible. This may also contribute to the smaller difference in amount adsorbed. As the
reactor cools, both components begin to re-adsorb on the surface, but as the surface
coverage of methanol increases, the benzene begins to desorb.
The components do not quite return to the same steady state after heating has
taken place. After the reactor temperature again reaches room temperature, the surface
coverages for both components have increased, so some change in the adsorbent must
117
have taken place; or, it is possible that there is some degree of chemisorption or a surface
reaction was taking place. It is known that methanol can react with surface hydroxyl
groups to methoxylate the surface [60, 66]. This influences the surface properties for
adsorption. There was no evidence of any chemical reaction taking place in the form of
new molecules being observed from the online mass spectrometer. It is likely that the
adsorbate molecules that react with the surface are bound there. There were no abnormal
temperature effects observed from the fiber optic temperature probe in the adsorbent bed.
Figure 6.10-1 Partial pressures of methanol and benzene on silicalite with conventional
heating
methanol and benzene on silicalite
Conventional heating
6.0E-07
120
5.0E-07
100
4.0E-07
80
3.0E-07
60
2.0E-07
40
1.0E-07
20
0.0E+00
845
0
895
945
time (minutes)
118
Temperature (C)
partial pressure (Torr)
methanol
benzene
Bed
temperature
effluent
temperature
Figure 6.10-2 Methanol and benzene on silicalite with conventional heating
Methanol and Benzene on Silicalite
Conventional Heating
14
120
100
12
90
11
80
10
70
60
9
50
8
Bed Temperature (C)
amount adsorbed
(molecules/silicalite unit cell)
methanol 1
110
13
benzene 1
methanol 2
benzene 2
40
7
6
840
30
20
890
940
bed
temperature
990
time (minutes)
6.11
Results Summary and Change in Adsorption Comparison; Accounting for
Changes in Temperature
A summary of the changes in surface coverage is expressed in table 6.11-1. The
temperatures listed are the measured bed temperature in °C. Theta 1 is the initial amount
adsorbed at steady state and room temperature, theta 2 is the amount adsorbed at steady
state during heating, and theta 3 is the amount adsorbed at steady state at room
temperature after heating. Therefore, theta2/theta1 is the amount adsorbed as steady state
during heating divided by the amount adsorbed at steady state before heating.
Theta3/theta1 is the amount adsorbed at steady state after the reactor has been heated and
cooled to room temperature divided by the amount adsorbed at steady state before
heating. The results for methanol and cyclohexane on silicalite are shown as a change in
119
molecules adsorbed/silicalite unit cell, since the absolute amount adsorbed could not be
calculated. The results for methanol and benzene are reported as both a change in the
amount adsorbed (molecules adsorbed/silicalite unit cell) and in amount adsorbed for the
sake of comparison.
The right half of table 2 compares the steady states before and after heating has
occurred. If the adsorption is completely reversible, the adsorbent surface remains the
same, and the flow conditions are the same, then the ratios of the surface coverage after
heating divided by the surface coverage before heating should be 1.0. The change in
amount adsorbed should be zero. In the experiments performed, in some cases there was
a change is the surface coverage. The explanation for this is that since the flow conditions
are the same, there must be some degree of chemisorption or a reaction taking place, to a
small extent, but enough to slightly alter the surface characteristics. There has been
evidence for the ability of methanol to react with surface hydroxide groups to
methoxylate the surface[60, 66].
For the conventional heating experiments, a rheostat and was used to control the
heating, and for the experiments using microwave heating, the heating was controlled by
using a fixed microwave power. This leads to a difference in temperature between
experiments. It was noted that the microwave power required to achieve the same bed
temperature was much less when using the 5.8 GHz microwave generator with this
apparatus. This is due to the fact that when the 2.45 GHz generator is operating in the
waveguide, it travels in a single mode. When the 5.8 GHz microwaves are in the same
waveguide, however, it travels in a multi-mode due to the smaller wavelength of the 5.8
GHz microwaves.
120
In comparing the amount adsorbed between these experiments the amount
adsorbed must be adjusted for changes in temperature. It was assumed that the heat of
adsorption was constant with respect to surface coverage (similar to a Langmuir-type
isotherm), only physical adsorption was taking place, and that the change in the amount
adsorbed due to a change in temperature is proportional to A*exp(-∆Hads/RT); where it
is assumed that the pre-exponential factor A is not a function of temperature. The change
in amount adsorbed due to heating was adjusted to what would be expected at 112.6°C,
since this was the highest temperature measured during any of the experiments. The
changes in amount adsorbed (with the surface coverage adjusted for the changes in
temperature) are shown in table 6.11-2. It should be noted that the measured bed
temperature might not be the same as the effective surface temperature during sorption.
