a novel solid oxide fuel cell tri-generation -

12th

International Conference on Sustainable Energy Technologies (SET-2013)

26 - 29th August, 2013

Hong Kong

Paper ID: SET2013 – 251

1

A Novel Solid Oxide Fuel Cell Tri-Generation System for Low Carbon Buildings

Theo Elmer1*

, Mark Worall2, Saffa B Riffat

3

1, 2, 3 Institute of Sustainable Energy Technology, The University of Nottingham, NG7 2RD

* Corresponding author email: [email protected]

ABSTRACT As a result of growing concerns of climate change and ever increasing cost and scarcity of fuel resources, fuel cells with their high

electrical efficiency, low emissions and useful heat output have been identified as a key technological option for improving both

building energy efficiency and reducing emissions. A 1.5kWe low temperature solid oxide fuel cell (SOFC) liquid desiccant tri-

generation system has been proposed as an environmentally friendly way of providing heating, cooling and electricity to low

carbon buildings. The described system will be constructed and trialled at The University of Nottingham as part of a European

Union (EU) funded project named Durable Solid Oxide Fuel Cell Tri-generation System for Low Carbon Buildings (TriSOFC).

The aim of this paper is to provide an introduction to the EU TriSOFC project. The paper is split into three parts. The first part,

section two, provides details of the TriSOFC system along with project aims and objectives. The second part, section three,

presents a review of the state-of-the art surrounding fuel cell and desiccant based tri-generation systems. Conclusions of the

review are discussed in terms of their impact on the TriSOFC system, most significantly (1) a considerable gap in the literature

currently exists surrounding tri-generation energy systems based on SOFC combined with liquid desiccant air conditioning

technology and fuel cell based tri-generation systems in residential applications (2) the thermal output of a fuel cell combined heat

and power (CHP) system is in good thermal agreement with desiccant regeneration temperatures, making them appropriate

companion technologies for integration in a tri-generation system concept, and (3) maximising thermal energy utilisation in CHP

and tri-generation systems is challenging, and can have a large impact on system efficiency. The third part of the paper, section

four, presents the parametric modelling results for a liquid desiccant air conditioner. Parametric studies have investigated the

influence that, cooling and heating water temperature, fluid flow rates, inlet air conditions and desiccant solutions (LiCl, CaCl2

and CHKO2), have on performance. There are four conclusions from the modelling work. First, the moisture removal rate of the

supply air is strongly controlled - assuming constant mass flow rate, by desiccant temperature and cooling water temperature.

Second, the air inlet conditions have a large influence on performance in terms of cooling output. The unit will generally perform

better in a hot and more humid climate such as southern China, as opposed to drier cooler climates such as the UK. Third, as the

regeneration heat source temperature increases, the mass of water vapour vaporised from the weak desiccant solution increases.

Fourth, in this model the CHKO2 desiccant solution shows the highest performance for the given conditions, and with its low

environmental impact, is a suitable solution for the desiccant unit in the TriSOFC system.

A separate paper presented at this conference (Paper ID: 334) reports the development of computer simulations carried out to

evaluate the novel desiccant membrane contactor design that will be used in TriSOFC and size the desiccant system.

KEYWORDS: Fuel cell, tri-generation, liquid desiccant air-conditioning, low carbon buildings

1. INTRODUCTION The aim of this paper is to provide an introduction to the EU project – Durable Solid Oxide Fuel Cell Tri-generation System for

Low Carbon Buildings (TriSOFC). The paper is split into three sections. The paper is split into three parts. The first part, section

two, provides details of the TriSOFC system along with the aims and objectives. The second part, section three, presents a review

of the state-of-the art surrounding fuel cell and desiccant based tri-generation systems. The third part, section four, presents the

parametric modelling results for the liquid desiccant air conditioner that will be used within the tri-generation system.

The central aim of the EU project is to design, develop and test a proof-of-concept tri-generation system based on solid oxide fuel

cell (SOFC) and liquid desiccant air conditioning technology, to provide heating, cooling and electricity to a low carbon building

application. Currently there is growing demand for energy in buildings and an increasing concern regarding the detrimental

environmental impact emissions of pollutants such as CO2 have. The TriSOFC system is a strong candidate for providing an

energy efficient, low carbon alternative to traditional building energy supply. The proposed TriSOFC system shown in Figure 1 is

truly multi-functional, designed to provide the energy needs of a building throughout all seasons. The SOFC is the core of the

system, generating electricity through electrochemical reactions using hydrogen from the natural gas stream. The only by-products

are water vapour, heat and a modest amount of CO2 – using a fuel cell is a more efficient process than simply combusting the fuel

[1]. The desiccant air conditioning cycle is predominately driven by heat for regeneration; therefore it is an attractive option for

integration with a SOFC. The desiccant air conditioning unit will be used to produce dehumidified cool air in summer, and

dehumidified warm air in winter. Furthermore, the utilisation of waste heat from the fuel cell can provide a means of effective

thermal management to the stack, ensuring both increased performance and longevity of the entire system. The use of desiccant as

a means of thermal storage has also been proposed and will feature in the final system. The fuel cell technology will be based on

solid oxide. The liquid desiccant dehumidifier will be based on a cellulose fibre membrane to hold the desiccant solution and

prevent carry-over. The desiccant dehumidifier will be incorporated alongside an indirect evaporative dew point cooler to provide

further sensible cooling to supply air.

mailto:[email protected]


Theo Elmer, Mark Worall, Saffa Riffat

2

Figure 1 TriSOFC system concept [2]

Within the EU project, the authors are responsible for the desiccant unit development. Therefore modelling work to date has

focussed on this. The initial modelling work uses a relatively simple method proposed by Gandhidasan [3] based on dimensionless

vapour pressure and temperature difference ratios for the preliminary design of a liquid desiccant dehumidifier and regenerator

system. The modelling work has been validated by well-regarded experimental data presented in the literature by Fumo and

Goswami [4]. Parametric analyses have been conducted and investigate the influence cooling water temperature, fluid flow rates,

inlet air conditions and desiccant solutions – LiCl, CaCl2 and CHKO2 have on the performance of the dehumidifier; namely

moisture removal rates and cooling output. The regenerator has also been studied to identify the influence regeneration

temperature has on performance. The results have been commented upon and their influence on system design and optimisation

discussed. As previously stated, the proposed desiccant air conditioning system used in TriSOFC will be based on a novel

cellulose fibre membrane contactor design. A separate paper presented at this conference (Paper ID: SET2013 – 334) reports the

development of computer simulations carried out to evaluate the novel membrane contactor (heat and mass exchange

effectiveness) and size the system.

