Subscriber access provided by TULANE UNIVERSITY
Article
The breathing cell: cyclic intermembrane distance variation in reverse electrodialysis Jordi Moreno, Esther Slouwerhof, David A. Vermaas, Michel Saakes, and Kitty Nijmeijer Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b02668 • Publication Date (Web): 19 Sep 2016 Downloaded from http://pubs.acs.org on September 21, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 27
Environmental Science & Technology
51x35mm (300 x 300 DPI)
ACS Paragon Plus Environment
Environmental Science & Technology
1
The breathing cell:
2
cyclic intermembrane distance variation in reverse electrodialysis
3
J. Moreno1,2, E. Slouwerhof1,2, D. A. Vermaas1, 2, M. Saakes1, K. Nijmeijer*, 2,3
4
1
Wetsus, European Centre of Excellence for Sustainable Water Technology, Oostergoweg 9, 8911MA Leeuwarden, The
5
Netherlands
6 7 8
Page 2 of 27
2
Membrane Science & Technology, University of Twente, P.O. Box 217, 7500AE Enschede, The Netherlands
3
Current address: Membrane Materials and Processes, Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands
9
*corresponding author:
[email protected] 10
Abstract
11
The breathing cell is a new concept design that operates a reverse electrodialysis stack by varying
12
in time the intermembrane distance. Reverse electrodialysis is used to harvest salinity gradient
13
energy; a rather unknown renewable energy source from controlled mixing of river water and
14
seawater. Traditionally, both river water and seawater compartments have a fixed
15
intermembrane distance. Especially the river water compartment thickness contributes to a large
16
extent to the resistance of the stack due to its low conductivity. In our cyclic approach, two stages
17
define the principle of the breathing concept; the initial stage, where both compartments
18
(seawater and river water) have the same thickness and the compressed stage, where river water
19
compartments are compressed by expanding the seawater compartments. This movement at a
20
tunable frequency allows reducing stack resistance by decreasing the thickness of the river water
21
compartment without increasing permanently the pumping losses. The breathing stacks clearly
22
benefit from the lower resistance values and low pumping power required, obtaining high net
23
power densities over a much broader flow rate range. The high frequency breathing stack (15 1 ACS Paragon Plus Environment
Page 3 of 27
Environmental Science & Technology
24
cycles/min) shows a maximum net power density of 1.3 W/m2. Although the maximum gross and
25
net power density ever registered (2.9 W/m2 and 1.5 W/m2, respectively) is achieved for a fixed
26
120 µm intermembrane distance stack (without movement of the membranes), it is only obtained
27
at a very narrow flow rate range due to the high pressure drops at small intermembrane distance.
28
The breathing cell concept offers a unique feature, namely physical movement of the membranes,
29
and thus the ability to adapt to the operational conditions and water quality.
30
31
Introduction
32
The world’s increasing demand for energy associated with CO2-emission and fossil fuel depletion
33
has raised the interest in renewable and sustainable energy. Cleaner alternatives need to be
34
found to reduce the emission of greenhouse gases1,2. Currently, solar-, wind-, biomass- and hydro-
35
energy are well-known and used as sustainable energy sources. A rather new and unknown
36
energy source is salinity gradient energy. Salinity gradient energy is a potentially large source for
37
electricity generation. The Gibbs free energy of mixing is available when concentrated and diluted
38
salt solutions mix. The potential power obtainable from mixing seawater and river water is large:
39
approximately 2 TW when using the global discharge of rivers into the seas3,4. Reverse
40
electrodialysis (RED) is used to harvest this energy. Other technologies to harvest salinity gradient
41
power (SGP) are pressure retarded osmosis (PRO)5,6 and capacitive mixing (CAPMIX)7,8. A RED
42
stack comprises an alternating series of cation exchange membranes (CEM) and anion exchange
43
membranes (AEM). CEMs contain fixed negative charges allowing ideally only the passage of
44
positively charged ions and AEMs contain fixed positive charges allowing ideally only the passage
45
of negatively charged ions. The membranes are traditionally kept at a fixed distance using non-
2 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 27
46
conductive net-spacers. The resulting potential difference over the membranes can be used to
47
power an electrical device.
48
Despite of the good prospects for this renewable energy source9, RED technology is not yet
49
commercially attractive because of the relatively low gross power densities obtained until now; a
50
net power density of 2 W/m2 is the target to be commercially attractive10–12. To achieve a high net
51
power density, the internal resistance of the RED stack and the hydraulic friction of the feedwater
52
should be as low as possible13.
