scieee Science in your language
[sp] (orig)

Revisiting the cycling stability of ferrocyanide in alkaline media for redox flow batteries

Abstract

Spanish Government (MINECO) through the Research Challenges Programme (Grant RTI2018-099228-A-I00) as well as the Comunidad de Madrid through the Talent Attraction Programme (2017-T1/AMB-5190). A. M.-C. thanks Ministerio de Ciencia, Innovación y Universidades and MINECO for the financial support (CTQ2017-87231-P and RYC-2017-22700). E.V. thanks the MINECO for the financial support (RYC2018-026086-I)

Read accessible full text

Revisiting the cycling stability of ferrocyanide in alkaline media for redox flow batteries

Author: Páez, Teresa,Martínez Cuezva, Alberto,Palma, Jesús,Ventosa Arbaizar, Edgar
Publisher: Elsevier
Year: 2020
DOI: 10.1016/j.jpowsour.2020.228453
Source: https://riubu.ubu.es/bitstream/10259/5824/1/Paez-jps_2020.pdf
1
Re isi ing he Cycling S abili y o Fe ocyanide in
Alkaline Media o Redox Flow Ba e ies
Te esa Páez,[a,b] Albe o Ma ínez-Cuez a,[c] Jesús Palma,[a] and Edga Ven osa[a,d,e]*
a IMDEA Ene gy, A da. Ramón de la Sag a 3, E-28935 Mós oles, Mad id, Spain
E-mail: [email p o ec ed]
b Depa amen o de ingenie ía Química Indus ial y del Medio Ambien e, Escuela Técnica
Supe io de Ingenie os Indus iales. Uni e sidad Poli écnica de Mad id, C/José Gu ie ez
Abascal, 2, 28006, Mad id, Spain.
c Depa amen o de Química O gánica, Facul ad de Química, Regional Campus o
In e na ional Excellence “Campus Ma e Nos um”, Uni e sidad de Mu cia, E-30100, Mu cia,
Spain
d Depa amen o de Química, Uni e sidad de Bu gos, Pza. Misael Bañuelos s/n, E-09001
Bu gos, Spain
e In e na ional Resea ch Cen e in C i ical Raw Ma e ials-ICCRAM, Uni e si y o Bu gos,
Plaza Misael Bañuelos s/n, E-09001, Bu gos, Spain
2
Abs ac
In he ques o sea ching o new edox- low ba e y chemis ies, cycling s abili y mus be
ca e ully e alua ed since i is one o he mos impo an pa ame e s o new ac i e species.
Howe e , i is challenging o elucida e he in insic s abili y du ing ope a ion o a edox low
ba e y. The symme ical low ba e y cell is a powe ul ool ha helps o unambiguously
de e mine he cycling s abili y. He ein, us wo hiness o his echnique is c i ically e-
e alua ed. Po assium e ocyanide in alkaline media is used as a case s udy since i) i is he bes
pe o ming species o he ca holy e o alkaline low ba e ies in e ms o e e sibili y, solubili y,
cos s and en i onmen al compa ibili y and ii) he cycling s abili y o his species is s ill unde
s ong deba e. Po assium e ocyanide is ound o be s able a pH 14 upon elec ochemical
cycling when he oxygen e olu ion eac ion is p e en ed, which should encou age esea che s
o esume he use o his species. The esul s also e eal ha ca e should be aken when
in e p e ing esul s om his powe ul echnique o a oid misleading conclusions.
TOC GRAPHICS
Keywo ds: edox low ba e ies; elec ochemis y; s abili y; e ocyanide; alkaline media
3
1. In oduc ion
Ene gy s o age echnologies a e conside ed o be a p omising solu ion o deal wi h he challenge
ha he in e mi en na u e o enewable ene gy sou ces ep esen s. Among a ious ene gy
s o age sys ems, edox low ba e ies (RFB) ha e eme ged as a sui able candida e o la ge
applica ions due o hei independen scalabili y o ene gy and powe and hei long cycle li e[1].