For the case of methanol and cyclohexane on silicalite, the methanol desorbs with
conventional heating, and the cyclohexane adsorbs. The desorption of methanol upon
heating is further accentuated by using microwave heating. This is because the ratio of
the bulk liquid permittivity for methanol to cyclohexane is very high (275) and the
methanol absorbs most of the microwave energy. Methanol and cyclohexane are
immiscible.
The results for the adsorption of methanol and benzene with conventional heating
were similar to that of methanol and cyclohexane; however, since the difference in the
heat of adsorption of methanol and benzene is smaller than that of methanol and
cyclohexane, the differences in the amount adsorbed were not as great. The miscibility of
methanol and benzene may also contribute to this.
For acetone and isopropanol on Aerosil 200, conventional heating yielded the
121
greatest desorption of isopropanol. The microwave heating at 2.45 GHz was not as
selective toward the heating of isopropanol as expected. The ratio of the bulk permittivity
of isopropanol to acetone is only 3.15 (compared to a factor of 275 for methanol to
cyclohexane) at 2.45 GHz.
It was expected based on the bulk permittivity frequency dependence that
microwave heating at 5.8 GHz would cause the acetone to desorb and the isopropanol to
adsorb. The ratio of the bulk liquid permittivities for acetone to isopropanol is at 5.8 GHz
is 1.71. At 5.8 GHz the isopropanol was still desorbed and the acetone adsorbed. The
reasons for this are that it is possible that since the surface coverage is low, the dielectric
properties of the adsorbates are not the same as the respective bulk liquid permittivities
and might be lower than the measured bulk permittivities. Also, from previous work[66],
the surface of the adsorbent Aerosil 200 absorbs a significant amount of the microwave
energy; which is then transferred to the adsorbates. This heating mechanism leads to
adsorption behavior that is similar to conventional heating, and the isopropanol is
desorbed more. The adsorption selectivity might also depend on the frequency
dependence of the permittivity of the surface, especially for Aerosil 200 due to the
surface hydroxyl groups. The smaller than expected change in adsorption selectivity with
microwave frequency might also be attributed to the miscibility of acetone and
isopropanol. If the adsorbed phase is also miscible, the adsorbed phase containing both
species might behave as if it were a solution with a single permittivity that is intermediate
of both acetone and isopropanol.
For the adsorption of acetone and isopropanol on silicalite, as the reactor is heated
conventionally, both acetone and isopropanol initially begin to desorb, but as more
122
isopropanol desorbs, acetone re-adsorbs on the surface as the temperature increases.
With microwave heating, the surface coverage of both adsorbates decreases upon
heating. This is because both components are absorbing the microwave energy
appreciably. Also, the silicalite absorbs less microwave energy than the Aerosil does.
This leads to a larger observed difference between microwave heating and conventional
heating when using silicalite. Changing the microwave frequency from 2.45 GHz to 5.8
GHz did not change the selectivity for adsorption of acetone and isopropanol. It was
expected that acetone would desorb more at 5.8 GHz (based on the bulk liquid
permittivities), but both the acetone and isopropanol desorbed more at 5.8 GHz than at
2.45 GHz. This was probably due to the fact that since the surface coverage is low, the
dielectric properties of the adsorbates would not the same as the respective bulk liquid
permittivities, and the frequency dependence of the permittivities for adsorbed species
would also be different than their respective bulk permittivity frequency dependences.
