La lecture en ligne est gratuite
Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres
Télécharger Lire

Optimizing adsorbents for heat storage applications [Elektronische Ressource] : estimation of thermodynamic limits and Monte Carlo simulations of water adsorption in nanopores = Optimierung von Adsorbentien für Wärmespeicheranwendungen / vorgelegt von Ferdinand Paul Schmidt

De
184 pages
Optimizing Adsorbents for Heat Storage Applications:Estimation of Thermodynamic Limits and Monte Carlo Simulationsof Water Adsorption in NanoporesOptimierung von Adsorbentien fur¨ Warmespeicheranwendungen:¨ Abschatzung¨ der thermodynamischenGrenzen und Monte Carlo Simulation der Wasseradsorption in NanoporenInaugural DissertationzurErlangung des DoktorgradesderFakultat¨ fur¨ Mathematik und PhysikderAlbert Ludwigs Universit at¨Freiburg im Breisgauvorgelegt vonFerdinand Paul Schmidtaus Verden / AllerJuli 2004Dekan: Prof. Dr. R. SchneiderLeiter der Arbeit: Prof. Dr. J. LutherReferent: Prof. Dr. J.Korreferent: Prof. Dr. H. HaberlandTag der Verkundigung¨des Prufungser¨ gebnisses: 1. September 2004PUBLICATIONS iiiIn the context of this thesis, the following articles have been published:Peer reviewed Journal Schmidt, F. P., J. Luther and E. D. Glandt: 2003.Influence of Adsorbent Characteristics on the Performance of an Adsorption Heat Storage Cycle.Industrial & Engineering Chemistry Research 42, 4910–4918Conference contributions and other publications Schmidt, F. P., S. K. Henninger , T. Nu´nez,˜ H. M. Henning and E. D. Glandt, 2004.Adsorbent Optimization for Heat Storage: Estimation of Achievable Energy Densities from Analyti cal Thermodynamic Model. Poster presentation at Fundamentals of Adsorption (FOA 8) conference,Sedona, Arizona, May 23–28, 2004, paper # 372 Henninger, S. K., F. P. Schmidt, T. Nu´nez˜ and H. M. Henning, 2004.
Voir plus Voir moins

Optimizing Adsorbents for Heat Storage Applications:
Estimation of Thermodynamic Limits and Monte Carlo Simulations
of Water Adsorption in Nanopores
Optimierung von Adsorbentien fur¨ Warmespeicheranwendungen:¨ Abschatzung¨ der thermodynamischen
Grenzen und Monte Carlo Simulation der Wasseradsorption in Nanoporen
Inaugural Dissertation
zur
Erlangung des Doktorgrades
der
Fakultat¨ fur¨ Mathematik und Physik
der
Albert Ludwigs Universit at¨
Freiburg im Breisgau
vorgelegt von
Ferdinand Paul Schmidt
aus Verden / Aller
Juli 2004Dekan: Prof. Dr. R. Schneider
Leiter der Arbeit: Prof. Dr. J. Luther
Referent: Prof. Dr. J.
Korreferent: Prof. Dr. H. Haberland
Tag der Verkundigung¨
des Prufungser¨ gebnisses: 1. September 2004PUBLICATIONS iii
In the context of this thesis, the following articles have been published:
Peer reviewed Journal
Schmidt, F. P., J. Luther and E. D. Glandt: 2003.
Influence of Adsorbent Characteristics on the Performance of an Adsorption Heat Storage Cycle.
Industrial & Engineering Chemistry Research 42, 4910–4918
Conference contributions and other publications
Schmidt, F. P., S. K. Henninger , T. Nu´nez,˜ H. M. Henning and E. D. Glandt, 2004.
Adsorbent Optimization for Heat Storage: Estimation of Achievable Energy Densities from Analyti
cal Thermodynamic Model. Poster presentation at Fundamentals of Adsorption (FOA 8) conference,
Sedona, Arizona, May 23–28, 2004, paper # 372
Henninger, S. K., F. P. Schmidt, T. Nu´nez˜ and H. M. Henning, 2004.