53
The internal resistance consists of the ohmic resistance and the non-ohmic resistance. The ohmic
54
resistance is determined by three components: the resistance of the membranes, that of the net-
55
spacers and that of the two feedwater compartments. The non-ohmic resistance includes the
56
boundary layer resistance (also known as concentration polarization) and the effects due to the
57
change in concentration of the bulk solution along the length of the channel. By reducing the
58
intermembrane distance of the feedwater compartments the ohmic resistance can be reduced13.
59
The dominant resistance is the river water resistance due to its low salt concentration and a
60
reduced intermembrane distance immediately induces a lower ohmic resistance13. At the same
61
time, the use of thin intermembrane distances increases the hydraulic friction and consequently
62
increases the pumping power. This effect is even stronger when net-spacers are used to keep
63
adjacent membranes at the same inter-mutual distance. Vermaas et al.13 achieved, when using
64
0.507 M (30 g NaCl per kg water) as seawater and 0.017 M (1 g NaCl per kg water) as river water,
65
a gross power density of 2.2 W/m2, but a net power density of only 1.2 W/m2 using a fixed
66
intermembrane distance of 100 µm13. Kim et al.14 achieved a slighty higher gross power density of
67
2.4 W/m2 but used much higher salt concentrations than the previous authors, i.e. 0.58 M as
68
seawater and 0.017 M as river water, at the same fixed intermembrane distance (i.e. 100 µm).
3 ACS Paragon Plus Environment
Page 5 of 27
Environmental Science & Technology
69
The use of net-spacers also increases the fouling behaviour of the stack when using natural feed
70
waters15,16. The high pumping power required at low intermembrane distances narrows the flow
71
rate range where the optimal net power output is achieved. Thin intermembrane distances ( 99% according to
173
the manufacturer) and because of their much higher flexibility compared to that of thicker
174
membranes. Also, these membranes have a low ionic resistance of 0.94 Ω·cm2 for the CEM and
175
0.34 Ω·cm2 for the AEM measured in Cl- form in 0.5 M NaCl at 25 °C.
176
Two titanium electrodes (mesh size 1.7 m2/m2) with 50 g/m2 galvanic platinum coating and with
177
an active area of 10x10 cm2 were used as the anode and cathode (Magneto Special Anodes B.V.,
178
The Netherlands). The electrode rinse solution used to facilitate the redox reactions at the
179
electrodes was composed of 0.3 M K3Fe(CN)6 , K4Fe(CN)6 and 0.5 M NaCl in demineralized water.
180
The electrode rinse solution was circulated at 3 cm3/s along the electrodes by using a peristaltic
181
pump (Cole-Parmer, Masterflex L/S Digital drive USA). The electrode rinse solution compartment
182
was created by placing a silicon rubber gasket with a thickness of 300 ± 5 µm. The compartment
183
was kept at an over pressure of 30 ± 10 kPa with the help of a needle valve in order to avoid
184
undesired bulging of the feed water compartments.
9 ACS Paragon Plus Environment
Page 11 of 27
Environmental Science & Technology
185
To introduce the ‘breathing’ of the cell, an electronic one-way valve (Bürkert, Germany) was
186
positioned at the outlet of the seawater compartments. This valve was automatically controlled
187
by a Raspberry Pi (Raspberry Pi Foundation, United Kingdom), using an open source python code.
188
The breathing was operated at two different on/off frequencies of 5 cycles/min and 15
189
cycles/min.
190
2. Feed water
191
Artificial seawater and river water were used with a concentration of 0.507 M (30 g NaCl per kg
192
water) and 0.017 M (1 g NaCl per kg water), respectively. These solutions were made with NaCl
193
(technical grade, ESCO, The Netherlands) dissolved in demineralized water. The solutions were
194
kept at 25 oC ± 0.5 oC with a heater (Tetratec HT300, Germany) and a pump. The artificial solutions
195
were pumped continuously through the stack by using two peristaltic pumps (Cole-Parmer,
196
Masterflex L/S Digital drive, USA). Measurements were performed in duplicate during 360
197
seconds per flow rate at 0.8, 1.6, 2.5, 3.3, 4.1, 5 and 5.8 cm3/s in random order to avoid history
198
effects. Flow velocities ranged from 0.69 cm/s to 5.55 cm/s for the 120 µm stack and 0.17 cm/s to
199
1.38 cm/s for the 480 µm stack. The Reynolds number (Re) is calculated being defined as the
200
Reynolds number for a wide empty channel (not corrected for the spacer porosity).
201
=
202
in which is the average flow velocity (m/s), Dh is the hydraulic diameter (m), is the density of
203
water (kg/m3), is the (dynamic) viscosity of water (kg/(m·s)), Φ is the flow rate per feedwater
204
compartment (m3/s) and b is the width of the feedwater compartment (m).
∙
∙
≈
∙
(eq.1) ∙
10 ACS Paragon Plus Environment
Environmental Science & Technology
205
Page 12 of 27
3. Pressure drop measurements
206
At the inlet and the outlet of the feed waters, pressure meters (Endress+Hauser Cerabar T,
207
Germany) were placed to measure the pressure drop over the seawater and river water
208
compartments. The pressure meters were connected to a data logger (Endress+Hauser Ecograph
209
T, Germany) to obtain the data. The data logger measured the pressure every second. Before
210
every experiment, air bubbles were removed from the tubes connected to the pressure meters, to
211
ensure an accurate measurement. The pressure drop values used to calculate the consumed
212
pumping energy are average pressure drop values measured during the full duration of the
213
experiment (300 seconds).
214
4. Electrochemical measurements
215
A chrono-amperometric series was applied using a potentiostat (Ivium Technologies, The
216
Netherlands). It involves an initial stage of 60 seconds without any current for the determination
217
of the open circuit voltage (OCV), followed by a stage of 300 seconds measuring the current at a
218
constant voltage equal to half of the OCV, i.e. at maximum current density. All experiments were
219
done in duplicate. During the stage without current, electrochemical impedance spectroscopy
220
(EIS) with an alternating sinusoidal signal of 5 KHz was applied to derive the ohmic resistance of
221
the stack. One hundred analyses per second were performed and the real component of the
222
impedance was used to extract the ohmic resistance. According to Długołęcki et al.
223
resistance measured at high frequencies and alternating current represents the resistances of the
224
membranes and the feed water compartments (i.e. ohmic resistance). Previous studies
225
investigated the ohmic resistance by applying different current steps13,25, but this method cannot
226
be applied in a breathing stack because the membranes move consequently changing the internal
227
resistance.
24
, the
11 ACS Paragon Plus Environment
Page 13 of 27
228 229 230
Environmental Science & Technology
5. Gross power density The gross power density can be calculated as follows: =
∙ (eq.2) !
231
where P is the gross power density (W/m2), V is the voltage (V), I is the current (A) and Astack is the
232
membrane stack area (0.2 m2).
233
6. Pumping energy
234
The power density required to pump the feed waters is determined from the measured pressure
235
drop over the feed water compartments (hydraulic losses). The pumping power per membrane
236
area (W/m2 membrane) can be calculated as 26,21:
237
"#" =
238
where Φ and Φ are the flow rates of river and seawater (m3/s) and ∆( and ∆( are the
239
pressure drops (Pa) over the river and seawater compartments.
240
∆"% ·% '∆" · (eq.3) !
7. Net power density
241
The net power density, Pnet (W/m2), is calculated as the difference between the gross power
242
density, Pgross (W/m2), and the power density consumed for pumping the feed waters, Ppump
243
(W/m2):
244
)*+ = − "#" (eq.4)
245
Results
246
1. Stack ohmic resistance
247
In order to predict individual contributions to the stack ohmic resistance, the ohmic resistance per
248
cell was calculated and compared with the experimentally measured values. In figure 3 the 12 ACS Paragon Plus Environment
Environmental Science & Technology
Page 14 of 27
249
calculated ohmic resistance per cell for the two stages of the breathing stack configuration is
250
presented. The resistance attributed to the membranes, the seawater compartment and the river
251
water compartment is calculated using the experimental membrane specifications and the
252
conductivity of the feed waters. To determine the contribution of the spacers (spacer shadow
253
effect 20,24,27) in Figure 3, the total ohmic resistance was measured experimentally both for the 120
254
µm and the 480 µm stacks and the difference between the calculated value and the measured
255
value is attributed to the effect of the spacers13. To account for this effect in the breathing stack,
256
the average value of this spacer effect of the 120 µm stack and that of the 480 µm stack is taken.
257
The membrane resistance is considered independent of the thickness of the two feed water
258
compartments, and consequently has the same resistance in both stages.