123
Table 6.11-1 Changes in amount adsorbed upon and after heating
Aerosil
Aerosil
Conventional 2.45 GHz
Temperature
86.3
59.6
Th2/Th1
Th2/Th1
acetone
1.10
1.14
isopropanol
0.68
0.85
Aerosil
5.8 GHz
42.0
Th2/Th1
1.17
0.94
Aerosil
Conventional
86.3
Th3/Th1
1.08
1.02
Aerosil
2.45 GHz
59.6
Th3/Th1
1.09
0.99
Aerosil
5.8 GHz
42.0
Th3/Th1
1.03
1.00
Silicalite
Silicalite
Conventional 2.45 GHz
Temperature
70.0
61.2
Th2/Th1
Th2/Th1
acetone
1.04
0.98
isopropanol
0.96
0.91
Silicalite
5.8 GHz
73.2
Th2/Th1
0.92
0.82
Silicalite
Conventional
70.0
Th3/Th1
1.09
0.99
Silicalite
2.45 GHz
61.2
Th3/Th1
1.07
1.06
Silicalite
5.8 GHz
73.2
Th3/Th1
1.15
1.08
Silicalite
Silicalite
Conventional 2.45 GHz
temperature
107.8
59.5
Silicalite
Silicalite
Conventional
2.45 GHz
107.8
59.5
change in amount adsorbed after heating
change in amount adsorbed w/ heating and cooling
methanol
-8.46
-6.89
-1.32
0.28
cyclohexane
2.79
1.38
-0.48
0.71
Silicalite
Conventional
temperature
112.6
Silicalite
Conventional
112.6
change in amount adsorbed after heating
change in amount adsorbed w/ heating and cooling
methanol
-3.89
0.86
benzene
3.04
0.56
Silicalite
Conventional
temperature
112.6
Th2/Th1
methanol
0.70
benzene
1.49
Silicalite
Conventional
112.6
Th3/Th1
1.07
1.09
124
6.11-2
Changes in amount adsorbed upon heating, adjusted for temperature
acetone
isopropanol
Aerosil
Conventional
Th2/Th1
1.14
0.57
Aerosil
2.45 GHz
Th2/Th1
1.30
0.68
Aerosil
5.8 GHz
Th2/Th1
1.73
0.73
acetone
isopropanol
Silicalite
Conventional
Th2/Th1
1.08
0.93
Silicalite
2.45 GHz
Th2/Th1
0.95
0.82
Silicalite
5.8 GHz
Th2/Th1
0.86
0.71
Silicalite
Silicalite
Conventional
2.45 GHz
change in amount adsorbed w/ heating (molecules/silicalite unit cell)
methanol
-8.84
-15.12
cyclohexane
2.91
3.03
methanol
benzene
methanol
benzene
Silicalite
Conventional
change in amount adsorbed w/ heating (molecules/silicalite unit cell)
-3.89
3.04
Silicalite
Conventional
Th2/Th1
0.70
1.49
125
6.12
Conclusions
Experiments were performed to study the effect of microwave frequency on
adsorption selectivity, and the differences between microwave and conventional heating.
It was found that:
1.
The adsorbent Aerosil 200 absorbed a significant amount of microwave energy.
This energy was then transferred to the adsorbates and contributed to a heating
mechanism that was similar to conventional heating. Larger differences between
conventional heating and microwave heating were observed when using silicalite.
2.
In the case of competitive adsorption with conventional heating the selectivity of
adsorption cannot be solely determined from the heats of adsorption of the
components.
3.
In the case of methanol and cyclohexane, microwave heating causes the methanol
to desorb more. This is because the methanol has a much greater permittivity (the
bulk liquid permittivities has a 275:1 ratio).
4.
The frequency dependence of the adsorption of acetone and isopropanol was not
as selective as expected based on the bulk permittivity differences (ratios
isopropanol:acetone 3.15:1 at 2.45 GHz and acetone:isopropanol 1.71:1 at 5.8
GHz). Since the surface coverage was low, the dielectric properties of the
adsorbates might not be the same as the respective bulk liquid permittivities, and
are expected to be lower than the measured bulk permittivities. Also, the
frequency dependence of the adsorbed phases may be different than that of the
bulk liquids. It might also depend on the frequency dependence of the permittivity
of the surface, especially for Aerosil 200 due to the surface hydroxyl groups. The
126
smaller than expected change in adsorption selectivity with microwave frequency
might also be attributed to the miscibility of acetone and isopropanol. If the
adsorbed phase is also miscible, the adsorbed phase containing both species might
behave as if it were a solution with a single permittivity that is intermediate of
both acetone and isopropanol.
5.
In some cases there was a slight change in the surface coverage after the reactor
was heated, compared to before the reactor was heated. There must be some
degree of chemisorption or a reaction taking place, to a small extent, but enough
to slightly alter the surface characteristics.