2. THE TriSOFC PROJECT

TriSOFC is a consortium based project, involving eight collaborators from across Europe. The overall aim is to define, develop

and deliver a novel low-cost durable low temperature SOFC tri-generation prototype. The system will be based on the

breakthroughs in low temperature SOFC and liquid desiccant air conditioning technology already made by the consortium

members.

The fuel cell innovation involves recent advances in low temperature SOFC technology using ceria-carbonate two or multi-phase

nanocomposite materials. These materials have illustrated that high electrochemical performance of up to 1.2 W/cm2 can be

achieved at reduced temperatures of 500°C, with excellent tolerance to sulphur poisoning and carbon deposition. These

developments are a clear advantage for the domestic sector where high temperature operation (750-1000°C) results in technical

complexity and consequent costs that inhibit commercialisation. Furthermore these developments mean SOFC stacks can be

manufactured for prices below 400 €/kWe compared to 1000 €/kWe for conventional SOFC systems [2]. The use of these new

materials also makes the prospect of using single cell construction technology a possibility, which would mean a simpler

fabrication procedure and further reductions to cost. The innovations surrounding the liquid desiccant air conditioning unit is the

development of a novel cellulose fibre membrane contactor design. The membrane can effectively prevent desiccant carry-over

into the supply air stream, a particular issue in previous contactor designs [5, 6]. Furthermore, the membrane design provides

greater contacting surface area between the air and desiccant solution, improving the efficiency of the air conditioning process.

There are several technical aims for TriSOFC; these have been summarised in Table 1.

Table 1 summary of technical aims of the tri-generation system [2]

Technical Parameter Goal

Fuel cell electrical power 1.5 kW

System efficiency (based on natural gas) > 90%

Electrical efficiency > 45%

It has been predicted that the described TriSOFC system running on hydrogen from the natural gas network will result in a 70%

CO2 emission reduction compared to a traditional energy production system comprising of separate condensate power plant, boiler

and compressor driven cooling unit. The TriSOFC system will be installed and tested in the Creative Energy Homes, at The

University of Nottingham. This work will provide a real world application study for fuel cell tri-generation system operating in a

stationary low carbon building environment. Such a study is a great opportunity as the majority of previous works related to the

topic have either been theoretical or laboratory based. However, as highlighted in the review of the literature presented in Section

Two, effective system integration in a real world application will be a challenging aspect of the project. A project website

documenting progress of the EU project has been set up at www.trisofc.com

http://www.trisofc.com/



3

3. TRI-GENERATION SYSTEMS Tri-generation can be deemed a subset of a term known as Combined Heat and Power (CHP). CHP is defined as the generation of

heat and power from a single fuel source, with a view to use both products. Tri-generation systems take this concept one step

further by producing a cooling output. By consuming the often wasted heat produced in the electrical generation process system

efficiency can be elevated from as low as 20% (depending on the electrical generator) to nearly 90% in tri-generation applications.

This results in reduced emissions and running costs. Because a fuel cell produces heat whilst generating electricity they are

particularly well suited to tri-generation applications.

Literature searches have shown that a fuel cell tri-generation system utilising liquid desiccant air conditioning technology will be

the first of its kind: no research activity is reported on fuel cell and liquid desiccant air conditioning. However, there are a

small number of research publications and patents that focus on SOFC tri-generation systems [7-11]. These works either focus on

the use of the fuel cells thermal output in a Rankin bottoming cycle or the use of vapour absorption cooling. Although little/no

work has been found directly relating to the proposed concept, the rationale and initial thinking behind the success of the proposed

system is that liquid desiccant regeneration temperatures are lower than that of solid desiccant media [3], therefore a well suited

technological partnership for a fuel cell CHP system, with a low temperature output of 50-80°C [12].

Because no research publications have been found detailing fuel cell-liquid desiccant systems, the review of the current literature

has been split into two key areas:

(1) Fuel cell tri-generation systems

(2) Liquid desiccant based tri-generation systems

It is hoped this approach will highlight important findings, such as: system design, performance, operational issues and evaluation

techniques. These lessons can then be applied to the TriSOFC system.

3.1 FUEL CELL TRI-GENERATION SYSTEMS

This section will examine tri-generation systems employing fuel cell technology. Fuel cells are well suited to tri-generation

applications because they produce heat when generating electricity, have low emissions, high efficiency and excellent load

following characteristics whilst maintaining high electrical efficiency [7, 13-15]. Furthermore, continued technological

improvements to fuel cells (durability, reliability) have facilitated an increased interest in fuel cell tri-generation systems [15].

Yu, Han et al. [15] have numerically investigated a tri-generation system incorporating a SOFC and a double-effect water/lithium

bromide absorption chiller. High system efficiencies of 84% or more were reported by the authors, illustrating the benefits of tri-

generation systems in applications where heating, cooling and power is required. Margalef and Samuelsen [16] numerically

examined a 300kW molten carbonate fuel cell tri-generation system using absorption cooling. The system achieved an overall

efficiency of 72%. However, the pairing of two ‘off the shelf’ technologies for tri-generation system construction can be

problematic. As Margalef [16] states, the molten carbonate fuel cell and absorption chiller chosen for the tri-generation system

were close, but not an ideal match. The fuel cell exhaust gas temperature was higher than the inlet temperature specified for the

chiller and the exhaust gas flow rate was not sufficient to achieve the required heat recovery within the chiller heat exchanger.