259 260 261 262
2
Figure 3. Calculated area resistance per cell (in Ω⋅ cm ) originating from the spacers, the river water, the seawater and the membranes for a breathing cell at the initial stage and the compressed stage. Experimental data for membrane resistance and conductivity of the feed waters are used as input values.
13 ACS Paragon Plus Environment
Page 15 of 27
263 264 265
Environmental Science & Technology
The thickness of the expanded seawater compartment (840 µm) originates from subtracting 120 µm (thickness of a totally compressed river water compartment) from the sum of the thickness of the two feed water compartments together at the initial stage (480 µm each = 960 µm).
266 267
One of the most beneficial aspects of the breathing cell concept is the inter-membrane distance
268
variation in time of the feed water compartments, especially for that of the river water
269
compartment. The calculated values suggest a 47% lower ohmic resistance between the initial
270
stage of the breathing cell (480 µm in both compartments) and the compressed stage (120 µm for
271
the compressed river water compartment and 840 µm for the expanded seawater compartment).
272
The thickness of the expanded seawater compartment (840 µm) originates from subtracting 120
273
µm (thickness of a totally compressed river water compartment) from the sum of the thickness of
274
the two feed water compartments together at the initial stage (480 µm each = 960 µm). The
275
expansion of the sea water compartment results in an increase in resistance when the initial
276
stage and the breathing stage are compared. However, this increase is minor compared to the
277
decrease in resistance of the river water compartment upon compression. The dominance of the
278
river water resistance, having a low electric conductivity due to the low salt concentration,
279
provokes that any variation of the intermembrane distance in the river water compartment
280
influences the overall stack performance 13.
281
In order to validate and demonstrate experimentally the expansion of the seawater
282
compartments and thus the compression of the river water compartments in time,
283
electrochemical impedance spectroscopy measurements were conducted. In Figure 4 the
284
experimentally measured ohmic resistance at time intervals of 0.01 s (in-phase component of
285
impedance at 5 KHz) for different stack configurations is plotted against the time.
14 ACS Paragon Plus Environment
Environmental Science & Technology
Page 16 of 27
286 287 288 289 290 291 292 293 294
Figure 4. Ohmic resistance per cell as a function of the time measured for every stack configuration. All 3 configurations were operated at the same flow rate per cell (5 cm /s). Dashed lines correspond to the calculated values of the initial stage (480 µm/480 µm) and the compressed stage (120 µm/840 µm) of the breathing cell. The straight lines at the bottom (120 µm) and at the top (480 µm) of the graph, represent the reference stacks, i.e, a stack with 120 µm spacers in both seawater and river water compartment and a stack with 480 µm spacers in both seawater and river water compartment, respectively (standard stack, no breathing). The time interval is 0.01 s for the impedance measurements at 5000Hz i.e. 100 analyses per second.
295 296
The ohmic resistance of the reference stacks without the breathing movement (lower and upper
297
line, 120 µm and 480 µm) remains constant during the measurement time, as expected because
298
these stacks do not have any breathing movement. The 480 µm line lies above the theoretical 480
299
µm line as calculated for the breathing stack. The reason for this is the spacer shadow effect. In
300
the breathing configuration the spacer shadow effect is diminished as the inner part of the spacer
301
is only 120 µm (only the inlet and outlet are 480 µm ) and thus the spacer does not completely
302
touches the membrane area.
15 ACS Paragon Plus Environment
Page 17 of 27
Environmental Science & Technology
303
The stacks with breathing movement (5 cycles/min and 15 cycles/min) present a strong cyclic
304
variation of the ohmic resistance in time, demonstrating the movement of the membranes. The
305
membrane movement of the breathing stack at a frequency of 5 cycles/min reduces the ohmic
306
resistance by 30% in the compressed stage compared to the initial stage. At a frequency of 15
307
cycles/min the reduction in ohmic resistance is only 15% in the compressed stage compared to
308
the initial stage. Although one would expect that the resistances of the breathing stacks lay in
309
between the resistance lines of the calculated values for the initial stage and that of the
310
compressed stage, this is not the case, because the calculated values for the non breathing stacks
311
include the spacer shadow effect. During expansion of the seawater compartments, free space is
312
created in the seawater compartment (which has a spacer of only 480 µm thick, while the
313
compartment expands to 840 µm), thus eliminating momentarily the spacer coverage of the
314
membrane (spacer shadow effect), thus decreasing the ohmic resistance of the stack further more
315
than the expected with the theoretical calculations20,28,24,29. This is probably the main reason why
316
the ohmic resistance values of the breathing cell stacks (cyclic lines) are lower than the calculated
317
values of the compressed stage (dashed line).
318
Moreover, we expect that the river water compartment remains somewhat compressed after the
319
compressed stage and does not return completely to the initial stage of 480 µm/480 µm stack,
320
resulting in an overall reduction in ohmic resistance. This could be a consequence of keeping the
321
electrode rinse solutions under a slight overpressure to avoid uncontrolled bulging of the
322
membranes, leading to a slight compression of the whole stack16. This compression of the river
323
water compartment also explains the difference in ohmic resistance values between the breathing
324
stacks. The breathing stack with 15 cycles/min, remains more compressed than the 5 cycles/min
325
breathing stack. Due to the high breathing frequency, the compartments do not return
326
completely to their original position after each cycle. However, this relatively small effect is not 16 ACS Paragon Plus Environment
Environmental Science & Technology
Page 18 of 27
327
visible in the pressure drop measurements for both compartments as a significant part of the
328
pressure drop comes from the in- and outflows of the compartments.
329
2. Gross power density
330
Figure 5 shows the time-averaged gross power density calculated following equation 2 (standard
331
deviation is shown as error bar) and plotted against the flow rate for the different stack
332
configurations.
333 334 335
Figure 5. Gross power density as a function of the flow rate for the different stack configurations. Trend lines are added to guide the eye.
336 337
For all stack configurations the gross power density output increases when the flow rate per cell is
338
increased. At higher flow rates there is better re-supply of feedwater (i.e., keeping the maximum
339
concentration gradient along the compartment) and better mixing (decreased boundary layer
340
resistance)13,30. The breathing cell with a breathing frequency of 5 cycles/min shows a gross power
341
density decay at higher flow rates. This is due to the closing of the valve in the seawater 17 ACS Paragon Plus Environment
Page 19 of 27
Environmental Science & Technology
342
compartment that provokes internal leakage. Seawater is transferred to the river water
343
compartment through internal leakage and thus reduces the salinity gradient. This is an undesired
344
effect, but the consequence of the concept at certain specific conditions. The breathing stacks
345
show gross power density values between those of the two stacks with a fixed intermembrane
346
distance of 120 µm and 480 µm. A maximum value of 1.8 W/m2 for the gross power density was
347
achieved for the breathing stack with a breathing frequency of 15 cycles/min and a maximum
348
value of 1.6 W/m2 was achieved for the stack operating at a frequency of 5 cycles/min. The values
349
achieved for the stack operated at 15 cycles/min are higher than those obtained for the stack with
350
5 cycles/min. This again suggests that the river water compartments do not completely return to
351
their initial stage position after the compression stage, and the river water compartments remain
352
slightly compressed. Therefore higher flow velocities inside the feed water compartments are
353
achieved, although at the same flow rate. This results in a higher gross power density output. At
354
the same time, the high flow velocities and the movement of the membranes of the breathing
355
stacks disturb the diffusion boundary layer and improve the mixing of the feed waters near the
356
membrane surface, especially in the river water compartment24. The highest gross power density
357
value was achieved for the 120 µm fixed intermembrane distance stack. This stack, that also
358
benefits from the high flow velocities in both compartments, results in a gross power density of
359
2.9 W/m2, exceeding the previously reported value for gross power density of 2.2 W/m2,13 for the
360
same feed water concentrations and establishes a new maximum value for the gross power
361
density in RED. The 480 µm fixed intermembrane distance stack achieved a maximum gross power
362
density value of only 0.5 W/m2, especially due to the thick river water compartment resulting in a
363
high resistance. This value is comparable with a previous research 13.
18 ACS Paragon Plus Environment
Environmental Science & Technology
364
Page 20 of 27
3. Pumping power
365
In Figure 6Figure 62 the consumed pumping power density against the flow rate is presented.
366
Pumping power was calculated from experimentally measured values following equation 3
367
(standard deviation is shown as error bar). The pumping power increases with increasing flow rate
368
for all stack configurations. The pumping power is significantly pronounced when using the stack
369
with a fixed intermembrane distance of 120 µm, as expected. All stacks have an inlet net-spacer of
370
480 µm but the stack with an intermembrane distance of 120 µm has in both the seawater and
371
the river water compartment an inlet net-spacer of only 120 µm. Consequently, the pressure drop
372
and thus the pumping power in this stack configuration is much higher than in the other
373
configurations. As observed before13,31,32, the pressure drop is strongly dependent on the spacer
374
thickness.
375 376 377
Figure 62. Pumping power density plotted against the flow rate for every stack configuration. Trend lines are added to guide the eye.
378
19 ACS Paragon Plus Environment
Page 21 of 27
Environmental Science & Technology
379
The major part of the energy consumed for pumping the feeds, and thus the pressure drop,
380
predominantly comes from the friction at the inlet and outlets of the stacks. Consequently, as
381
both the breathing stacks and the stack with a fixed membrane distance of 480 µm have a much
382
wider inflow slit than the 120 µm stack, the three 480 µm stacks experience less friction and have
383
a much lower pressure drop than the narrower stack . Still the lowest pressure drop values are
384
achieved for the stack with a fixed intermembrane distance of 480 µm, due to the wide stack
385
compartments and consequently low friction inside the compartments. During a small fraction of
386
time, the breathing stacks, are in the compressed state, resulting in a temporary, short reduction
387
of the intermembrane distance and consequently a slight increase in pressure drop (which is
388
relatively minor compared to the pressure drop due to inflow effects). Since the pressure drop
389
values are averaged over one complete cycle, the average additional pressure drop due to
390
compression is also low. This demonstrates that the intermembrane distance variation in time
391
contributes to minimize the pumping power while keeping high gross power densities.
392
4. Net power density
393
In Figure 7Figure the net power density values, calculated as the difference between the gross
394
power density and the power consumed for pumping of the feed waters following equation 4
395
(standard deviation is shown as error bar), are shown as function of the flow rate for every stack
396
configuration.
20 ACS Paragon Plus Environment
Environmental Science & Technology
Page 22 of 27
397 398 399
Figure 7. Net power density plotted against the flow rate for every stack configuration. Trend lines are added to guide the eye.
400 401
The breathing stacks clearly benefit from the low pumping power values obtaining high net power
402
densities over a broader flow rate range. The optimal flow rate giving the highest net power
403
density is between 2 cm3/s and 4 cm3/s in the breathing stack configurations. The high frequency
404
breathing stack with 15 cycles/min shows a maximum value of 1.3 W/m2 and the stack with a low
405
breathing frequency of 5 cycles/min has a maximum net power density of 1.1 W/m2.
406
The stack with a fixed intermembrane distance of 120 µm registers a maximum net power density
407
value of 1.5 W/m2. This is currently the highest net power density measured experimentally
408
known to the authors. To the best of our knowledge, the maximum net power density reported in
409
literature was 1.2 W/m2 using the same feed water concentrations13. At slighty higher flow rates,
410
the pumping power increases tremendously (Figure 6), directly resulting in a significantly lower
411
net power density as well. 21 ACS Paragon Plus Environment
Page 23 of 27
Environmental Science & Technology
412
The stack with the wide compartments (a fixed membrane distance of 480 µm) shows the lowest
413
net power density due to the low gross power densities as a consequence of the dominance of
414
the high resistance of the river water compartment.
415 416
Outlook
417
The breathing cell concept offers a unique feature, namely physical movement of the membranes.
418
By compressing the river water compartment, the internal resistance of the stack decreases and
419
higher net power outputs can be achieved, despite the momentarily increase in pressure drop
420
along the compartment. This concept increases the robustness towards commercial applications
421
using real feed waters where stacks can be operated at a wide range of different flow rate
422
regimes if needed (e.g. adapting high or low flow rate or breathing frequency to the water
423
parameters and characteristics). However, new challenges concerning the (mechanical) lifetime
424
and integrity of the membranes have risen with this new stack concept and emphasizes the need
425
of more research about ion exchange membrane resistance towards continuous expansion and
426
stretching movements.
427
As the concentration of river water increases along the length of the stack, the effect of the
428
breathing will be more pronounced at the first stages of the stack than at the end of the stack,
429
where the salt concentration in the river water is significantly higher in real large-scale operation.
430
Earlier studies with sea and river water however, clearly showed the strong dominance of the low
431
conductivity of the river water on the power output in that case. As such for the seawater-river
432
water case, the breathing concept is an interesting option. When other feed waters with higher
433
conductivities are applied (e.g. seawater and water from the dead sea), this can be different, but
434
this needs to be considered for each specific case.
22 ACS Paragon Plus Environment
Environmental Science & Technology
Page 24 of 27
435 436
437
Acknowledgements
438
This work was performed in the cooperation framework of Wetsus, European Centre of Excellence
439
for Sustainable Water Technology (www.wetsus.eu). Wetsus is co-funded by the Dutch Ministry of
440
Economic Affairs and Ministry of Infrastructure and Environment, the Province of Fryslân, and the
441
Northern Netherlands Provinces. The authors like to thank the participants of the research theme
442
“Blue Energy” for the fruitful discussions and their financial support.
443
References
444 445 446
(1)
Olivier, J. G. J.; Van Aardenne, J. A.; Dentener, F. J.; Pagliari, V.; Ganzeveld, L. N.; Peters, J. A. H. W. Recent trends in global greenhouse gas emissions:regional trends 1970–2000 and spatial distributionof key sources in 2000. Environ. Sci. 2005, 2 (2–3), 81–99.
447 448 449
(2)
Change, U. N.-F. convention on climate. Paris Agreement http://unfccc.int/files/meetings/paris_nov_2015/application/pdf/paris_agreement_english _.pdf (accessed Apr 12, 2016).
450 451
(3)
Weinstein, J. N.; Leitz, F. B. Electric Power from Differences in Salinity: The Dialytic Battery. Science (80-. ). 1976, 191 (4227), 557–559.
452 453
(4)
Logan, B. E.; Elimelech, M. Membrane-based processes for sustainable power generation using water. Nature 2012, 488 (7411), 313–319.
454 455 456
(5)
Loeb, S.; Hessen, F. Van; Shahaf, D. Production of Energy from concentrated brines by Pressure-Retarded Osmosis, II. Experimental results and projected energy costs. J. Memb. Sci. 1976, 1, 249–269.
457 458
(6)
Gerstandt, K.; Peinemann, K. V; Skilhagen, S. E.; Thorsen, T.; Holt, T. Membrane processes in energy supply for an osmotic power plant. Desalination 2008, 224, 64–70.
459 460
(7)
Brogioli, D. Extracting Renewable Energy from a Salinity Difference Using a Capacitor. Phys. Rev. Lett. 2009, 58501, 1–4.
461 462 463
(8)
Brogioli, D.; Zhao, R.; Biesheuvel, P. M. A prototype cell for extracting energy from a water salinity difference by means of double layer expansion in nanoporous carbon electrodes. Energy Environ. Sci. 2011, 4 (3), 772–777.
464 465
(9)
Post, J. W. Blue Energy: electricity production from salinity gradients by reverse electrodialysis. Thesis, Wageningen University, 2009, Vol. PhD thesis. 23 ACS Paragon Plus Environment
Page 25 of 27
Environmental Science & Technology
466 467
(10)
Turek, M.; Bandura, B. Renewable energy by reverse electrodialysis. Desalination 2007, 205, 67–74.
468 469 470
(11)
Post, J. W.; Goeting, C. H.; Valk, J.; Goinga, S.; Veerman, J.; Hack, P. J. F. M. Towards implementation of reverse electrodialysis for power generation from salinity gradients. Desalin. water Treat. 2010, 16, 182–193.
471 472
(12)
Daniilidis, A.; Herber, R.; Vermaas, D. A. Upscale potential and financial feasibility of a reverse electrodialysis power plant. Appl. Energy 2014, 119, 257–265.
473 474
(13)
Vermaas, D. A.; Saakes, M.; Nijmeijer, K. Double Power Densities from Salinity Gradients at Reduced Intermembrane Distance. Environ. Sci. Technol. 2011, 45 (16), 7089–7095.
475 476 477
(14)
Kim, H.-K.; Lee, M.-S.; Lee, S.-Y.; Choi, Y.-W.; Jeong, N.-J.; Kim, C.-S. High power density of reverse electrodialysis with pore-filling ion exchange membranes and a high-open-area spacer. J. Mater. Chem. A 2015, 3 (31), 16302–16306.
478 479
(15)
Vermaas, D. A.; Kunteng, D.; Saakes, M.; Nijmeijer, K. Fouling in reverse electrodialysis under natural conditions. Water Res. 2013, 47 (3), 1289–1298.