One cannot predict what will be selectively desorbed due to a change in
temperature based solely on the heats of adsorption of the adsorbates. The influence of
microwave frequency was not significant for changing selectivity; however, the
desorption efficiency appeared greater at 5.8 GHz than at 2.45 GHz. This effect may be
due to an inherent efficiency of microwaves to influence interfacial interactions in nonresonant applications wherein the exposure can vary in intensity with time.
127
APPENDIX A
NITROGEN ADSORPTION ISOTHERMS
Figure A-1
Nitrogen on Aerosil 200 silica
Nitro gen on Aeros il Silic a
160
Vol u me Adsorb ed (cc/g )
140
120
Ads
100
80
60
40
Des
20
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Re l ati ve Pressure (P/Po)
128
0.8
0.9
1.0
Figure A-2
Nitrogen on Silicalite
Nitrog en o n Silicalite @ 77K
160
Vol u me Adsorb ed (cc/g )
140
120
adsorption
100
80
60
desorption
40
20
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Rel a tive Pre ssu re (P/Po )
129
0.8
0.9
1.0
Figure A-3
Nitrogen on temperature treated Aerosil 200 silica
Nitro gen on Temp erature Treated (700C) Aeros il 200 Silica
180
160
Vol u me Adsorb ed (cc/g )
140
Ads
120
100
80
60
Des
40
20
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Re l ati ve Pressure (P/Po)
130
0.8
0.9
1.0
APPENDIX B
PERMITTIVITIES OF MATERIALS AS A FUNCTION OF FREQUENCY
Table B-1 Permittivity of materials at 2.45 GHz [61]
Microwave power
Aerosil 200 silica
silicalite
n-pentane
cyclohexane
dichloromethane
isopropanol
methanol
*values are +/- 0.1
relative permittivity (2.45 GHz at 22C)
*(e')
*(e")
1.43
-0.08
2.41
0.01
1.80
NA
2.02
0.05
9.08
0.39
18.30
3.19
23.04
13.77
131
Figure B-1
Permittivity of Aerosil 200 silica versus frequency at 22 C [57, 59]
1. 5
1. 45
e'
1. 4
1. 35
1. 3
1. 25
1. 2
0
2
4
6
8
10
12
14
Frequency [GHz]
0. 2
0. 1
0
e' '
- 0. 1
- 0. 2
- 0. 3
- 0. 4
- 0. 5
0
2
4
6
8
Fr equency [ GHz]
132
10
12
14
16
Figure B-2
Permittivity of silicalite versus frequency at 22 C [57, 59]
2. 55
2. 50
2. 45
2. 40
e'
2. 35
2. 30
2. 25
2. 20
2. 15
2. 10
2. 05
0
2
4
6
0
2
4
6
8
10
Fr equency [ GHz]
12
14
16
12
14
16
0. 4
0. 3
0. 2
e' '
0. 1
0
- 0. 1
- 0. 2
- 0. 3
- 0. 4
8
Fr equency [ GHz]
133
10
Figure B-3
Permittivity of cyclohexane versus frequency at 22 C [57, 59]
2. 4
2. 35
e'
2. 3
2. 25
2. 2
2. 15
0
2
4
6
8
10
12
14
16
10
12
14
16
Fr eq [ GHz]
0. 35
0. 30
0. 25
e' '
0. 20
0. 15
0. 10
0. 05
0. 00
0
2
4
6
8
Fr eq [ GHz]
134
Figure B-4
Permittivity of dichloromethane versus frequency at 22 C [57, 59]
10. 2
10. 0
9. 8
9. 6
e'
9. 4
9. 2
9. 0
8. 8
8. 6
8. 4
0
2
4
6
8
10
12
14
16
10
12
14
16
Fr eq [ GHz]
2. 0
1. 8
1. 6
1. 4
e' '
1. 2
1. 0
0. 8
0. 6
0. 4
0. 2
0. 0
0
2
4
6
8
Fr eq [ GHz]
135
Figure B-5
Permittivity of isopropanol versus frequency at 22 C [57, 59]
14
12
10
e'
8
6
4
2
0
0
2
4
6
8
10
12
14
16
18
12
14
16
18
Freque ncy (GHz)
9
8
7
6
e"
5
4
3
2
1
0
0
2
4
6
8
10
Freque ncy (GHz)
136
Figure B-6
Permittivity of methanol versus frequency at 22 C [57, 59]
35
30
25
e'
20
15
10
5
0
0
2
4
6
8
10
12
14
16
18
12
14
16
18
Fr eque ncy (GHz)
16
14
12
e"
10
8
6
4
2
0
0
2
4
6
8
10
Freque ncy (GHz)
137
Figure B-7
59]
Permittivity as a function of frequency for acetone and isopropanol [57,
Permittivity vs. Frequency
9
Acetone
Isopropanol
8
Permittivity (e")
7
6
5
4
3
2
1
0
0
2
4
6
8
10
12
Frequency (GHz)
138
14
16
18
APPENDIX C
LITERATURE VALUES FOR HEATS OF ADSORPTION
Table C-1 Literature values for heats of adsorption
Hads summary
Acetone
Isopropanol
methanol
cyclohexane
benzene
Aerosil 200
silicalite
Hads (kJ/mol)
50.2 [67], 56.1 [68], 59.2 [69]
56.4 [68], 60.5 [66]
50.2 [68]
30.3 [66]
40.6 [68]
Hads (kJ/mol)
67.0 [70, 71]
45.5 [71], 47.1[66]
43.0 [72]
63.0 [73]
52.0 - 57.8 [74]
139
BIBLIOGRAPHY
1.