Therefore two strategies to overcome this were presented (1) blending the fuel cell exhaust gases with ambient air, and (2) mixing

the fuel cell exhaust gases with a fraction of the chiller exhaust gas. Both options worked, however the second option yielded

better performance and was thus used.

Al-Sulaiman, Dincer et al. [7] presents an energy analysis of a tri-generation plant incorporating a 520kW SOFC, organic Rankin

cycle, heat exchanger and single effect absorption chiller. The investigation shows that incorporating the cooling cycle system

efficiency is improved by 22% compared to just having the SOFC and organic Rankin cycle running together. A maximum tri-

generation efficiency of 74% has been achieved.

The above investigations have however been for larger commercial applications. Limited work has been carried out on residential

scale tri-generation systems due to system size, cost and complexity. The work that has been completed has mainly focussed on

simulation studies. The most appropriate are summarised below.

Bhatti, O'Brien et al. [8] present a patent publication for a SOFC assisted air conditioning system based on solid desiccant and

evaporative air conditioning technology. The aim of the system is to provide both comfort cooling and heating. The authors state

that the most common technology to provide cooling/heating is a reversible heat pump; however these use electric power to drive

the compressor. Desiccant enhanced systems are a low energy alternative to this. The thermal energy from the SOFC is used to

regenerate a solid desiccant wheel, which addresses the buildings latent load, whilst an evaporative cooler is used to address the

sensible load. No performance data is available, however it is expected the system will provide high co-generation efficiency and a

measurable energy reduction in comparison to a separate vapour compression system, boiler and grid electricity system.

Braun, Klein et al. [17] have developed a system model to evaluate the performance of different residential scale SOFC CHP

systems. The authors explain that currently a considerable gap exists in the area of techniques to maximise the benefits of fuel cell

systems, both in terms of electrical generation and thermal utilisation. The authors state that the majority of previous work has



4

been focussed on large applications, not residential. Braun, Klein et al. looked at a SOFC CHP system with an electrical power

output of around 1.5kW, a thermal output of 60°C, and a heat to power (H:P) ratio of 0.7 to 1.0. Five system configurations were

studied: (1) hydrogen fuelled (2) methane fuelled with external reforming (ER), and internal reforming (IR) (3) methane fuelled

with cathode gas recycle (CGR) (4) methane fuelled with anode gas recycle (AGR), and (5) integration of CGR, AGR and IR. The

results are based on thermodynamic models which simulate the efficiency performance of the fuel cell co-generator. It was found

that optimisation of each system is dependent upon the cell power density and the required H:P of the application. A summary of

the results are as follows: (1) the hydrogen fuelled SOFC did not offer any efficiency advantages over methane for both IR and

ER, in some cases the magnitude of efficiency advantage of methane can be as high as 6% (2) using IR in the SOFC reduces the

system air flow by 50%, thus producing a higher electrical efficiency, but the CHP efficiency remains low (3) CGR enhances

performance (system efficiency), by eliminating the system air input and pre-heater requirement (4) AGR and CGR with IR

produces the highest CHP efficiency and a H:P ratio near one, whilst minimising the air-pre-heater and air blower requirement.

These results show potential optimisation considerations for the TriSOFC system and that the proposed system should not suffer

any efficiency losses due to its operation on natural gas.

Pilatowsky, Romero et al. [18] have carried out simulation based work on a 1kWe Proton Exchange Membrane Fuel Cell

(PEMFC) coupled to an absorption cooling system. The simulations were completed to determine the optimum operating

conditions of the air conditioning system during the co-generation process; primarily the cooling capacity at maximum power

output from the PEMFC. The absorption cycle was operated with a monomethylaminre-water solution, with a low vapour

generation temperatures of around 80°C, ideal for PEMFC CHP applications. Results show that the co-generation process

increases total efficiency of the system, illustrating the feasibility of using fuel cells in small scale combined cooling applications.

Results from the simulation include (1) the Coefficient of Performance (CoP) of the absorption cooler increases as the generation

temperature (from the fuel cell) increases, reaching a maximum value and then dropping off (2) the cooling power increase with

an increase of the electrical power from the PEMFC and evaporation temperature of the absorption cooler, but decreases with an

increase of generation temperature, and (3) the co-generation efficiency is almost independent of the electrical power, but strongly

dependant on evaporation and generation temperatures of the absorption cooler.

With recent advances in fuel cell CHP systems for residential applications, the feasibility of fuel cell tri-generation systems in

domestic homes is strong. Fuel cell tri-generation systems can create substantial energy savings leading to reduced atmospheric

pollution and operational costs. However an issue such as accurate pairing of technologies still needs careful consideration.

Currently a large gap in the literature exists regarding SOFC tri-generation systems for residential applications. Published

literature documenting SOFC tri-generation systems has focussed on either simulation or larger commercial based studies,

utilising vapour absorption cooling. Little or no work has been found on experimental based domestic scale studies. There is a

clear need for experimental work in both; fuel cell tri-generation systems operating in residential applications and fuel cell tri-

generation systems incorporating liquid desiccant air conditioning technology. This need is due to the technologies suitability for

integration and the patent advantages they can bring to total system efficiency. It is hoped work from the TriSOFC project will

assist in contributing towards filling this significant gap. Next, liquid desiccant tri-generation systems will be examined.