Monte Carlo Investigation of the water adsorption behavior in MFI Type Zeolites for different Si/Al
ratios with regards to heat transformation applications. Poster presentation at Fundamentals of Ad
sorption (FOA 8) conference, Sedona, Arizona, May 23–28, 2004, paper # 363
Schmidt, F. P. and J. Luther, 2002.
Monte Carlo Simulation of Water Adsorption in Micropores
Freiburg Materials Research Center: Annual Report 2002, p. 67
Henning, H. M., F. P. Schmidt, S. Henninger and T. Nu´nez,˜ 2001.
Materials Research for Adsorption Heat Storage: Application of Molecular Computer Simulation.
Fraunhofer ISE: Achievements and Results – Annual Report 2001, p. 32Contents
Publications iii
1 Introduction 3
2 Thermodynamics of adsorption heat storage 9
2.1 Heat storage: State of the Art . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.1 Hot water storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.2 Phase Change Materials (PCMs) . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.1.3 Adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.1.4 System concepts for adsorption heat storage . . . . . . . . . . . . . . . . . . . . . 13
2.2 General adsorption thermodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2.1 Classification of adsorption isotherms . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.2 Henry’s constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 Adsorption heat storage cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 The “Dubinin approach” to materials optimization . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Statistical thermodynamics approach to adsorbent optimization . . . . . . . . . . . . . . . 26
2.5.1 Langmuir model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.5.2 Reference conditions for cycle modeling . . . . . . . . . . . . . . . . . . . . . . 31
2.5.3 Parametric study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5.4 Modeling energetic heterogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.5.5 Lattice gas models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.5.6 Variation of temperature lift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3 Experimental results and comparison with model predictions 53
3.1 Thermogravimetric measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2 Henry’s constants of water on silica gels . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.3 Comparison with Langmuir and lattice gas models . . . . . . . . . . . . . . . . . . . . . 59
3.4 Volumetric energy densities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.5 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4 Monte Carlo Simulations: Background and models employed 71
4.1 Monte Carlo sampling techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.1.1 Random sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.1.2 Importance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.1.3 The Metropolis algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.1.4 The condition of detailed balance . . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.1.5 Simulations in the grand canonical ensemble . . . . . . . . . . . . . . . . . . . . 78
4.1.6 Simulation of non spherical molecules . . . . . . . . . . . . . . . . . . . . . . . . 79
4.2 Molecular modeling of water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.2.1 Partial charge models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.2.2 Ewald summation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.3 Tetrahedral square well (TSW) water model . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.4 How to increase Sampling Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 862 CONTENTS
4.4.1 Configurational bias algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
4.4.2 Tests of particle exchange efficiency . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.5 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
24.6 Force fields and zeolite modeling in Cerius /Sorption . . . . . . . . . . . . . . . . . . . . 91
4.6.1 Generation of zeolite host grids . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.6.2 Force fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5 Simulation results and comparison with experiment 97
25.1 Simulations of water in zeolites with Cerius . . . . . . . . . . . . . . . . . . . . . . . . 98
5.1.1 Water adsorption in zeolite NaX . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
5.1.2 Water in ZSM 5 . . . . . . . . . . . . . . . . . . . . . . . . . 100
5.2 Properties of the TSW water model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.2.1 Virial coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.2.2 Radial distribution function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
5.3 From activated carbon to silica surface models . . . . . . . . . . . . . . . . . . . . . . . . 105
5.3.1 Activated carbon model used previously with TSW water . . . . . . . . . . . . . . 106
5.3.2 Previous results on adsorption properties of activated carbon model . . . . . . . . 107
5.3.3 Adaptation of slit pore model to silica gel surface structure . . . . . . . . . . . . . 109
5.3.4 Cristobalite model surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.3.5 Conclusions for modeling silica surfaces with respect to TSW water model . . . . 115
5.4 Slit pore model with amorphous silica surface . . . . . . . . . . . . . . . . . . . . . . . . 116
5.5 Adsorption properties of slit pore model . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.5.1 Reference pore and scheme of parametric study . . . . . . . . . . . . . . . . . . . 120