Turner, M.D., Laurence, R.L., and Conner, W.C., "Microwave Radiation's
Influence on Sorption and Competitive Sorption in Zeolites". AIChE Journal.
46(4): p. 758-768 (2000).
2.
Turner, M.D., Laurence, R.L., Yngvesson, K.S., and Conner, W.C., "The effect of
microwave energy on three-way automotive catalysts poisoned by SO2".
Catalysis Letters. 71(3-4): p. 133-138 (2000).
3.
Ulaby, F.T., Fundamentals of Applied Electromagnetics. Prentice Hall. Upper
Saddle River, NJ (2001).
4.
Pozar, D.M., Microwave Engineering. John Wiley & Sons, Inc. Toronto (1998).
5.
Galema, S.A., "Microwave Chemistry". Chemical Society Reviews. 26: p. 233238 (1997).
6.
Stuerga and Gaillard, P., "Microwave Athermal Effects in Chemistry: A myth's
Autopsy". Journal of Microwave Power and Electromagnetic Energy. 31(2): p.
87-100 (1996).
7.
Gregg, S.J. and Sing, K.W., Adsorption, Surface Area and Porosity. Academic
Press Inc. San Diego (1982).
8.
Rouquerol, F., Rouquerol, J., and Sing, K., Adsorption by powders and porous
solids. Academic Press. London (1999).
9.
Fogler, H.S., Elements of Chemical Reaction Engineering. 3rd ed.: Prentice Hall
PTR. Saddle River, NJ (1999).
10.
Tester, J.W. and Model, M., Thermodynamics and Its Applications. 3rd ed.:
Prentice Hall. Upper Saddle River, NJ (1997).
11.
Gruermeur, R. and Jacolin, C., "Influence of Surface Silanols on Dielectric
Properties of Nitrogen Adsorbed on Activated Silica". Surface Science. 315: p.
323 (1994).
140
12.
Kobayashi, S., Kim, Y., Kenmizaki, C., Kushiyama, S., and Mizuno, K., "Control
of Adsorption by Microwave Irradiation". Chemistry Letters: p. 769-770 (1996).
13.
Lee, K., Experiments using a spectrum analyzer to measure permittivity versus
frequency, in Electrical and Computer Engineering. University of Massachusetts:
Amherst (2001-2003).
14.
Barthel, J., Bachhuber, K., Buchner, R., Gill, J.B., and Kleebauer, M., "Dielectric
Spectra of Some Common Solvents in the Microwave Region, Dipolar Aprotic
Solvents and Amides". Chemical Physics Letters. 167(1,2): p. 62-66 (1990).
15.
Barthel, J., Bachhuber, K., Buchner, R., and Hetzenauer, H., "Dielectric Spectra
of some common solvents in the microwave region, water and lower alcohols".
Chemical Physics Letters. 165(4): p. 369-373 (1990).
16.
Lopez, A., Diamy, E., Legrand, A., and Fraissard, J., "Sorption of Volatile
Organic Compounds on Zeolites with Microwave Irradiation". Studies in Surface
Science and Catalysis. 154B: p. 1866-1871 (2004).
17.