3.2 LIQUID DESICCANT BASED TRI-GENERATION SYSTEMS

This section will provide a brief review of tri-generation systems employing liquid desiccant air conditioning technology. An

interesting study of a system employing an internal combustion engine and vapour absorption cooling is provided by Liu, Geng et

al. [19]. The study illustrates the benefits a desiccant based air conditioning system can bring to a tri-generation system setup. Liu,

Geng et al. [19] investigated a hybrid tri-generation HVAC system in a multi-storey demonstration building, employing an internal

combustion engine, liquid desiccant and vapour absorption cooling technology. The investigation highlighted issues with

conventional tri-generation systems that use a straight absorption cooling cycle, these include: (1) low grade waste heat from the

prime mover is used directly to drive the cycle, leading to low energy performance and thus a CoP (thermal) of 0.7 or less (2) how

to match electrical and thermal loads of the building and lengthen the operating hours of the tri-generation system (thus improving

benefits from the system), and (3) dehumidification in an absorption cooling systems is achieved by cooling the air below its dew

point, then re-heating the air before supplying it to the room, this lowers the CoP because supply air temperature is required at 7-

12°C, it is also easier to cause health problems; condensed water makes the coil surface a hot bed for bacteria. However, the

authors found a way of avoiding the above problems when using a hybrid system consisting of liquid desiccant, absorption and

thermal/desiccant storage. The solutions are as follows: (1) the regeneration of liquid desiccant does not need high grade heat, only

70-80°C, thus CoP improvements of ~1.2 to 1.3 can be achieved (2) a thermal/desiccant storage tank can be used to match

differences between thermal supply and demand, and (3) the liquid desiccant system will remove the latent load, therefore

avoiding the need to cool then re-heat. A simulation was carried out to analyse the performance of the building with the different

system configurations i.e. with and without liquid desiccant. Results show that CO2 emissions are reduced by about 40%, and

energy storage plays an important role in the hybrid liquid desiccant system.

Qiu, Liu et al. [20] have carried out an experimental investigation of a liquid desiccant air conditioning system driven by the flue

gases of a biomass boiler. The air conditioning system consists of two parts; first, the dehumidifier and secondly the dew point

cooler. The desiccant dehumidifier used was of a direct contact counter flow falling film design, using a potassium formate

solution. Internal desiccant cooling was provided from an evaporative device. The system was operated in autumn in Nottingham,



5

UK and was found to be able decrease air temperature by 4°C and relative humidity by 13%. The total cooling capacity of the

system was found to be 2381W, the evaporative cooling device alone had a cooling capacity of 1049W, and illustrating the

dehumidifier plays a significant role in air conditioning by reducing the supply airs internal energy through dehumidification.

Electrical consumption for fans and pumps totalled ~ 200W; therefore the desiccant cooling systems CoP was in the range of 11-

12, an encouraging value. It was found that the heat from the flu gas was insufficient to heat the desiccant solution to the required

40°C, only 30°C was reached, as a result future work will look at improving the flue gas heat exchanger. This work has proved

that desiccant systems are effective cooling devices in waste heat scenarios, however issues such as waste heat recovery

effectiveness needs to be addressed in order to maximise operating performance. Finally, the experimental results confirm that

cooling is a result of lowering both humidity and air temperature, and high CoP values are attainable in a working environment –

critical in tri-generation system applications.

Liquid desiccant systems employed in tri-generation applications can offer environmental, economic and operational advantages

over other heat driven cooling cycles. Furthermore, vapour absorption cooling is rarely used in tri-generation applications of less

than 10kW [21], therefore liquid desiccant is a particularly well suited air conditioning option for domestic scale tri-generation

applications.

3.3 CO-GENERATION THERMAL OPTIMISATION FOR HOUSING INTEGRATION

This section will deal with how the full potential of co-generation systems may be realised. When a CHP system is grid interactive

i.e. electrical energy can be imported or exported, maximising the use of generated electrical energy is relatively simple. Excess

generation may be exported, and in periods of insufficient generation, electricity may be imported. However as Elmer [22] states it

is of greater value to the user to consume all/as much self-generated electricity onsite because the export tariff is lower than the

import tariff. Maximising the use of thermal energy generated by a CHP systems is however more complex. The degree of thermal

utilisation is a decisive factor in CHP system efficiency values, and will impact on cost and emission savings. The heat led nature

of combustion based CHP systems (internal combustion engine) means they should only be operational when demand for thermal

energy is present [23]. Fuel cells, particularly SOFCs have very different operational characteristics, most significantly a low H:P

ratio and the requirement of constant operation to avoid thermal cycling. Thermal cycling can lead to accelerated voltage

degradation within the cells. However, the majority of the UK housing stock is characterised by peak morning and evening heating

demands, shown in Figure 2a. How fuel cell CHP systems with their preference for constant operation are effectively integrated

into buildings is a challenging task.

Figure 2 (a) conventional UK heating profile left, and (b) slow heating demand right [24]

A number of studies have been carried out trying to address this issue. Kato, Iida et al. (2003) investigated the use of thermal

energy storage (TES) in a residential PEMFC application. It was found that a high level of temporal precision was required in

order to match the domestic hot water demand and to correctly size the system components. For a 2kWe installation, a 200-350

litre TES was required to effectively make use of the PEMFCs thermal output. Hawkes, Aguiar et al. (2007) present a SOFC CHP

model for use in a techno-economic analysis to determine options in the provision of residential heat demand in the UK. Four

different UK heating demand profiles were considered to see which profile best suited that of a SOFC CHP system using a cost

minimisation model. The model also included the use of TES. The base case model, shown in Figure 2a was based on a typical

UK heat load profile. Results showed that a ‘slow’ space heating demand, seen in Figure 2b, running constantly in winter with

under floor heating was most suitable for the SOFC due to maximum thermal energy utilisation. This profile achieved highest cost

and CO2 savings compared to the reference study of grid electricity and gas fired boiler. The slower heating demand is also a

better match for SOFC technology due to the avoidance of thermal cycling. TES provided added benefit to the system.

3.5 LITERATURE SEARCH CONCLUSIONS AND IMPACT ON TriSOFC

The review of literature presented in this paper has been used to provide a background study on the state of the art surrounding:

fuel cell tri-generation systems, desiccant tri-generation systems and thermal optimisation methods for residential co-generation

applications. The literature has been selected based on its appropriateness and suitability in addressing the central research aims of

the project. Three main conclusions of the review are summarised below:

(1) Currently there is large gap in the literature regarding fuel cell based tri-generation systems incorporating liquid desiccant

air conditioning technology. This is believed to be a result of three key factors: cost, system size/complexity and relative

infancy of technological development, particularly for domestic applications. It is intended that the TriSOFC project will



6

assist in contributing to filling this gap and addressing the proposed barriers of development, thus justifying the future use

of the proposed system concept.