To da e, all- anadium edox low ba e y ep esen s he mos de eloped and comme cialized
RFB[2]. Howe e , he use o anadium elec oly e has se e al d awbacks, such as sca ci y and
co osi eness[3], which has igge ed he in e es in inding al e na i e chemis ies o RFB.
Alkaline low ba e ies ha e a ac ed much a en ion as an al e na i e o anadium elec oly e.
All alkaline chemis ies epo ed so a a e based on he use o po assium e ocyanide as ac i e
species in he posi i e compa men in combina ion wi h elec oac i e o ganic compounds in
he nega i e one, including an h aquinone – K4Fe(CN)6[4],[5],[6], i amin B- de i a i es –
K4Fe(CN)6[7], alloxazine – K4Fe(CN)6[8], phenazine-de i a i es – K4Fe(CN)6[9],[10]. Since
po assium e ocyanide is he only ma e ial ound o be ac i e o he posi i e compa men , he
in insic elec ochemical p ope ies o his species has become o high scien i ic in e es .
Cycling s abili y is one o he mos impo an pa ame e s o he ac i e species o edox low
ba e ies. Howe e , i is challenging o elucida e he in insic s abili y du ing ope a ion o a
edox low ba e y. Hyd ogen e olu ion, species c osso e h ough he memb ane o
deg ada ion o he ac i e species used in he opposi e compa men a e o he sou ces o capaci y
ading, which hinde an unambiguous de e mina ion o he in insic cycling s abili y o an ac i e
species. Recen ly, he symme ical low ba e y cell was p oposed o o e come his challenge
and unambiguously e alua e he cycling s abili y o new edox low ba e y chemis ies (Figu e
1a).[11][12][13] In his con igu a ion, he same elec oly e in i s oxidized and educed o m is
used o he posi i e and nega i e compa men . In his way, o he sou ces o capaci y ading
a e a oided. Consequen ly, a symme ical low ba e y cell has been used o in es iga e he
elec ochemical s abili y o e ocyanide a a ious pH [14]. The main conclusion o he s udy
4
was ha e ocyanide is no s able a s ong alkaline media (pH = 14) eleasing ee CN- and
causing a apid capaci y ading (Figu e 1b and 1c). These esul s ha e gene a ed an in ensi e
deba e in he communi y since se e al alkaline low ba e ies based on e ocyanide, e.g. Zn-
Fe low ba e y, ha e shown o deli e long cycle li e.[15], [16]
He ein, we c i ically e-e alua e he symme ical low ba e y cell o s abili y es s o ac i e
species using he in e es ing s udy case o e ocyanide in alkaline media. Two impo an
conclusions a e eached: I) The symme ical low ba e y cell is no a lawless ool o he
e alua ion o s abili y o elec oac i e species o edox low ba e y. Al hough he symme ical
low ba e y cell is a e y use ul ool, cau ion needs o be exe cised when analyzing he esul s.
II) The e ocyanide in s ong alkaline media (pH 14) is no uns able as p e iously p oposed,
which should e-encou age esea ches o esume hei sea ch o new alkaline low ba e y
chemis ies based on i .
Figu e 1. Cycling s abili y o e icyanide/ e ocyanide conduc ed in a symme ical low
ba e y cell a di e en pH condi ions. (a) Schema ic ep esen a ion o a K3Fe(CN)6/ K4Fe(CN)6
symme ical edox low ba e y cell. Capaci y e sus cycling numbe 0.2 M K3Fe(CN)6/
K4Fe(CN)6 symme ical cell (11-13 mL o elec oly es) in (b) 1.0 M KCl solu ion, and (c) 1.0
M KOH solu ion a 40 mA/cm2.[14] Rep oduced wi h pe mission om Else ie .
5
2. Expe imen al P ocedu es
Ma e ials
All chemicals we e pu chased om Sigma Ald ich and Al a Aesa , and used as ecei ed.
P epa a ion o elec oly es
In he symme ic low cell, he ca holy e was p epa ed by dissol ing po assium e ocyanide
(98%, Al a Aesa ) in 1 M KOH (85 %, Sigma Ald ich) o a o d 12 mL o 0.2 M e ocyanide
elec oly e. The anoly e was p epa ed by dissol ing po assium e ocyanide and po assium
e icyanide (99%, Al a Aesa ) in 1 M KOH o a o d ei he 12 mL o 0.2 M e icyanide
elec oly e (balanced cell) o 45 mL o 0.1 M e ocyanide + 0.2 M e icyanide elec oly e
(o e size coun e -compa men ).
Flow ba e y cell
Fil e -p essed low cell desc ibed elsewhe e,[17]using Na ion 212, expanded g aphi e (SGL
Ca bon) and g aphi e el (SGL Ca bon) as he ion selec i e memb ane, cu en collec o and
elec ode, espec i ely, we e used in his s udy. The p ojec ed a ea o he cell was 9 cm2. A
pe is al ic pump (Mas e Flex L/S) was used o p o ide a low a e o 50 mL min-1.
Elec ochemical cha ac e iza ion.
Gal anos a ic cha ge-discha ge measu emen s we e conduc ed using a Biologic VMP
mul ichannel po en ios a . The ba e y was gal anos a ically cycled a ± 20 mAcm-2 wi h
ol age limi s o ± 0.4 V.
3. Resul s and Discussion
3.1. Oxygen e olu ion eac ion in alkaline media.
One o he e iden d awbacks o all aqueous-based ba e ies, including edox low ba e ies, is
he ela i ely na ow he modynamic elec ochemical s abili y window (1.23 V), which is
limi ed by he hyd ogen e olu ion eac ion (HER) and he oxygen e olu ion eac ion (OER).

6
The he modynamic po en ial o OER and HER is s ongly dependen on p o on ac i i y
(Ne ns ian dependency) and, as a consequence, he pH o he elec oly e plays a undamen al
ole in p ocesses associa ed wi h bo h eac ions.
The HER a he nega i e elec ode is kine ically as e han he OER, which in mos ba e ies
make he HER he limi ing side eac ion, [18] e.g. i on-ch omium (E0 o C 2+/C 3+ = −0.41
V)[19] and all- anadium (E0 o V2+/V3+ = −0.26 V).[19] The e o e, he use o a symme ical
cell o he s udy o ca holy es a oids he HER enabling he decon olu ion o chemical s abili y
on he o e all pe o mance o a g ea e ex en . When e ocyanide is used as ca holy e he HER
does no occu in a symme ical cell, which may conclude ha elec oly e decomposi ion does
no play an impo an ole when e alua ing he cycling s abili y o e ocyanide in his cell
con igu a ion. Howe e , he occu ence o he OER canno be easily neglec ed, and should be
conside ed. Scheme 1 illus a es he he modynamic s abili y window o he elec oly e a a pH
8.4 and pH 14 as well as he edox po en ial o e ocyanide a hese wo alues o pH.
Ob iously, he he modynamic window (EOER - EHER) does no change wi h pH, bu he absolu e
po en ials shi owa ds lowe alues as he pH inc eases. Howe e , he edox po en ial o
po assium e ocyanide is no dependen on he pH, bu i is on he ac i i y o po assium ca ions.
I 1 M KCl is used in neu al pH, o main aining a good ionic conduc i i y, he edox po en ial
o he e ocyanide will no a y signi ican ly (EFe o/Fe i (1 M KOH) ≈ EFe o/Fe i (1 M KCl) ≈
0.42 V s SHE). As a esul , he edox po en ial o e ocyanide a pH=8.4 will be wi hin he
s abili y window o he elec oly e (330 mV below he po en ial o he OER), while he po en ial
o bo h eac ions ( e ocyanide and he OER) will o e lap in alkaline media. Thus, i is sa e o
neglec he occu ence o he OER du ing elec ochemical cycling o e ocyanide a neu al
pH, bu i canno be uled ou in alkaline media.
7
Scheme 1. Schema ic illus a ion o he pH dependency o he s abili y window in aqueous
media wi h espec o he pH independen beha iou o po en ial o e ocyanide o (a) neu al
media, b) alkaline media. In pH 8.4 media, po en ial o he couple K3Fe(CN)6/ K4Fe(CN)6 will
be wi hin he elec ochemical s abili y window, while in pH 14 media, i will be on he bounda y
o he OER.
3.2. Capaci y unbalancing o a ba e y igge ed by side eac ion.
I e e sible side eac ions such as he OER lead o speci ic capaci y unbalance be ween posi i e
and nega i e elec ode. Scheme 2 illus a es he gene al mechanism o capaci y unbalancing by
he OER du ing a cha ge/discha ge cycle. The posi i e elec ode su e ing a side eac ion will
no be ully cha ged (99 %) as some elec ons (1 %) a e consumed by he pa asi ic eac ion,
while he nega i e elec ode will be able o be ully cha ged (100 %). Du ing he discha ge
p ocess, he posi i e elec ode will be he limi ing elec ode since i was no ully cha ged. As
a esul , he discha ge capaci y o he ba e y will be limi ed o 99 % o i s capaci y. Mos
impo an ly, in he subsequen cha ge p ocess, he nega i e elec ode will no s a om a ully
discha ge s a e (100 % - 99 % = 1 %). Consequen ly, he ba e y will no be able o be ully
cha ged in he subsequen cha ge p ocess since only 99 % o he species in he nega i e
8
elec ode a e in i s discha ge s a e and he ba e y will ha e los his 1 % o i s cha ge s o age
capaci y. I a side eac ion is occu ing du ing cha ge/discha ge o he species o in e es , he
capaci y unbalancing will be accele a ed when using an alkaline e ocyanide - e icyanide
symme ical low. The accele a ed ading is due o he ac ha he side eac ion no only will
occu in one elec ode, bu i will occu in bo h elec odes (in one elec ode du ing cha ge and
in he ano he one du ing discha ge). As a esul , he capaci y ading will be accele a ed by a
ac o o 2.
In he case o e ocyanide, i was discussed abo e ha he OER will likely no occu in neu al
pH. The ques ion is whe he i akes place a pH 14. The modynamically, he OER migh occu
bu i is kine ically e y sluggish. Howe e , he OER migh be p omo ed by se e al ac o s (o
combina ion o hem), e.g. p esence o elec oca alys s in he elec ode ha imp o es he
kine ics o he OER, small cu en densi ies ha p omo e he kine ically slow eac ion o la ge
ope a ing o e po en ials de i ed om a high in e nal esis ance (memb ane, el s, low ac o ,
e c) ha inc eases he d i ing o ce o he OER.
Scheme 2. Schema ic illus a ion o he concep o ba e y unbalance by he OER in aqueous
low ba e ies du ing a cha ge-discha ge cycle.
3.3. Cycling s abili y o a e ocyanide elec oly e a pH 14 in a symme ical low cell.
9
A symme ical low cell con aining a balanced amoun o e o- and e icyanide was used o
an easy e alua ion o e ocyanide cycling s abili y a pH 14. Thus, 0.2 M e ocyanide and 0.2
M e icyanide dissol ed in a 1 M KOH aqueous solu ion a e used as elec oac i e species o
he posi i e and nega i e compa men , espec i ely. A shadow capaci y ading was obse ed
du ing he 300 cycles and 90 hou s o he es (Figu e 2), e aining 87 % o i s ini ial cha ge
capaci y. Al hough he capaci y e en ion is no ou s anding, his alue is d as ically be e han
he capaci y e en ion p e iously epo ed in alkaline media using symme ical cell[14], in
which he e en ion d opped below 50 % a e only 100 cycles. This disc epancy clea ly shows
ha he symme ical low cell is no lawless and he e mus be some o he ac o s ha a e no
been conside ed. As abo e-men ioned, ba e y unbalancing due o he OER could be
esponsible o capaci y ading. Al hough bo h expe imen s, he p e iously epo ed and he
one epo ed he e, we e ca ied ou a he same condi ions (cu en densi y, low a e), he e a e
ac o s which a e e y di icul o con ol. We no ice ha he o e po en ials in ou expe imen s
a e signi ican ly lowe han ha o he p e ious epo , which lead us o p opose he OER as
he sou ce o he disc epancies. To co obo a e his hypo hesis, an iden ical symme ical low
cell was cycled unde a o able condi ions o he OER, i.e. he g aphi e el s we e in en ionally
con amina ed wi h an OER elec oca alys (Ni(OH)2 pa icles)[[20], [21], [22]] and was cycled
a lowe cu en densi y (20 mA cm-2 ins ead o 40 mA cm-2) o p omo e he kine ically slowe
p ocess ( he OER), as i occu s in all- anadium low ba e ies wi h he HER. Figu e 3 shows
he i s 200 cycles o an ini ially balanced symme ical low cell (64 mAh o e e sible cha ge
o each compa men ). The capaci y e en ion a e 200 cycles was 78 %, which con i ms he
accele a ed ading unde a o able condi ions o he OER. Thus, side eac ions canno be ule
ou when e alua ing he cycling s abili y using a symme ical low cell. In he pa icula case
o e ocyanide, he s ong capaci y ading p e iously epo ed appea s o be ela ed o he OER
occu ing unde hose expe imen al condi ions, and i does no ep esen a gene al beha io in
alkaline media. Howe e , ou expe imen can nei he con i m no ule ou ha e ocyanide is
16
h ps://doi.o g/10.1126/science.aab3033.
[5] D.G. Kwabi, K. Lin, Y. Ji, A. Aspu u-Guzik, R.G. Go don, M.J. Aziz, E.F. Ke ,
M.-A. Goule , D. De Po cellinis, D.P. Tabo , D.A. Pollack, Alkaline Quinone
Flow Ba e y wi h Long Li e ime a pH 12, Joule. 2 (2018) 1894–1906.
h ps://doi.o g/10.1016/j.joule.2018.07.005.
[6] Y. Ji, M. Goule , D.A. Pollack, D.G. Kwabi, S. Jin, D. Po cellinis, E.F. Ke ,
R.G. Go don, M.J. Aziz, A Phosphona e‐Func ionalized Quinone Redox Flow
Ba e y a Nea ‐Neu al pH wi h Reco d Capaci y Re en ion Ra e, Ad . Ene gy
Ma e . 9 (2019) 1900039. h ps://doi.o g/10.1002/aenm.201900039.
[7] A. O i a, M.G. Ve de, M. Sakai, Y.S. Meng, A biomime ic edox low ba e y
based on la in mononucleo ide, Na . Commun. 7 (2016) 13230.
h ps://doi.o g/10.1038/ncomms13230.
[8] K. Lin, R. Gómez-Bomba elli, E.S. Beh, L. Tong, Q. Chen, A. Valle, A.
Aspu u-Guzik, M.J. Aziz, R.G. Go don, A edox- low ba e y wi h an
alloxazine-based o ganic elec oly e, Na . Ene gy. 1 (2016) 1–8.
h ps://doi.o g/10.1038/nene gy.2016.102.
[9] A. Hollas, X. Wei, V. Mu ugesan, Z. Nie, B. Li, D. Reed, J. Liu, V. Sp enkle,
W. Wang, A biomime ic high-capaci y phenazine-based anoly e o aqueous
o ganic edox low ba e ies, Na . Ene gy. 3 (2018) 508–514.
h ps://doi.o g/10.1038/s41560-018-0167-3.
[10] C. Wang, X. Li, B. Yu, Y. Wang, Z. Yang, H. Wang, H. Lin, J. Ma, G. Li, Z.
Jin, Molecula Design o Fused-Ring Phenazine De i a i es o Long-Cycling
Alkaline Redox Flow Ba e ies, ACS Ene gy Le . (2020) 411–417.
h ps://doi.o g/10.1021/acsene gyle .9b02676.

17
[11] M.-A. Goule , M.J. Aziz, Flow Ba e y Molecula Reac an S abili y De e mined
by Symme ic Cell Cycling Me hods, J. Elec ochem. Soc. 165 (2018) A1466–
A1477. h ps://doi.o g/10.1149/2.0891807jes.
[12] J.D. Milsh ein, A.P. Kau , M.D. Casselman, J.A. Kowalski, S. Modek u i, P.L.
Zhang, N. Ha sha A anayake, C.F. Ellio , S.R. Pa kin, C. Risko, F.R. B ushe ,
S.A. Odom, High cu en densi y, long du a ion cycling o soluble o ganic ac i e
species o non-aqueous edox low ba e ies, Ene gy En i on. Sci. 9 (2016)
3531–3543. h ps://doi.o g/10.1039/c6ee02027e.
[13] J.D. Milsh ein, J.L. Ba on, R.M. Da ling, F.R. B ushe , 4-ace amido-2,2,6,6-
e ame hylpipe idine-1-oxyl as a model o ganic edox ac i e compound o
nonaqueous low ba e ies, J. Powe Sou ces. 327 (2016) 151–159.
h ps://doi.o g/10.1016/j.jpowsou .2016.06.125.
[14] J. Luo, A. Sam, B. Hu, C. DeB ule , X. Wei, W. Wang, T.L. Liu, Un a eling pH
dependen cycling s abili y o e icyanide/ e ocyanide in edox low ba e ies,
Nano Ene gy. 42 (2017) 215–221.
h ps://doi.o g/10.1016/J.NANOEN.2017.10.057.
[15] Z. Yuan, Y. Duan, T. Liu, H. Zhang, X. Li, Towa d a Low-Cos Alkaline Zinc-
I on Flow Ba e y wi h a Polybenzimidazole Cus om Memb ane o S a iona y
Ene gy S o age, IScience. 3 (2018) 40–49.
h ps://doi.o g/10.1016/j.isci.2018.04.006.
[16] T. Páez, A. Ma ínez-Cuez a, J. Palma, E. Ven osa, Media ed alkaline low
ba e ies: F om undamen als o applica ion, ACS Appl. Ene gy Ma e . 2 (2019)
8328–8336. h ps://doi.o g/10.1021/acsaem.9b01826.
[17] E. Ven osa, M. Skoumal, F.J. Vázquez, C. Flox, J.R. Mo an e, Ope ando s udies
18
o all- anadium low ba e ies: Easy- o-make e e ence elec ode based on
sil e -sil e sul a e, J. Powe Sou ces. 271 (2014) 556–560.
h ps://doi.o g/10.1016/j.jpowsou .2014.08.029.
[18] R. Schweiss, A. P i zl, C. Meise , Pa asi ic hyd ogen e olu ion a di e en
ca bon ibe elec odes in anadium edox low ba e ies, J. Elec ochem. Soc.
163 (2016) A2089–A2094. h ps://doi.o g/10.1149/2.1281609jes.
[19] POURBAIX, M., A las o Elec ochemical Equilib ia in Aqueous Solu ions,
NACA. (1984).
[20] R.D.L. Smi h, M.S. P é o , R.D. Fagan, Z. Zhang, P.A. Sedach, M.K.J. Siu, S.
T udel, C.P. Be lingue e, Pho ochemical ou e o accessing amo phous me al
oxide ma e ials o wa e oxida ion ca alysis, Science. 340 (2013) 60–63.
h ps://doi.o g/10.1126/science.1233638.
[21] C.C.L. McC o y, S. Jung, J.C. Pe e s, T.F. Ja amillo, Benchma king
He e ogeneous Elec oca alys s o he Oxygen E olu ion Reac ion, J. Am.
Chem. Soc. 135 (2013) 16977–16987. h ps://doi.o g/10.1021/ja407115p.
[22] M. Gao, W. Sheng, Z. Zhuang, Q. Fang, S. Gu, J. Jiang, Y. Yan, E icien Wa e
Oxida ion Using Nanos uc u ed α-Nickel-Hyd oxide as an Elec oca alys , J.
Am. Chem. Soc. 136 (2014) 7077–7084. h ps://doi.o g/10.1021/ja502128j.
[23] U. Himmel eich, P.W. Kuchel, 13C-NMR s udies o ansmemb ane elec on
ans e o ex acellula e icyanide in human e y h ocy es, Eu . J. Biochem.
246 (1997) 638–645. h ps://doi.o g/10.1111/j.1432-1033.1997. 01-1-00638.x.