Conde, L.D., Marun, C., Suib, S.L., and Fathi, Z., "Frequency Effects in the
Catalytic Oligomerization of Methane via Microwave Heating". Journal of
Catalysis. 204: p. 234-332 (2001).
18.
Deng, S.G. and Lin, Y.S., "Microwave Heating Synthesis of Supported Sorbents".
Chemical Engineering Science. 52(10): p. 1563-1575 (1996).
19.
Kingston, H.M. and Haswell, S.J., Microwave-Enhanced Chemistry:
Fundamentals, Sample Preparation, and Applications. American Chemical
Society. Washington D.C. (1997).
20.
Gallis, K.W. and Landry, C.C., "Rapid Calcination of Nanostructured Silicate
Composites by Microwave Irradiation". Advanced Materials. 13(1): p. 23-26
(2001).
21.
Friedman, R.M. and Dannhardt, H.C., "Versatile Infared cell for in situ
transmission IR studies of heterogeneous catalysis". Rev. Sci. Instrum. 8: p. 15891592 (1985).
141
22.
White, R.L. and Nair, A., "Diffuse Reflectance Infared Spectroscopic
Characterization of Silica Dehydroxylation". Applied Spectroscopy. 44(1): p. 6975 (1990).
23.
Miecznikowski, A. and Hanuza, J., "Infared and Raman studies of ZSM-5 and
Silicalite-1 at room, liquid nitrogen, and helium temperatures". Zeolites. 7: p. 249254 (1987).
24.
Knozinger, H. and Huber, S., "IR spectroscopy of small and weakly interacting
molecular probes for acidic and basic zeolites". J. Chem. Soc. 94(15): p. 20472059 (1998).
25.
Holmes, S.M., Zholobenko, V.L., Thursfield, A., Plaisted, R.J., Cundy, C.S., and
Dwyer, J., "In-Situ FTIR study of the formation of MCM-41". J. Chem. Soc.
94(14): p. 2025-2032 (1998).
26.
Calabro, D.C., Valyocsik, E.W., and Ryan, F.X., "In Situ ATR/FTIR study of
mesoporous silicate synthesis". Microporous Materials. 7: p. 243-259 (1996).
27.
Tompsett, G., Laboratory Report. University of Massachusetts, Amherst (2004).
28.
Smith, R.A., Jones, F.E., and Chasmar, R.P., The Detection and Measurement of
Infra-red Radiation. Oxford University Press. Oxford (1960).
29.
Bittencourt, J.A., Fundamentals of Plasma Physics. Pergamon Press Inc.
Elmsford, NY (1986).
30.
Aliev, Y.M., Schluter, H., and Shivarova, A., Guided-Wave-Produced Plasmas.
Atoms + Plasmas, ed. G. Ecker. Springer-Verlag. Berlin (2000).
31.
Zgrablich, G., Mendioroz, S., Daza, L., Pajares, J., Mayagoitia, V., Rojas, F., and
Conner, W.C., "Effect of Porous Structure on the Determination of Pore Size
Distribution by Mercury Porosimetry and Nitrogen Sorption". Langmuir. 7(4): p.
779-785 (1991).
32.
Muller, J.A. and Conner, W.C., "Cyclohexane in ZSM-5, 1. FTIR and X-ray
Studies". J. Phys. Chem. 97(7): p. 1451-1454 (1992).
142
33.
Maglara, E., Pullen, A., Sullivan, D., and Conner, W.C., "Characterization of
Microporous Solids by Adsorption: Measurement of High-Resolution Adsorption
Isotherms". Langmuir. 10: p. 4167-4173 (1994).
34.
Kaminsky, R.D., Maglara, E., and Conner, W.C., "A Direct Assessment of MeanField Methods of Determining Pore Size Distributions of Microporous Media
from Adsorption Isotherm Data". Langmuir. 10: p. 1556-1565 (1994).
35.
Groen, J.C., Peffer, L.A.A., and Perez-Ramirez, "Pore size determination in
modified micro- and mesoporous materials. Pitfalls and limitations in gas
adsorption data analysis". Microporous and Mesoporous Materials. 60: p. 1-17
(2003).
36.
Halasz, I., Kim, S., and Marcus, B., "Hydrophilic and hydrophobic adsorption on
Y zeolites". Molecular Physics. 100(19): p. 3123-3132 (2002).
37.