(2) An opportunity that the literature review has highlighted is the rationale for TriSOFC. A fuel cell CHP system output

temperature (~60°C) is in good thermal agreement with the required desiccant regeneration temperature (45-50°C),

making integration of the two technologies a feasible and attractive option for low carbon buildings. However caution

will be taken due to the issues other researches have encountered when coupling ‘off the shelf’ components.

(3) Challenges the literature review has highlighted is that system integration, optimisation and energy utilisation is not a

large issue in theoretical or laboratory based projects, thus high system efficiencies have been reported. However in a real

world working environment, effective energy utilisation can pose a serious challenge to the system, specifically how to

maximise the utilisation of the energy outputs of the system in order to improve system efficiency. If this cannot be

effectively addressed it will have a large impact on the final feasibility of the system.

4. COMPONENT MODELLING Modelling work is an essential task for the design, building, testing and optimisation of the liquid desiccant air conditioning unit

used in the TriSOFC system. Modelling work will be used to: identify the important operating parameters of the desiccant unit to

facilitate its optimisation, for predicting the performance of the unit under set conditions and to assist in sizing a suitable unit for

TriSOFC. The 1-D model described below has been used to identify important operating parameters and establish initial

performance estimates.

The performance of a desiccant dehumidifier is strongly related to the quantity of water condensed from the humid air to the

desiccant solution in the dehumidifier, this quantity should be balanced with the amount of water vaporised in the regenerator [3].

In the literature there are numerous validated models predicting this quantity, however it should be stated that the majority of these

correlations are only valid to the specific desiccant unit and solutions for which the correlations were obtained. Therefore, a

simplified model to estimate the preliminary performance of the novel cellulose fibre packed bed system based on fundamental

equations would be valuable at this stage. The preliminary modelling task has called upon the work of Gandhidasan [3] and

Gandhidasan [25]. Gandhidasan describes a relatively simple model for the preliminary design of a liquid desiccant dehumidifier

and regenerator system, using dimensionless vapour pressure and temperature difference ratios Eq. (5) and (6). These ratios are

used to derive an expression to determine the mass of water vapour condensed from the air to the solution in the liquid desiccant

dehumidifier Eq. (3Error! Reference source not found., and for the mass of water vaporised from the desiccant solution to the

scavenging air stream in the regenerator Eq. (4). Both of these equations are used in terms of known operating parameters. The

model has been validated by well-regarded experimental data presented in the literature provided by Fumo and Goswami [4] with

very good agreement.

The dimensionless temperature difference ratio (β) is defined for the dehumidifier in Eq. (1) and for the regenerator in Eq. (2)

(1)

(2)

For the dehumidifier, the water condensation rate (m) as presented by Gandhidasan [3] is given in Eq. (3):

(3)

For the regenerator, the water evaporation rate (m) as presented by Gandhidasan [25] is given in Eq. (4):

(4)

The heat capacity rate of the desiccant solution (Cs) is defined in Eq. (5):

(5)

The heat capacity rate of the humid air (Ca) is defined in Eq. (6):

(6)



7

Where;

m = mass of water vapour condensed/vaporised (g/s) or per unit cross-sectional area (g/m2.s)

hfg = latent heat of vaporisation (kJ/kg)

G = mass flux or flow rate per unit cross-sectional area (kg/m2.s)

C = heat capacity rate (kW/°C)

Cp = Specific heat capacity (kJ/kg.K)

t = temperature (°C)

Greek letters

λ = latent heat of condensation (kJ/kg)

η = heat exchanger effectiveness (%)

β = dimensionless temperature difference ratio – Eq. (5)

Subscripts

a = air

c = cooling medium (water)

d = dehumidifier

HE = Heat exchanger

i = inlet

o = outlet

r = regenerator

s = desiccant solution

Data for the desiccant solutions needed to be acquired to use the aforementioned relationships. The specific heat capacities and

enthalpy of dilution is required for the desiccant solutions at different concentrations and temperatures. Curve fitting and linear

regression analysis has been used to do determine the required properties from the data presented by Conde [26] for LiCl and

CaCl2. Melinder [27] was used for the lesser known physical characteristics of the CHKO2 (Potassium Formate) solution. Figure 3

shows the simplified schematic of the desiccant system used in the modelling process, with the labelled desiccant solution and

airflows.

Figure 3 model schematic

3.1 DESICCANT SYSTEM WORKING PRINCIPLE

In the dehumidifier mass transfer (dehumidification) occurs due to differences in vapour pressures. The cool concentrated

desiccant solution exhibits a vapour pressure which is lower than that of the water vapour present in the humid air, therefore the

moisture moves from the air to the solution. In addition to this, in summer, it is expected that the temperature of the humid air that

enters the desiccant dehumidifier is higher than the desiccant solution temperature, therefore sensible heat transfer will also occur.

The greater the cooling of the desiccant solution prior to entering the dehumidifier, the greater the heat and mass transfer, and thus

the greater cooling output. Following dehumidification, the weak desiccant solutions needs to be regenerated i.e. re-concentrated.

This takes place in the regenerator. Prior to entering the regenerator the desiccant is heated, first in the desiccant to desiccant heat

exchanger (also used to pre-cool the desiccant flowing to the dehumidifier), then using an external source e.g. the fuel cell. In the

regenerator, the heated weak desiccant solution will have a vapour pressure higher than that of the scavenging air stream, thus the

water vapour present in the solution will transfer (vaporise) to the air stream, thus re-concentrating the desiccant solution ready for

dehumidification in the dehumidifier. The modelling work was carried out using Engineering Equation Solver (EES), a general

equation solving program that can numerically solve thousands of coupled non-linear algebraic and differential equations. Air and

water property routines are inbuilt functions in EES, making calculations involving psychrometric functions much easier. Once the

model had been developed, and adjusted such that the output results were in good agreement with the experimental and theoretical

results presented by Gandhidasan [3], Gandhidasan [25] and Fumo and Goswami [4], parametric analyses were carried out.

