Composite materials for
solid sorption heat pumps: measurements, model calculations and
consequences for system performance M. van der Pal
J.B.J. Veldhuis
R. de Boer
G.D. Elzinga
June 2014
ECN-L--14-028
2014 INTERNATIONAL SORPTION HEAT PUMP CONFERENCE
(ISHPC2014), March 31 – April 3, 2014
University of Maryland, College Park, MD, U.S.A.
Composite materials for solid sorption heat
pumps: measurements, model calculations
and consequences for system performance
Michel van der Pal, Jakobert Veldhuis,
Robert de Boer, Gerard Elzinga
1 Paper ID Number 79
Contents
• Introduction: why composite materials?
• Introduction: experience from earlier work
• Composite materials: selection and preparation
• Analysis: thermal conductivity: measurements and results
• Analysis: cyclic stability: measurements and results
• Discussion & conclusions
Introduction
• ECN is working on thermochemical heat transformers based on: – LiCl(1-3)NH3 reaction in combination with MgCl2(2-6)NH3 for upgrading industrial
waste heat to process heat
– CaCl2(2-4-8)NH3 reaction in combination with MnCl2(2-6)NH3 and a compressor for upgrading low temperature waste heat to process heat
• Crucial for success of heat transformer: – Sufficient transfer(rate) of sorption heat to/from salt from/to Hex
– Stable operation over many cycles
Why composite materials?
• Typical ammonia-salts properties are: – Poor thermal conductivity: only reasonable power density with thin layers = low COP
– Poor strength: cannot create thick layers of salts with stable conditions
– Volume changes: resulting in poorer connection with HEx
– Tendency to become smaller and smaller crystals: loss of salt
• Composite materials can: – Improve thermal conductivity, resulting in shorter cycle times = higher power density
– Provide structure for salts, including porosity and room to swell/shrink
– Immobilise the salt
Attempts so far…
Composite selection and preparation
• Based on literature most promising material for salt-composite: ENG
• Preparation based on literature info – Method 1: mixing of ENG and salts followed by pressing to desired density
– Method 2: pressing (or purchasing) ENG to desired density followed by impregnating with salt from solution
Results preparation
Solids mixing of ENG and salts Impregnating ENG with salts
Analyses
• Thermal conductivity using Hotdisk thermal conductivity meter
• Cycle stability using micro flow reactor
Hotdisk measurement
Principle: - Transient plane method - Constant heat flux provided by
sensor for set period - Simultaneous with heat flux:
measurement of electrical resistance of sensor proportional to temperature
Hotdisk measurement
• Assumed: – lradial ≠ laxial
– Heat capacity = sum of heat capacity ENG and heat capacity CaCl2
– No heat flow at sides of the sample
– Temperature top of the sample is equal to ambient
• Conditions: – Temperature increase measured in 1 to 80 seconds intervals with power input chosen
for optimal temperature rise (too low = poor signal/noise ratio, too high = too many second-order effects)
– ENG measured with and without CaCl2
• Analysis: – Comparison with COMSOL model with various values for lradial and laxial
Hotdisk results for ENG-CaCl2.xNH3
0
1
2
3
4
5
0.01 0.1 1 10 100
tem
pe
ratu
re c
han
ge p
er
un
it o
f p
ow
er
(K/W
att)
time (s)
0.4W 80s
0.4W 80s
0.4W 80s
0.7W 20s
0.7W 20s
0.7W 20s
1W 5s
1W 5s
1W 5s
2W 1s
2W 1s
2W 1s
0
1
2
3
4
5
0.01 0.1 1 10 100
tem
pe
ratu
re c
han
ge p
er
un
it o
f p
ow
er
(K/W
att)
time (s)
measured
COMSOL-1
COMSOL-2
COMSOL-3
COMSOL-4
COMSOL-1: 8/2.5 W/mK
COMSOL-2: 8/1.5 W/mK
COMSOL-3: 8/1.0 W/mK
COMSOL-4: 16/1 W/mK
Setup: Micro flow reactor
N2
NH3Off gas
12
Furnace with quartz reactor (ID 4-20 mm)
Pos 1 is K-type thermocouple to control
Pos 2 is K-type thermocouple to measure
NH3/N2
Mass flow controllers Mass flow meter
Sample area
Swagelok cap
gas inlet
gas outlet
Position for measurement
T-couple
Position
control
T-couple Porous Quartz
support
Composite sample
Micro flow reactor measurement
• Reactor and gas temperature set at constant value of 40oC
• Mass flow controller set at 10 ml/min
• Sample using ENG + CaCl2
• Over 100 times: 1 hour NH3 flow (adsorption), 1 hour N2 flow (desorption)
• Measured parameters: – Flow out
– Temperature composite
– Temperature gas
• Visual inspection after 100+ cycles
Temperature profiles cycle tests
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0 10 20 30 40 50 60
Te
mp
era
ture
dif
fere
nce
[
C]
Time /minutes
cycle 1, TI02
cycle 10, TI02
cycle 20, TI02
cycle 30, TI02
cycle 40, TI02
cycle 50, TI02
cycle 60, TI02
cycle 70, TI02
cycle 80, TI02
cycle 90, TI02
cycle 95, TI02
-7.0
-6.0
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
0 10 20 30 40 50 60Te
mp
era
ture
dif
fere
nce
[ C
]Time /minutes
cycle 1, TI02
cycle 10, TI02
cycle 20, TI02
cycle 30, TI02
cycle 40, TI02
cycle 50, TI02
cycle 60, TI02
cycle 70, TI02
cycle 80, TI02
cycle 90, TI02
cycle 95, TI02
NH3-flow (adsorption) N2-flow (desorption)
NH3 flow out profiles cycle tests
0
2
4
6
8
10
12
0 10 20 30 40 50 60
Ou
tle
t fl
ow
[m
l/m
in]
Time /minutes
cycle 1, flow out
cycle 10, flow out
cycle 20, flow out
cycle 30, flow out
cycle 40, flow out
cycle 50, flow out
cycle 60, flow out
cycle 70, flow out
cycle 80, flow out
cycle 90, flow out
cycle 95, flow out
NH3-flow (adsorption)
Materials before and after 100+ cycles
Discussion and conclusions
• Hotdisk measurements: – Thermal conductivity of ENG+CaCl2 seems very comparable with ENG properties
– Assumed/unknown: do ENG and salt heat up at the same time?
– COMSOL model calculations give good insight but less suited for quantifying thermal conductivity very precisely
• Microflow measurements: – Experimental method successful for determining composite stability regarding
sorption behavior
– Stable operation after first 10 cycles
– Effect on material itself cannot be determined, visual inspection required
– Setup allows for long-term testing of material (10,000+ cycles)
Questions?
8
ECN
Westerduinweg 3 P.O. Box 1
1755 LE Petten 1755 LG Petten
The Netherlands The Netherlands
T +31 88 515 4949
F +31 88 515 8338
info@ ecn.nl
www.ecn.nl