5.5.2 Variation of silanol surface density . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.5.3 Finite size effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.5.4 Variation of pore width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.6 Conclusions for heat storage and suggested model refinements . . . . . . . . . . . . . . . 130
6 Summary and Outlook 137
6.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.2 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
A Adsorbents 147
A.1 Zeolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
A.1.1 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
A.1.2 Water adsorption properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
A.2 Silica Gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
A.2.1 Pore structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
A.2.2 Use of silica gels for adsorptive heat transformation . . . . . . . . . . . . . . . . . 150
A.2.3 SWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
A.3 Others . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
A.3.1 MCM type adsorbents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
A.3.2 Aluminophosphates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
A.3.3 Activated Alumina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
A.3.4 Modified carbonaceous adsorbents . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Nomenclature 155
Bibliography 161
Acknowledgements 179Chapter 1
Introduction
Motivation
stThe necessary transition to sustainable energy systems in the 21 century poses tremendous scientific, tech
nological and political challenges. Various scenarios for future energy systems and a range of “transition
1paths” towards sustainability are being discussed. With regards to short and medium term predictions a
common denominator of many of these scenarios is that a more decentralized and flexible energy infrastruc
ture will lead to a significantly improved overall energy efficiency, especially through an increased fraction
of power generated in combined heat and power (CHP) plants. In the near future fuel cells will provide
an attractive option (among others) for realizing such cogeneration plants. With a decentralized infrastruc
ture and new modes of controlling and operating distributed power plants it would also be easier to feed
an increasing fraction of power from renewable sources into the electricity grid. Today’s highly central
ized infrastructure would eventually face problems in coping with the fluctuating supply of wind and solar
power.
The time averaged, overall efficiency of combined heat and power systems will only be high if the
generated heat can serve useful purposes throughout the year. In many climate zones, namely where heat
demand is very low in summer but cooling and air conditioning systems consume a large share of generated
power during the warmest months of the year, heat driven cooling are a very attractive option. They
both reduce peak power demand and allow the use of heat from cogeneration plants that would otherwise
be wasted under these conditions. Cooling and air conditioning systems based on the adsorption of vapors
on porous solids can play an important role in this context (see Henning, 2004, for an overview on the state
of these technologies).
Thermodynamically speaking, the supply of “cooling power” means to provide heat sinks at suitable
temperature levels for the respective applications. Including cooling and additional energy services in the
concepts of distributed power generation and cogeneration leads to the concept of “polygeneration”, which
the European Commission (2002) defines as follows:
Polygeneration encompasses the combined production of electricity, heat, cold and products
(hydrogen or other fuels or chemicals), district heating or cooling systems or other advanced
2energy services.
Recognizing the efficiency potentials of these novel concepts the European Commission has designated
th“polygeneration” as one of the targeted research areas in its 6 Framework Program (FP6) for Research,
Technological Development and Demonstration.
1Such a debate is necessarily not a purely scientific and impartial one, but also one in which stakeholders participate whose interests
are affected by the outcomes. Only a few key sources for tracing this debate are given here: The “World Energy Outlook” series of the
International Energy Agency (IEA, 2002); the “Assessment Reports” (especially Metz et al., 2001) of the Intergovernmental Panel on
Climate Change (IPCC), an international scientific body established by the World Meteorological Organization (WMO) and the United
Nations Environment Program (UNEP); the work of the enquete commission on sustainable energy supply of the german Bundestag,
especially the commissioned report by Schlesinger et al. (2002); the long term solar energy scenario for Germany by Langniss, Luther,
Nitsch, and Wiemken (1997).
2The EU funded research projects already underway within this framework are linked into a research cluster “Integration of Re
newable Energy Systems and Distributed Generation” (http://www.clusterintegration.org). The largest research activity within this
cluster is called “DISPOWER” (http://www.dispower.org).4 CHAPTER 1. INTRODUCTION
As a consequence of the success of the “distributed generation paradigm” (Borbely and Kreider, 2001),
there is also a growing interest in efficient, high energy density heat storage systems. In a distributed
generation system the need for heat storage arises from the separate demand profiles for heat and power: If
cogeneration plants are operated in a power controlled mode in order to match power supply and demand
in a grid with a large fraction of renewable energy sources, then the co generated heat needs to be stored
until there is a demand for its use.