Parametric analysis for the dehumidifier and regenerator has looked at the variation of mass condensed/evaporated in the

dehumidifier/regenerator respectively with:



8

Cooling water temperature

Regeneration heating temperature

Fluid flow rates (air and desiccant)

Fluid flow rate ratios

Heat exchanger effectiveness

Desiccant solutions – LiCl, CaCl2 and CHKO2

Furthermore a parametric analysis focussing on the variation of the inlet air conditions to the desiccant dehumidifier was

investigated. This was to see the effect environmental conditions had on performance. In this analysis, the following were assessed

in terms of cooling output:

Air inlet temperature (ta,i)

Air inlet relative humidity (RHa,i)

Model boundary conditions (unless a varied parameter):

Air inlet temperature = 35°C

Air inlet RH = 70%

Desiccant inlet concentration = 35%

Desiccant inlet temperature = 30°C

Air mass flux = 1 (kg/m2.s)

Desiccant mass flux = 2 (kg/m2.s)

Air to desiccant flow rate ratio = 2

Heat exchanger effectiveness = 60%

Beta value = 0.5 – taken from the literature

Cooling water temperature = 29°C

Regeneration heating temperature = 70°C

All the solutions flow rates are given as mass flux values i.e. per m2 of contactor, thus all results are presented in terms of per m

2

of contactor area, either; mass condensed/vaporised or cooling output. The parametric analyses have been carried out for the LiCl

solution as this work has been validated with the well regarded work of Fumo and Goswami (2002). This study will allow the

working principles and underlying relationships involved in a desiccant air conditioning cycle to be identified. Following this, a

comparison of the cooling output potential of the three different desiccant solutions is made – LiCl, CaCl2 and CHKO2. The

modelling results are split into two sections: dehumidifier and regenerator.

3.2 DEHUMIDIFIER ANALYSIS

(i) Influence of cooling water temperature on mass condensed - LiCl

Figure 4 the effect of cooling water temperatures on dehumidification

Figure 4 shows how decreasing the cooling water temperature increases the mass flux condensed in the dehumidifier. This is due it

creating a lower desiccant temperature and thus increased dehumidification potential – the lower the temperature, the lower the

vapour pressure exhibited by the desiccant. For internally cooled cycles, it also means greater removal of the latent heat of

condensation. Gandhidasan [3] states that the performance of a desiccant dehumidifier is improved with pre-cooling, the lower the

better, thus a lower cooling water temperature yields improved dehumidification performance.

2 3 4 5 6 70

2

4

6

8

10

12

14

16

Desiccant flow rate (kg/m2.s)

tc,i = 29.5°C

tc,i = 29°C

tc,i = 28.5°C

ma

ss flu

x co

nd

en

se

d (

g/m

2.s

)



9

(ii) Influence of solution flow rate ratio on mass condensed and outlet RH - LiCl

Figure 5 the effect of desiccant solution flow rate ratio on dehumidification

Figure 5 shows that as the solution to air flow rate ratio increases, the mass flux condensed increases, and thus the RH of outlet air

decreases. This is because more desiccant solution is supplied per unit of air. However, as Lowenstein [6] states, the parasitic

energy load of supplying the desiccant solution and air is one of the largest issues regarding liquid desiccant systems as this has a

large influence on the CoP of the cycle. Therefore fluid supply flow rates need to be balanced against the parasitic energy load.

Furthermore if desiccant and air supply flow rates are too high it can create issues of desiccant carry-over to the supply air stream.

The cooling output of an air conditioning device is made up of both sensible and latent parts, and is usually a mixture of the two.

In the desiccant dehumidifier, latent cooling is due to mass transfer, and sensible cooling is due to the temperature difference

between the humid air and desiccant solution. The cooling output of the desiccant dehumidifier has thus been assessed using the

mass flux of air supplied multiplied by the difference in enthalpy between the inlet and outlet air, seen in Eq. (7). This gives a

cooling output per m2 of absorber.

(7)

Where:

Ga,d = mass flux of supplied air to the dehumidifier (kg/m2.s)

ha,in = inlet air specific enthalpy (J/kg)

ha,out = outlet air specific enthalpy (J/kg)

(iii) Influence of inlet air temperature on cooling output - LiCl

Figure 6 the effect of inlet air temperatures on cooling output

0.5 1 1.5 2 2.5 30

1

2

3

4

5

0.46

0.48

0.5

0.52

0.54

0.56

0.58

0.6

Yd kg solin / kg airin

Mass flu

x c

ondensed (

g/m

2.s

)

mass flux condensedmass flux condensed

RH

a,o

ut,

d (

%)

RHa,out,dRHa,out,d

Qcooling = Ga,d · ( ha,in – ha,out )

8400

8450

8500

8550

8600

8650

8700

8750

8800

8850

23 28 33 38 43

Co

oli

ng o

utp

ut

(W/m

2 a

bso

rber

)

Air inlet temperature (°C)



10

Figure 6 shows that as the inlet air temperature increases the cooling output for the LiCl solution increase, this is because there is a

greater temperature difference between the solution and inlet air, thus sensible heat transfer is increased, leading to an increased

cooling output.

(iv) The influence of inlet air relative humidity on cooling output - LiCl

Figure 7 the effect of inlet air RH on cooling output

Figure 7 shows as the relative humidity of the inlet air increases the cooling output also increases. This is because as the relative

humidity of the air increase so does its partial pressure, therefore representing a greater dehumidification potential for the

desiccant solution interacting with the humid air.

(v) A cooling output comparison for different desiccant solutions

Figure 8 cooling output comparisons for different desiccant solutions

Figure 8 shows that at the given boundary conditions, the Potassium Formate desiccant solution has the best performance, with

LiCl performing slightly better than the CaCl2 over the varied desiccant solution flow rates. This is because at the stated boundary

conditions, potassium formate has a higher specific heat capacity. As seen in Eq.(1) the specific heat capacity of the desiccant

solution has a large influence on mass condensed in the dehumidifier, which in turn has a large influence on cooling output.