480 481 482
(16)
Vermaas, D. A.; Kunteng, D.; Veerman, J.; Saakes, M.; Nijmeijer, K. Periodic Feedwater Reversal and Air Sparging As Antifouling Strategies in Reverse Electrodialysis. Environ. Sci. Technol. 2014, 48 (5), 3065–3073.
483 484
(17)
Kedem, O.; Rubinstein, I.; Segel, L. A. Reduction of polarization by ion-conductive spacers: theoretical evaluation of a model system. Desalination 1978, 27, 143–156.
485 486
(18)
Kedem, O. Reduction of polarization in Electrodialysis by ion-conducting spacers. Desalination 1975, 16, 105–118.
487 488 489
(19)
Shahi, V. K.; Thampy, S. K.; Rangarajan, R. The effect of conducting spacers on transport properties of ion-exchange membranes in electrodriven separation. Desalination 2001, 133, 245–258.
490 491 492
(20)
Długołęcki, P. E.; Dąbrowska, J.; Nijmeijer, K.; Wessling, M. Ion conductive spacers for increased power generation in reverse electrodialysis . J. Memb. Sci. 2010, 347 (1–2), 101– 107.
493 494
(21)
Vermaas, D. A.; Saakes, M.; Nijmeijer, K. Power generation using profiled membranes in reverse electrodialysis. J. Memb. Sci. 2011, 385–386 (0), 234–242.
495 496
(22)
Güler, E.; Elizen, R.; Saakes, M.; Nijmeijer, K. Micro-structured membranes for electricity generation by reverse electrodialysis. J. Memb. Sci. 2014, 458, 136–148.
497 498
(23)
Vermaas, D. A.; Hitsov, I.; Schaetzle, O.; Galama, O.; Saakes, M. Device and method for performing an ion-exchange process, 2013.
499 500 501
(24)
Długołęcki, P. E.; Ogonowski, P.; Metz, S. J.; Saakes, M.; Nijmeijer, K.; Wessling, M. On the resistances of membrane, diffusion boundary layer and double layer in ion exchange membrane transport. J. Memb. Sci. 2010, 349 (1–2), 369–379.
502 503 504
(25)
Sistat, P.; Pourcelly, G. Chronopotentiometric response of an ion-exchange membrane in the underlimiting current-range. Transport phenomena within the diffusion layers. J. Memb. Sci. 1997, 123 (1), 121–131.
505
(26)
Veerman, J.; Saakes, M.; Metz, S. J.; Harmsen, G. J. Electrical Power from Sea and River 24 ACS Paragon Plus Environment
Environmental Science & Technology
506 507
Page 26 of 27
Water by Reverse Electrodialysis: A First Step from the Laboratory to a Real Power Plant. Environ. Sci. Technol. 2010, 44 (23), 9207–9212.
508 509 510
(27)
Długołęcki, P. E.; Gambier, A.; Nijmeijer, K.; Wessling, M. Practical Potential of Reverse Electrodialysis As Process for Sustainable Energy Generation. Environ. Sci. Technol. 2009, 43 (17), 6888–6894.
511 512
(28)
Długołęcki, P. E.; Anet, B.; Metz, S. J.; Nijmeijer, K.; Wessling, M. Transport limitations in ion exchange membranes at low salt concentrations. J. Memb. Sci. 2010, 346 (1), 163–171.
513 514 515
(29)
Pawlowski, S.; Sistat, P.; Crespo, J. G.; Velizarov, S. Mass transfer in reverse electrodialysis: Flow entrance effects and diffusion boundary layer thickness. J. Memb. Sci. 2014, 471, 72– 83.
516 517
(30)
Vermaas, D. A.; Saakes, M.; Nijmeijer, K. Enhanced mixing in the diffusive boundary layer for energy generation in reverse electrodialysis. J. Memb. Sci. 2014, 453, 312–319.
518 519
(31)
Gurreri, L.; Tamburini, A.; Cipollina, A.; Micale, G. CFD analysis of the fluid flow behavior in a reverse electrodialysis stack. Desalin. Water Treat. 2012, 48 (1–3), 390–403.
520 521 522
(32)
Pawlowski, S.; Geraldes, V.; Crespo, J. G.; Velizarov, S. Computational fluid dynamics (CFD) assisted analysis of profiled membranes performance in reverse electrodialysis. J. Memb. Sci. 2015, 502, 179–190.
523 524
25 ACS Paragon Plus Environment
Page 27 of 27
525
Environmental Science & Technology
For table of contents only
526
26 ACS Paragon Plus Environment