Another field in which heat storage plays a key role is solar room heating. Under climatic conditions
of central Europe – and even more so at higher geographic latitudes – most of the annual solar insolation
occurs during summer, while demand for room heating peaks in the winter months. Therefore, long term
heat storage becomes necessary if more than about 20% of the heat demand is to be met by solar energy
(Lindenberger et al., 2000).
Adsorption heat storage
The most important criteria for the choice of a heat storage technology (besides cost) are the storage density,
defined as energy stored per unit volume, and the efficiency, defined as the fraction of energy recov
erable after the storage period. Adsorption heat storage offers the technological advantage that its storage
efficiency does not depend (in principle) on storage time: The adsorption of a gas on the surface of a solid is
an exothermal process that here provides the heat released or “discharged” from the storage. Preventing this
discharge simply requires to keep the dry solid adsorbent and the adsorbing fluid, called adsorptive, in sepa
rate closed vessels. In order to maximize the storage density, water is the most suitableve due to its
m 3exceptionally high heat of condensation per unit volume of the bulk liquid (h % = 676 kWh/mcond liq:
at room temperature, T = 298 K). A further advantage of using water as the working fluid is that this
choice leads to ecologically benign, non hazardous systems which are well suited for use in a residential
environment – which would not be the case for some other adsorptives such as ammonia.
The adsorbents to be considered should have a very large internal surface area and should release a large
amount of heat upon adsorption of the fluid. This leads to microporous, hydrophilic adsorbents as the most
suitable candidates, and the most prominent among such materials are zeolites and silica gels.
Situation of research field
Research needs related to adsorption heat storage strongly overlap with those for the related applications
of adsorptive heat pumping and cooling. These three applications are often collectively referred to as “heat
3transformation applications” of adsorption (see e.g. Nu´nez˜ , 2001). Consequently, research on all of these
applications is to a large extent carried out by the same working groups. There is also some overlap with
resarch into absorption (gas/liquid) systems for heat transformation. A review of the state of the art in ther-
mally driven heat pumping and cooling technology, to which several European research groups contributed,
was presented by Pons et al. (1999). An updated review of the heat pumping field was given by Wongsuwan
et al. (2001).
As far as adsorption systems are concerned, research activities are generally focussed on two different
aspects: Firstly, finding adsorbents (or “working pairs” of adsorptive/adsorbent) with improved equilibrium
properties for a given adsorption cycle and secondly, improving the dynamical behavior of adsorption reac
tors or systems. This thesis is only concerned with the first of these aspects, adsorption equilibria. For long
term heat storage this aspect is the most important one since it determines the achievable energy density.
For heat pumps or cooling machines with short cycle times dynamical properties like the power density or
the coefficient of performance (COP) become more important. Improvements of these dynamical properties
can be achieved e.g. by improving the heat and mass transfer in adsorbent layers, or by combining several
adsorption vessels in a way allowing recovery of a large part of the sensible heat involved in heating and
cooling the different adsorption reactors. The problems to be solved along this way mostly lie on a process
and systems engineering level.
Regarding adsorption equilibria, an important research task that was largely accomplished by the end of
2001 was a systematic screening of available adsorbents and evaluation of their performance e.g. in a heat
3In a narrow sense, “heat transformation” refers to the splitting of a heat flow at an intermediate temperature into two flows at
higher and lower temperature (Rothmeyer, 1985).5
4storage cycle. Our working group at the Fraunhofer Institute for Solar Energy Systems (ISE) was strongly
involved in this effort of characterizing materials (Nu´nez˜ et al., 1999; Nu´nez˜ , 2001). Methods adapted from
the phenomenological “theory of volume filling in micropores” (Dubinin, 1975) have played a key role in
adsorbent evaluation. They can be considered the state of the art in predicting adsorption cycle performance
in a specific application from a limited set of adsorption equilibrium data.
Parallel to the screening of materials, research and development work has been conducted on demonstra
tion systems for adsorption heat storage. A seasonal solar heat storage system was developed in a research
project entitled “High Energy Density Sorption Heat Storage for Solar Space Heating” (HYDES) in which
our group participated (Mittelbach and Henning, 1997; Mittelbach et al., 2000). Despite the generally
promising features of adsorption heat storage systems, the energy densities realized in the demonstration
units built within the HYDES project were somewhat disappointing. It turned out that technical adsorbents
presently available on the market do not lead to the desired increased energy density compared to conven
tional hot water storage. Only a significantly increased energy density would justify the additional effort
and expense related to technical equipment and control in an adsorption system. Most technical adsorbents
available today have been optimized for applications very different from heat transformation, primarily for
gas separation and catalytic processes in the chemical industry. It was concluded from the HYDES ex
perience that the lack of availability of microporous adsorbents optimized specifically for this application
produces a bottleneck that further research should focus on (Nu´nez,˜ Henning, and Mittelbach, 2003).
The adsorbent used in the HYDES demonstration system was a microporous silica gel, and the energy
density measured for this system was consistent with the predictions based on water adsorption data ob
tained in our laboratory on a small sample of the same material. Since evaluation of water adsorption data
of other silica gels led to a prediction of a similar (and disappointing) storage density achievable with these
materials, the hypothesis was formulated that the interaction of water with the adsorption sites in silica gels
was simply not in the right energy range to enable a high storage density. A task to be addressed here is to
test this hypothesis by obtaining and interpreting adsorption data of water vapor on silica gel at very low
pressures. The result of this test would determine which class of adsorbents should be selected as the most
suitable candidate for the effort of “tailoring” an optimized adsorbent for heat storage.
Need for interdisciplinary approach
While heat transformation applications of adsorption have until now been a research domain of engineers
and physicists it is clear that the development of new, tailored adsorbents would be a task to be addressed
by chemists. This need for interdisciplinary co operation creates a need for the “heat transformation com
munity” to embrace new methods and concepts that would allow to link the desired adsorption properties
to desired characteristics in the adsorbent microstructure. The latter would provide a starting point for
chemists, who could then attempt to gear their synthesis procedures towards producing these desired struc
tural features.
Thus an optimization of adsorbents for heat storage or heat transformation in general can only be at
tempted on the basis of a solid understanding of “structure property relationships”. The analysis of such
relationships is one of the core tasks in many branches of materials science, and in recent years, computer
simulations are being increasingly employed to aid this analysis. With regards to adsorbents, molecular
5simulations have been successfully used for this purpose (see e.g. Davies and Seaton, 2000). Research
on molecular simulations of adsorption has mostly been carried out by theoretically oriented chemical en
6gineers. This can be understood from the high relevance of specially tailored adsorbents in the chemical
industry (see Barton et al., 1999 for a review).
7I was able to establish a link between our group at Fraunhofer ISE and the chemical engineering com
4This group at Fraunhofer ISE is headed by Dr. Hans Martin Henning, who also co ordinates an international research collaboration
on “Solar Assisted Air Conditioning of Buildings” within the International Energy Agency’s Solar Heating and Cooling Programme
(SHC Programme Task 25).
5The publication by Davies and Seaton (2000) deals with the separation of ethane and methane mixtures by adsorption in microp
orous carbons. The paper is presented by the authors as a case study for how molecular simulations can be used as a tool for selecting
the best adsorbent for a given application.
6 thIn 2002, the Journal “Molecular Physics” devoted a special issue to celebrate the 65 birthday of a chemical engineer, Keith E.
Gubbins. In the foreword to this special issue, Cummings, Jackson, and Rowlinson (2002) trace the influence of Gubbins’ work on
statistical and molecular physics.
7I joined the research group of H. M. Henning in November 1997 for a study project (“Hauptpraktikum”, Schmidt, 1998), and
stayed on for my diploma thesis (Schmidt, 1999).6 CHAPTER 1. INTRODUCTION
8munity when I met Eduardo Glandt of the University of Pennsylvania at the “Fundamentals of Adsorp
tion 6” conference in 1999, while still working towards my diploma degree. Upon invitation by E. Glandt I
have spent a total of nine months (in 2000 and 2001/02) working with him in his research group at Penn.