Conner, W.C., Tompsett, G., and Vallee, S., Effect of Temperature and Vibration
on Isotherm Hysteresis. University of Massachusetts: Amherst (2003).
38.
Avery, R.G. and Ramsay, J.D., "The Sorption of Nitrogen in Porous Compacts of
Silica and Zirconia Powders". Journal of Colloid and Interface Science. 42(3): p.
597-606 (1973).
39.
Al-Muhtaseb, S.A. and Ritter, J.A., "Further Modification of the Antoine
Equation for Correlation of Adsorption Equilibria". Langmuir. 14(18): p. 53175323 (1998).
40.
Roussy, G., Zoulalian, A., Charreyre, M., and Thiebaut, J.-M., "How Microwaves
Dehydrate Zeolites". J. Phys. Chem. 88(23): p. 5702-5708 (1984).
41.
Heggs, P.J. and Bao, J., "Heat and Mass Transfer in an Adsorbent Pellet Within a
Microwave Field". The 1994 ICHEME Research Event: p. 1062-1065 (1994).
42.
Ayappa, K.G., "Modelling Transport Processes During Microwave Heating: A
Review". Reviews in Chemical Engineering. 13(2): p. 1-69 (1997).
43.
Ayappa, K.G., Davis, H.T., Davis, E.A., and Gordon, J., "Analysis of Microwave
Heating of Materials with Temperature-Dependent Properties". AIChE Journal.
37(3): p. 313-322 (1991).
143
44.
Blanco, C. and Auerbach, S.M., "Microwave-Driven Zeolite-Guest Systems Show
Athermal Effects from Nonequilibrium Molecular Dynamics". J. Am. Chem. Soc.
124: p. 6250-6251 (2002).
45.
Tang, J., Zhang, T., Ma, L., Li, N., Liang, D., and Lin, L., "Microwave-Assisted
Purification of Automotive Emissions". Journal of Catalysis. 211: p. 560-564
(2002).
46.
Kuroda, Y., Okamoto, T., Kumashiro, R., Yoshikawa, Y., and Nagao, M.,
"Microwave-assisted simple ion-exchange of ZSM-5-type zeolites with copper
ions and their specific adsorption properties for N2 molecules at room
temperature". Chem. Commun.: p. 1758-1759 (2002).
47.
Nagaraj, N., Narayan, A.V., Srinivas, N.D., and Raghavarao, K.S., "Microwavefield-assisted enhanced demixing of aqueous two-phase systems". Analytical
Biochemistry. 312: p. 134-140 (2003).
48.
Ania, C.O., Menendez, J.A., Parra, J.B., and Pis, J.J., "Microwave-induced
Regeneration of Activated Carbons Polluted with Phenol. A Comparison with
Conventional Thermal Regeneration". Carbon. 42: p. 1383-1387 (2004).
49.
Carrott, P.J., Nabais, J.M., Carrott, M.R., and Menendez, J.A., "Thermal
treatments of activated carbon fibres using a microwave furnace". Microporous
and Mesoporous Materials. 47: p. 243-252 (2001).
50.
Hirata, M., Kawasaki, N., Nakamura, T., Matsumoto, K., Kabayama, M., Tamura,
T., and Tanada, S., "Adsorption of Dyes onto Carbonaceous Materials Produced
from Coffee Grounds by Microwave Treatment". Journal of Colloid and Interface
Science. 254: p. 17-22 (2002).
51.
Kiselev, A.V. and Lygin, V.I., Infared Spectra of Surface Compounds, ed. D.
Slutzkin. Keter Publishing House Jerusalem Ltd. Jerusalem (1975).
52.
Van der Maas, J.H., Basic infrared spectroscopy. Sadtler Research Laboratories.
Philadelphia (1969).
53.
Conley, R.T., Infrared spectroscopy. Allyn and Bacon. Boston (1966).
144
54.
Hair, M.L., Infrared spectroscopy in surface chemistry. M. Dekker. New York,
NY (1967).
55.
Chen, H.Y., Chen, L., Lin, J., and Tan, K.L., "Copper Sites in Copper-Exchanged
ZSM-5 for CO Activation and Methanol Synthesis: XPS and FTIR Studies".
Inorganic Chemistry. 36: p. 1417-1423 (1997).
56.