8700

8720

8740

8760

8780

8800

0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9

Co

oli

ng o

utp

ut

(W/m

2 a

bso

rber

)

Inlet air relative humidity (%)

6000

11000

16000

21000

26000

31000

1.5 2.5 3.5 4.5 5.5 6.5

Co

oli

ng o

utp

ut

(W/m

2)

Desiccant flow rate (kg/m2.s)

Cooling output (W/m2) - LiCl Cooling output (W/m2) - CaCl2 Cooling output (W/m2) - CHKO2



11

3.3 REGENERATOR ANALYSIS

(i) The influence of regeneration temperature on mass vaporised - LiCl

Figure 9 influence of regeneration temperature on regeneration performance

From Figure 9 it is clear that as the desiccant flow rate and regeneration temperature increases, the mass vaporised in the

regenerator increases. This is because as the regeneration temperature increases, the desiccant temperature increases, leading to a

greater vaporisation potential. This shows how the maximum recoverable heat should be obtained from the fuel cell in order to

improve the desiccant system performance. An interesting point from Liu [28] is that when not using waste heat i.e. fuel cell, for

the regeneration of the liquid desiccant, the lowest usable concentration of desiccant solution is preferred, as this creates the

highest CoP. However when waste or renewable energy is available, the maximum recoverable heat possible should be targeted as

this will result in a higher concentration of desiccant, resulting in improved cooling capacity and system CoP.

(ii) Influence of heat exchanger effectiveness on mass vaporised - LiCl

Figure 10 influence of heat exchanger effectiveness on regeneration performance

Figure 10 shows that an increase in heat exchanger effectiveness improves mass vaporised; again this is due to the increase in

desiccant temperature i.e. more heat is exchanged to the desiccant solution. Both in Figure 9 and Figure 10, mass vaporised

increases with descant solution flow rate, however as stated with the dehumidifier, increasing this needs to be balanced against

parasitic energy consumption.

2.5 3 3.5 4 4.5 5 5.5 60

10

20

30

40

50

60

Desiccant flow rate [kg/m2.s]

Eva

po

ratio

n r

ate

[g

/m2

.s]

th,r,i = 68°C

th,r,i = 70°C

th,r,i = 72°C

Y = 2 kg solin / kgair in

Concin = 34%

Air flow Ga = 1 kg/m2.s

2 2.5 3 3.5 4 4.5 5 5.5 60

10

20

30

40

50

60

70

Desiccant solution flow rate [kg/m2.s]

Eva

po

ratio

n r

ate

[g

/m2

.s]

HXe = 0.6

Concin = 34%

HXe = 0.65

HXe = 0.7



12

4.4 MODELLING CONCLUSIONS

The parametric analyses have been used to identify the parameters that have the greatest influence on the performance of the

desiccant dehumidifier and regenerator. The work has also shown the optimum environmental conditions for which the system

should be operated in.

The moisture removal rate of the supply air is strongly controlled - assuming constant mass flow rate, by desiccant temperature

and cooling water temperature. The air inlet conditions have a large influence on performance in terms of cooling output. The unit

will generally perform better in a hot and more humid climate such as southern China, as opposed to drier cooler climates such as

the UK. In this model, the potassium formate desiccant solution shows the highest performance for the given conditions, and with

its lower environmental impact compared to LiCl and CaCl2, is a suitable solution for the desiccant unit in the TriSOFC system.

As the regeneration heat source temperature increases, the mass of water vapour vaporised from the weak desiccant solution

increases, although as Liu [28] explains, depending on the source of regeneration heat, too high a desiccant concentration is not

always beneficial to system performance. If waste heat is available the maximum concentration of desiccant solution should be

obtained. A higher regeneration temperature can also mean a lower desiccant solution and air flow rate are required (beneficial to

CoP) in order to yield a similar solution outlet concentration and for the mass balance between the dehumidifier and regenerator to

be satisfied.

Although emergent, this modelling has illustrated the fundamentals of liquid desiccant air conditioning systems, namely the

interactions observed in the parametric analysis. The current model does however have its limitations and is not particularly

robust; for example it does not include the heat and mass exchanger effectiveness of the cellulose fibre membrane, geometry of the

contactor or desiccant carry-over. Furthermore, there have been some issues encountered with the described model such as

assessing the variation of desiccant concentration and mass flux condensed/vaporised. However the model has provided a valuable

quick evaluation tool for the operational performance of liquid desiccant dehumidification systems. As previously stated, a

separate paper presented at this conference (Paper ID: SET2013 – 334) reports the development of computer simulations carried

out to evaluate the novel fibre membrane contacting design used in the liquid desiccant air conditioning unit (heat and mass

exchange effectiveness) and size the system.

5. CONCLUSIONS This paper has sought to provide an introduction to the EU funded project - Durable Solid Oxide Fuel Cell Tri-generation System

for Low Carbon Buildings (TriSOFC), through a project description, a review of the relevant literature and initial modelling work.

The system is a novel concept on two levels: (1) the integration of SOFC and liquid desiccant air conditioning technology in a low

carbon domestic building environment, and (2) the use of novel materials and construction techniques in the fuel cell stack, and

novel materials and design in the liquid desiccant air conditioning system.

A concise review of the literature surrounding the topics of fuel cell tri-generation and desiccant based tri-generation systems has

been presented. The most significant conclusions from this are: (1) a considerable gap in the literature currently exists

surrounding: tri-generation energy systems based on SOFC - liquid desiccant air conditioning technology and fuel cell based tri-

generation energy systems in residential applications, (2) the thermal output of a fuel cell CHP system is in good thermal

agreement with desiccant regeneration temperature, making them appropriate companion technologies for integration in a tri-

generation system concept, and (3) thermal energy utilisation in CHP and tri-generation systems is challenging, and has a large

impact on system efficiency. Energy utilisation issues are not so prevalent in simulation and laboratory based projects, however in

real world application studies such as TriSOFC they are very much apparent. Initial modelling work focussed on the desiccant air

conditioning system components has been presented. Parametric analyses have been carried out, and the parameters with the most

influence on performance commented upon. The modelling has been carried out for both the desiccant dehumidifier and

regenerator. Further modelling work related to the desiccant air conditioning system that will be used in TriSOFC can be referred

to in: Paper ID: SET2013 – 334.

The TriSOFC system is expected to provide an environmentally friendly way of providing heating, cooling and electricity to low

carbon buildings. Future work that can be expected from the project will include system modelling focussing on tri-generation

system capacities and efficiencies, experimental results from the desiccant air conditioning and tri-generation system and energetic

and economic analysis. This will be the work of future papers.

ACKNOWEDGEMENTS

The TriSOFC project is supported by the European Commission, under the Fuel Cell and Hydrogen joint Undertaking (FCH-JU)

initiative. Please refer to www.trisofc.com for more information. The authors would also like to thank the EPSRC and DTC in

hydrogen, fuel cells and their applications for their continued financial and academic support.

http://www.trisofc.com/



13

REFERENCES

1. Steele, B., Fuel-cell technology: Running on natural gas. Nature, 1999. 400: p. 619-621.

2. Riffat, S., Durable Solid Oxide Fuel Cell Tri-generation System for Low Carbon Buildings. 2012.

3. Gandhidasan, P., A simplified model for air dehumidification with liquid desiccant. Solar Energy, 2004. 76(4): p. 409-

416.

4. Fumo, N. and D.Y. Goswami, Study of an aqueous lithium chloride desiccant system: air dehumidification and desiccant

regeneration. Solar Energy, 2002. 72(4): p. 351-361.

5. Conde, M. LIQUID DESICCANT-BASED AIR-CONDITIONING SYSTEMS — LDACS. in 1st European Conference on

Polygeneration. 2007. Tarragona, Spain.

6. Lowenstein, A., Review of Liquid Desiccant Technology for HVAC Applications. American Society of Heating,

Refrigerating and Air-Conditioning Engineers, 2008. 14(6).

7. Al-Sulaiman, F.A., I. Dincer, and F. Hamdullahpur, Energy analysis of a trigeneration plant based on solid oxide fuel

cell and organic Rankine cycle. International Journal of Hydrogen Energy, 2010. 35(10): p. 5104-5113.

8. Bhatti, M.S., et al., Solid oxide fuel cell assisted air conditioning system. 2010: USA.

9. Jeong, J., et al., Performance analysis of desiccant dehumidification systems driven by low-grade heat source.

International Journal of Refrigeration, 2011. 34(4): p. 928-945.

10. Tse, L.K.C., et al., Solid oxide fuel cell/gas turbine trigeneration system for marine applications. Journal of Power

Sources, 2011. 196(6): p. 3149-3162.

11. Wang, Y., et al., An investigation of a household size trigeneration running with hydrogen. Applied Energy, 2011. 88(6):

p. 2176-2182.

12. CFCL, BlueGEN Modular Generator - Power and Heat. 2009.

13. Larmine, J. and A. Dicks, Fuel Cell Systems Explained. Second ed. 2003, Chichester: John Wiley & Sons Ltd.

14. Al-Sulaiman, F.A., I. Dincer, and F. Hamdullahpur, Exergy analysis of an integrated solid oxide fuel cell and organic

Rankine cycle for cooling, heating and power production. Journal of Power Sources, 2010. 195(8): p. 2346-2354.

15. Yu, Z., et al., Analysis of total energy system based on solid oxide fuel cell for combined cooling and power applications.

International Journal of Hydrogen Energy, 2010. 35(7): p. 2703-2707.

16. Margalef, P. and S. Samuelsen, Integration of a molten carbonate fuel cell with a direct exhaust absorption chiller.

Journal of Power Sources, 2010. 195(17): p. 5674-5685.

17. Braun, R.J., S.A. Klein, and D.T. Reindl, Evaluation of system configurations for solid oxide fuel cell-based micro-

combined heat and power generators in residential applications. Journal of Power Sources, 2006. 158(2): p. 1290-1305.

18. Pilatowsky, I., et al., Simulation of an air conditioning absorption refrigeration system in a co-generation process

combining a proton exchange membrane fuel cell. International Journal of Hydrogen Energy, 2007. 32(15): p. 3174-

3182.

19. Liu, X.H., et al., Combined cogeneration and liquid-desiccant system applied in a demonstration building. Energy and

Buildings, 2004. 36(9): p. 945-953.

20. Qiu, G., H. Liu, and S. Riffat, Experimental investigation of a liquid desiccant cooling system driven by flue gas waste

heat of a biomass boiler International Journal of Low-Carbon Technologies, 2012.

21. Pietruschka, D., et al., Experimental performance analysis and modelling of liquid desiccant cooling systems for air

conditioning in residential buildings. International Journal of Refrigeration, 2006. 29(1): p. 110-124.

22. Elmer, T. and S. Riffat, STATE OF THE ART REVIEW: FUEL CELL TECHNOLOGIES IN THE DOMESTIC BUILT

ENVIRONMENT, in International Conference on Sustainable Energy Technologies. 2012: Vancouver, Canada.

23. Beaussoleil-Morrison, I. (2008) An Experimental and Simulation-Based Investigation of the Performance of Small Scale

Fuel Cell and Combustion-Based Cogeneration Devices Serving Residential Buildings. Annex 42 of the International

Energy Agency’s Energy Conservation in Buildings and Community Systems Programme.

24. Hawkes, A.D., et al., Solid oxide fuel cell micro combined heat and power system operating strategy: Options for

provision of residential space and water heating. Journal of Power Sources, 2007. 164(1): p. 260-271.

25. Gandhidasan, P., Quick performance prediction of liquid desiccant regeneration in a packed bed. Solar Energy, 2005.

79(1): p. 47-55.

26. Conde, M., Aqueous solutions of lithium and calcium chlorides: property formulations for use in air conditioning

equipment design, M.C. Engineering, Editor. 2009.

27. Melinder, A., Thermophysical Properties of Aqueous Solutions Used as Secondary Working Fluids, in Energy

Technology. 2007, KTH Energy and Environmental Technology: Stockholm.

28. Liu, S., A Novel Heat Recovery/Desiccant Cooling System, in Architecture and Built Environment. 2008, The University

of Nottingham: Nottingham.

a novel solid oxide fuel cell tri-generation -

Documents