Given the experiences with the HYDES project, it was unclear at the outset of this work how much could
be gained by efforts in targeted adsorbent design. Molecular simulation of the relevant water/adsorbent
systems and even more so the actual synthesis of modified adsorbents are tasks that require large scale,
co ordinated efforts to be resolved. To determine whether the results can possibly be worth this effort, it
is therefore of high interest to estimate the thermodynamic limits to adsorbent optimization. Specifically,
in the framework of this thesis we are interested in the maximum energy density that could be achieved
in an adsorption heat storage cycle under certain reference conditions, using water and an ideal adsorbent.
The reference conditions considered here are chosen to be representative for a long term heat storage in a
solar thermal system. However, the methods of analysis employed in this thesis may as well be applied to
adsorptive heat pumping and cooling systems.
Outline of this thesis
This thesis can be divided into two parts. The first part (chapters 2 and 3) is concerned with answering
the question whether the effort of adsorbent optimization for heat storage is worthwhile to be undertaken.
Provided that the answer to that question is positive, the main task in the second part of this thesis is to
prepare the ground for a targeted adsorbent design (chapters 4 and 5). This involves a review of molecular
simulation methods that can be employed to analyze structure property relationships of the systems of our
interest. A simulation program that enables the prediction of water adsorption equilibria in microporous
solids must be found or developed in the course of this work. The suitability of this program is to be
demonstrated by showing that it enables the calculation of equilibria under the conditions of interest for
heat storage applications. Specifically, this includes the challenge of reaching thermodynamic equilibrium
for a relatively large, disordered adsorbent system at a high density of adsorbed water within an acceptable
amount of computing time. Once this task is mastered, the validity of the molecular models and force
fields employed in the simulation can be assessed trough comparison with experimental data. Within the
selected model system, a first analysis of structure property relationships can be undertaken by performing a
parametric study in which the adsorbent structure is systematically varied and the effect on water adsorption
equilibria is recorded and analyzed.
More specifically, the flow of the four main chapters is as follows: In chapter 2, the foundations for
phrasing the problem in thermodynamic terms are laid. After a brief review of the state of the art in low
temperature heat storage technology including the HYDES project, the thermodynamic relationships rele
vant to such adsorption cycles are summarized. The method of adsorbent characterization and evaluation
that is today widely employed in the “heat transformation community” is discussed. It is then made clear
why this method, based on Dubinin’s “theory of volume filling in micropores” (Dubinin, 1975), is too nar-
row in scope with regards to the task of adsorbent optimization. In an alternative approach, the relationship
between adsorbent properties and performance in a heat storage cycle is discussed within two statistical
mechanical models of adsorption, the Langmuir model and the Bragg Williams model. A parametric study
of the cycle peformance of model adsorbents is presented, and it is shown that these simple model systems
are suitable for estimating the thermodynamic limits to adsorbent optimization. This part of the chapter
is largely based on results that have previously been published in a journal article (Schmidt, Luther, and
Glandt, 2003).
Chapter 3 presents experimental work in which adsorbent properties are related to the model systems.
The temperature dependence of Henry’s constant for water adsorption in a microporous silica gel is de
termined through thermogravimetric measurements in combination with a dosing of water vapor. These
results are related to the statistical adsorption models by analyzing model systems in which the tempera
ture dependence of Henry’s constant agrees with the experimental data. This procedure allows to test the
hypothesis that the interaction of water with silica gel surfaces would be unsuitable to achieve a high en
ergy density in a heat storage system using a silica gel based adsorbent. In a separate comparison between
experimental data of various adsorbents and model results, the volumetric energy density achievable in an
8Eduardo D. Glandt is Dean of the School of Engineering and Applied Science (SEAS) at the University of Pennsylvania (Philadel
phia, PA) and Professor of Chemical Engineering. The focus of his work has been on the statistical physics of fluids in disordered
media (see e.g. Madden and Glandt, 1988) and on molecular simulations of gas adsorption (Matranga, Myers, and Glandt, 1992;
Gordon and Glandt, 1997).

Un pour Un
Permettre à tous d'accéder à la lecture
Pour chaque accès à la bibliothèque, YouScribe donne un accès à une personne dans le besoin