Pelmenschikov, A.G., Morosi, G., Gamba, A., Zecchina, A., Bordiga, S., and
Paukshtis, E.A., "Mechanisms of Methanol Adsorption on Silicalite and Silica: IR
Spectra and ab-Initio Calculations". J. Phys. Chem. 97(46): p. 11979-11986
(1993).
57.
Wovchko, E.A., Camp, J.C., Glass, J.A., Jr., and Yates, J.T., Jr., "Active Sites on
SiO2: Role in CH3OH Decomposition". Langmuir. 11: p. 2592-2599 (1995).
58.
Conner, W.C., "Apparatus and Method for Efficient Determination of Equilibrium
Adsorption Isotherms at Low Pressures". United States Patent. 5637810 (1997).
59.
Lu, F., Experiments using a automatic network spectrum analyzer to measure
permittivity versus frequency, in Electrical and Computer Engineering. University
of Massachusetts, Amherst (2004).
60.
Meier, M., Studies on Desorption Under the Use of Microwave Energy in a 1
Meter Column, in Chemical Engineering. University of Massachusetts: Amherst
(2001).
61.
Lee, K., Experiments using a automatic network spectrum analyzer to measure
permittivity versus frequency, in Electrical and Computer Engineering. University
of Massachusetts: Amherst (2003).
62.
Kiselev, A.V., Lygin, V.I., and Solomonova, I.N., "Infrared Spectra of Aerosil
with a Chemically Modified Surface". Kolloidnyi Zhurnal. 26(3): p. 324-329
(1964).
63.
Jones, S.B. and Or, D., "Surface Area, Geometrical and Configurational Effects
on Permittivity of Porous Media". Journal of Non-Crystalline Solids. 305: p. 247254 (2002).
145
64.
Thorp, J.M., "The Dielectric Behavior of Vapors Adsorbed on Porous Solids".
Journal of Physical Chemistry. 66(6): p. 1086-1092 (1962).
65.
Thiebaut, J.-M., Akyel, C., Roussy, G., and Bosisio, R., "Dehydration and
Dielectric Permittivity Measurements of a Porous, Inorganic Material (13X
Zeolite) Heated with Microwave Power". IEEE Transactions on Instrumentation
and Measurement. 37(1): p. 114-120 (1988).
66.
Vallee, S.J. and Conner, W.C., "Microwaves and Sorption on Oxides: A Surface
Temperature Investigation". J. Phys. Chem. B. 110(31): p. 15459-15470 (2006).
67.
Hair, M.L., The Molecular Nature of Adsorption on Silica Surfaces. Xerox
Corporation: Rochester, NY.
68.
Bardina, I.A., Kovaleva, N.V., and Nikitin, Y.S., "The Adsorption Properties of
Initial and Modified Silica Gels". Russian Journal of Physical Chemistry. 74(3):
p. 421-425 (1999).
69.
Kokunova, E.Y., Lanin, S.N., Nikitin, Y.S., and Shoniya, N.K., "GasChromatographic Study of the Temperature Effect on Heats of Retention and
Adsorption for Oxygen Containing Organic Compounds on Silica Adsorbents".
Russian Journal of Physical Chemistry. 67(8): p. 1508-1513 (1992).
70.
Lee, C., Gorte, R.J., and White, D., "Carbonyl13C Shielding Tensors and Heats
of Adsorption of Acetone Adsorbed in Silicalite and the 1:1 Stoichiometric
Complex in H-ZSM-5 ". Journal of Physical Chemistry. 100(47): p. 1851518523 (1996).
71.
Gorte, R.J. and White, D., "Interactions of Chemical Species with acid sites in
zeolites". Topics in Catalysis. 14(1): p. 57-69 (1997).
72.
Thamm, "Calorimetric Study on the State of C-1-C-4 Alcohols Sorbed on Silicalite". Journal of the
Chemical Society - Faraday Transactions I. 85: p. 1-9 (1989).
73.
Cavalcante, "Adsorption of Branched and Cyclic Paraffins in Silicalite".
Industrial & Engineering Chemistry Research. 34(1): p. 177-184 (1995).
146
74.
Klemm, E., Wang, J., and Emig, G., "A comparative study of the sorption of
benzene and phenol in silicalite, HAlZSM-5 and NaAlZSM-5 by computer
simulation ". Microporous and Mesoporous Materials. 26(1-3): p. 11-21 (1998).
147
Документ
Категория
Без категории
Просмотров
0
Размер файла
4 290 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа