Tải bản đầy đủ - 0 (trang)
6 Polymer Electrolyte Membrane Fuel Cell

6 Polymer Electrolyte Membrane Fuel Cell

Tải bản đầy đủ - 0trang

155



Fuel Cells

4H+ + 4e– + O2 → 2H2O

Cathode flow channel



Air (O2)



Cathode GDL/MPL

Cathode electrocatalyst layer

MEA



H+



H2O



e–



Ionomer membrane

Anode electrocatalyst layer



Load



Anode GDL

Anode flow channel



H2



H2 → 2H+ + 2e–



FIGURE 6.9  Schematic of a PEM fuel cell. GDL, gas diffusion layer; MPL, microporous

layer; MEA, membrane electrode assembly. (Reprinted with permission from Borup, R. et al.

2007. Scientific aspects of polymer electrolyte fuel cell durability and degradation. Chem.

Rev. 107 (10):3904–3951. Copyright 2007, American Chemical Society.)



hydrogen initially and the carbon cost in making the fuel cell materials). However,

not only are technological challenges associated with the generation of the necessarily huge volume of hydrogen fuel, the development of a sufficiently robust PEMFC

remains a substantial challenge.

What are the constraints faced by polymer chemists as they design a polymer for

use as an electrolyte membrane in a fuel cell?

• Perhaps foremost, the polymer membrane must be mechanically stable,

having a reasonably high Tg for operation at relatively high temperatures

(up to 120°C; U.S. Department of Energy 2011a).

• Water is a critical component in the successful operation of a PEMFC so

the polymer must perform well and be durable under conditions of high

humidity.

• In addition to thermal/mechanical stability, the polymer must be robust

with respect to its chemical reactivity.

• As with any electrolyte, it must act as an insulator to electrons.

• It must also be impervious to the fuel, preventing its crossover through the

electrolyte membrane to the cathode which would lower the VOC.

Because PEMs are used as the electrolyte in both the hydrogen/oxygen fuel cell

and the DMFC, much polymer development has been motivated by trying to minimize the methanol crossover problem: diffusion of methanol through the PEM to the

cathode. Crossover amounts of up to 10% can occur leading to a significant decrease



156



Chemistry of Sustainable Energy



in the open circuit voltage and requiring increased catalyst loadings. The thinner the

PEM, the lower the ohmic losses, but the membrane then becomes more susceptible

to crossover and mechanical failure. Finally, the electrocatalysts must be engineered

into a PEM membrane electrode assembly (MEA, discussed below). Of course, it is

also desirable that the PEM should be inexpensive, easy to fabricate, and synthesized

in a sustainable manner! These challenges have led to (and continue to lead to) a

multitude of approaches in the development of a better PEM; we describe some of

the more important ones below.



6.6.2  General Considerations

6.6.2.1  Membrane Electrode Assembly

The heart of a PEMFC is the MEA consisting of the electrocatalysts and the electrolyte membrane (the “ionomer membrane” core shown in Figure 6.9, where an ionomer is simply an ion-conducting polymer). The design and improvement of MEAs is

very much an empirical undertaking, and both chemical and mechanical variables

play large roles in its effectiveness. Because the PEMFC may use gases for both the

fuel and the oxidant (H2 and O2 or air), the electrodes typically consist of a porous

carbon gas diffusion layer to which the catalyst layer is bonded. Fabrication of the

MEA takes place either by bonding the catalyst layer (ca. 10 µm thick) to the ionomer membrane first, or by bonding it to the carbon electrode layer first. The ionomer

membrane itself is also quite thin—usually around 10–175 µm. In either case, good

bonding is essential so that the proper conductivity (and minimal loss due to ohmic

resistance) can be achieved. We will look at MEAs in more detail in the context of

the DMFC in Section 6.6.4.

6.6.2.2  Water Management

Water management is an important issue in PEMFCs and both the physical and

chemical properties of the polymer play important roles. Moisture is necessary to

provide the level of proton conductivity needed to sustain the electrochemical reaction. A dry membrane has low conductivity and increased resistance leading to a

decreased current. Evaluation of PEM performance, therefore, must somehow reflect

the conductivity, reported as s in units of S/cm at specific conditions of temperature

and relative humidity (RH). Exposure of the membrane to a sufficient amount of

moisture allows for the development of networks of water (clusters, pools, channels,

etc.) that provide a conduit for the conduction of protons by diffusion and protonhopping, hence (as usual) the morphology of the polymer is a crucial variable in

its performance as an ion conductor (Jorn et al. 2012). Dehydration can occur via

electro-osmotic drag, where water molecules from the anode are literally dragged

to the cathode by the proton current (Cheah et  al. 2011). On the other hand, too

much water “drowns” the electrodes, flooding the gas flow channels and interfering

with mass transport. A hydrophobic agent such as PTFE (polytetrafluoroethylene,

aka Teflon®, −(CF2)n−) is added to prevent the pores from being choked with water

(Wang et al. 2011).



157



Fuel Cells



6.6.3  Polymer Development

Given the demands upon the polymer electrolyte, what kind of materials have

been developed that can work effectively in a fuel cell? The ionomers used as

electrolytes possess highly polar functionality that allow for the required proton

conductivity through the channels of the membrane. Because the role of the ionomer in a fuel cell is to facilitate proton conductivity, these membranes consist of

a polymer backbone with acidic functionality (typically sulfonic or phosphonic

acids) scattered throughout. The backbone may vary from perfluorinated, partially fluorinated, or without fluorine; completely aliphatic to primarily aromatic;

primarily made up of carbon and hydrogen or a polymer interspersed with a high

proportion of heteroatoms. A special concern in the synthesis and fabrication of

PEM ionomers is their unusually high polarity, making them particularly slow

to dissolve in relatively nonpolar organic solvents. In any case, the development

of a relatively inexpensive, easily fabricated ionomer with a high proton conductivity, low electron conductivity, and good resistance to degradation is a challenge. There is a tremendous variety of ionomers that have been investigated for

use in PEMFCs but we will focus on four classes: perfluorosulfonic acids, sulfonated poly(arylene) ethers/ketones/sulfones, polyimides, and polybenzimidazoles

(PBI).

6.6.3.1  Perfluorosulfonic Acid Membranes

If platinum on carbon is the standard for electrocatalyst composition in fuel cells,

then Nafion® (DuPont Co.) can be considered the standard for the ionomer membrane material of PEMFCs. Nafion membranes are a series of perfluorosulfonic acid/

PTFE copolymer membranes of different thicknesses and strengths (Figure 6.10).

The common acronym for these membranes is PFSA or PFSI (for perfluorinated

sulfonic acid or ionomer). The poly-CF2 backbone presents excellent chemical stability and the sulfonic acid groups allow for good proton conductivity when hydrated

(up to 0.10 S/cm; Li et al. 2003). PFSA membranes also offer low O2 (g) and H2 (g)

permeability but, unfortunately, significant permeability for methanol, making them



F2

C





C

F2



F

C



F2

C



y



x



CF3

O



O

F2

C



CF

C

F2



FIGURE 6.10  The Nafion copolymer structure.



O

z



O

S



C

F2



OH



158



Chemistry of Sustainable Energy



poor candidates for the DMFC. In addition, not only is Nafion expensive, it also

has a relatively low Tg and its operating temperature range is quite limited: PFSA

membranes are effectively unusable below 0°C, at which point the water necessary

for proton conductance is frozen, or above 100°C, where the stability of Nafion is

poor (Nafion’s ordinary operating temperature is ≈80°C). Nafion also presents serious problems in its environmental impact as it is not biodegradable and incineration

generates SO2, HF, and CO2 (DuPont 2009). Thus, while they are widely used PEMs,

much research effort has focused on modifying, or finding a replacement for, PFSA

membranes.

Researchers have attempted to modify the Nafion membrane by impregnating

inorganic materials (e.g., SiO2, TiO2, or zirconium phosphates) or functionalized fillers (carbon nanotubes, montmorillonite clays, or zeolites) into the polymer, primarily

to improve thermal stability (Li et al. 2003; Zhang and Shen 2012). Doping of Nafion

with an ionic liquid (triethylammonium trifluorosulfonate (CH 3CH 2 )3 NH + CF3SO3− )

was shown to match the proton conductivity of regular Nafion under anhydrous

conditions, but its addition acted as a plasticizer, preventing its use at elevated

temperatures (Sood et al. 2012). Attempts to modify the PFSA membrane by radiation-grafting (exposing the fluoropolymer to UV or gamma-ray radiation to prepare

a reactive radical surface on the film, upon which a new monomer (e.g., styrene) can

be grafted and further functionalized) have presented greatly mixed results in terms

of durability and conductivity. Proton conductivities of up to 0.25 S/cm have been

reported for these modified PFSA membranes, but with substantial accompanying

degradation (Zhang and Shen 2012).

6.6.3.2  Poly(Arylene Ether) Membranes

Another prominent class of PEM membranes is based on sulfonated poly(arylene

ether) ionomers of which there are several varieties: poly(arylene ether) ketones,

poly(arylene ether) sulfones, poly(arylene ether)s with heterocyclic functionality, and

so on. Poly(arylene ethers) offer more flexibility in terms of polymer synthesis and

modification, are less expensive than PFSA membranes, and have shown good performance overall with proton conductivities that match or surpass PFSA membranes

under similar conditions. One particularly well-known example is the semicrystalline sulfonated poly(ether)ether ketone SPEEK (Figure 6.11). SPEEK membranes,

however, are prone to hydrolysis or radical-induced degradation and can become

brittle at high temperatures. As with other ionomer classes, addressing the issue of



O





C



O



SO3H



FIGURE 6.11  The sulfonated poly(ether)ether ketone repeating unit.







O



n



159



Fuel Cells

O





O





C











C





n



n

H

CH2





OH





n











C



H-atom abstraction

n

CH2





OH









C

CH2







n





n



n

Cross-linked polymer



FIGURE 6.12  Radical-induced cross-linking of a SPEEK membrane.



methanol crossover is a high priority. Some modified poly(arylene ether) PEMs that

have been prepared to address these concerns are discussed below.

Cross-linking of SPEEK membranes greatly increases the durability of the membrane but can have a detrimental impact on the proton conductivity. As shown in the

light-induced radical coupling of Figure 6.12, cross-linking can take place between

the benzophenone moiety and suitable substituent groups. As a result, this polymer

showed improved selectivity for proton conduction relative to methanol permeability (Ishikawa et al. 2007). The synthesis of a cross-linked organic/inorganic hybrid

poly(ether) ether membrane is shown in Figure 6.13. In this example, sulfonation of

the PEEK precursor (1) provides for the development of a sulfonate-bridged SPEEK

network (2). This network can then undergo ortholithiation with n-butyllithium followed by treatment with tetrachlorosilane and hydrolysis to give the silated/sulfonated ionomer product. This material demonstrated greatly improved mechanical

properties (e.g., an elastic modulus of 260 ± 40 MPa, compared to 25 for Nafion),

attributed to the silanol functionality and extra cross-linking. However, the enhanced

mechanical rigidity lowered the proton conductivity relative to Nafion (DiVona et al.

2006). In another approach to a modified poly(ether)sulfone membrane, a bisazide

was used as a coupling agent with a vinyl-appended polyethersulfone (PES) as

shown in Figure 6.14. The resultant ionomer gave well-enhanced proton conductivity of 0.79 S/cm at 100°C, compared to 0.48 S/cm for Nafion 112 under identical

conditions (Oh et al. 2008).



O



1. BuLi (–60°C), THF

2. SiCl4, NaHCO3



1



O







O



(HO)3Si



n







SO3



HO



S



O



O



Cl



FIGURE 6.13  A silated, cross-linked SPEEK derivative.







O



O







O



O



Si(OH)3



O



SO3Cl



x



O







O



SO3Cl



O



O



O

S



O



Si(OH)3



O



O



O



SO3



O



2



x



O



O



O



O



O



Si(OH)3



S



O



y











O



O



SO3Cl



O



O



y











160

Chemistry of Sustainable Energy



N3



O



PES



O



S



O



O



O



N



HO3S



O

x



N3



O



O

N



O



PES = poly(ether) sulfone



2. Heat



1. Film casting



O



S



O



FIGURE 6.14  Poly(ether)sulfone coupling to an aromatic bisazide for enhanced mechanical stability.



O



O



PES



y



O



Fuel Cells

161



162



Chemistry of Sustainable Energy



While promising inroads have been made with respect to methanol crossover,

the issue of PEM degradation by hydrolysis remains. Researchers have found that

modification of the arene units by the incorporation of additional bulky substituents

reduces the tendency toward hydrolysis. The poly(arylene ether sulfone) polymer

electrolyte membrane shown in Figure 6.15 gave results comparable to Nafion in

terms of both conductivity and durability (Miyatake et al. 2007). Stability to hydrolysis is even more improved when the polymer backbone includes heterocycles such

as the oxazole- or triazole-containing copolymers shown in Figure 6.16; compound

b also showed an increased proton conductivity relative to SPEEK at the same level

of sulfonation (Li and Yu 2007; Ponce et al. 2008).

6.6.3.3  Polyimides and Imidazoles

Sufonated polyimides (SPI). Imides are not particularly resistant to hydrolysis, but

SPI offer high mechanical strength, minimal crossover, and excellent thermal and

chemical stability. Naphthalenic polyimides (Figure 6.17a) have been prepared and

tested in order to address the hydrolysis issue. Ar1 and Ar2 represent any of a number

of sulfonated oligomers, primarily of the arene ether variety. Binaphthyl SPIs (Figure

6.17b) appear to be even more resistant to hydrolysis and oxidation. For example, the

naphthyl SPI (A, Figure 6.18) demonstrated a hydrolytic stability of 34 h at 90°C and

the binaphthyl PEM B was stable for more than 1000 h at this temperature (Zhang

and Shen 2012).

Polybenzimidazoles. Beyond concerns of methanol crossover and hydrolysis lies

the issue of higher-temperature operation to take advantage of improved kinetics and

reduced propensity to poisoning. PBI (Figure 6.19) are particularly attractive PEM candidates for higher-temperature PEMFC operation (≈150–200°C) due to their enhanced

chemical and thermal stability. These membranes are doped with phosphoric acid to

make the proton-conducting acid–base complex (Figure 6.19). Both meta- and paraPBI have been prepared, with the para isomer exhibiting better mechanical properties

that persist even at a high level of doping with H3PO4 (Zhang and Shen 2012).

Like the SPI and PFSA membranes, much research has been devoted to modifying

the basic PBI skeleton to improve the ionomer characteristics. Additionally, research

has been done examining blends of PBI with other polymers. An SPI blended with

a PBI membrane and doped with phosphoric acid (Figure 6.20) gave a proton conductivity of about 0.5 S/cm (120°C/45% RH), with no sign of decreased performance

even after 1000 h of operation at 120°C and 0% RH (Suzuki et  al. 2012). A 50:50

blend of PBI and poly(vinyl-1,2,4-triazole) doped with phosphoric acid (Figure 6.21)

showed moderate conductivity at 120°C and no humidification (≈0.09 S/cm) compared

to the conductivity of neat PBI under the same conditions of roughly 0.02 S/cm. The

researchers postulated that the reason for the increased conductivity relative to PBI was

due to the incorporation of more phosphoric acid in the blend, as well as to the fact that

the blend includes both the triazole and imidazole rings. This material also exhibited

excellent thermal stability (Hazarika and Jana 2012).

6.6.3.4  Metal–Organic Frameworks

In Chapter 5, we learned about how metal–organic frameworks are well suited for

potential use as adsorbents for H2 storage. Not only does the porosity of an MOF lend



SO3H



y



S



O



O



FIGURE 6.15  A hydrolysis-resistant PEM.



O



O



CH3



C



CH3



0.5n



O

SO3H

y



O



S



O

O



2x



HO3S



0.5n



SO3H



x



Fuel Cells

163



N



O



N



O



O



N



F3C



x



C



N



CF3



O



N



O



N



HO3S



O



O



y



N

N



CF3



C



CF3



N



O



O



HO3S



1–n



z



C



O



N

N



N

O



O

n



n

Sporadic sulfonation at dashed line



CF3



C



SO3H

CF3



FIGURE 6.16  Polymers with improved hydrolytic resistance. (a) An oxazole-containing copolymer, (b) a triazole-containing copolymer.



(b)



(a)



164

Chemistry of Sustainable Energy



165



Fuel Cells

(a)







O



N



O



N



O



Ar1



O



x



O



O



N



N



O



O



Ar2







1–x



(b)

O







N



O



O



N



O



Ar1



x



O



O



N



N



O



O



Ar2







1–x



FIGURE 6.17  (a) Naphthalenic SPI. (b) Binaphthyl SPI.



itself to incorporation of H2, but also the well-defined channels within MOFs make

them excellent candidates for proton conductivity and use as a membrane electrolyte

in a fuel cell. Three examples follow.

One approach to taking advantage of MOFs in PEMFCs is to chemically incorporate MOFs in the polymer membrane. For example, a sulfonated poly(2,6-dimethyl1,4-phenylene oxide) membrane was coupled with an iron-aminoterephthalate MOF

(Fe-MIL-101-NH2) to give a “mixed matrix membrane” as depicted in Figure 6.22.

This mixed matrix membrane showed a proton conductivity of 0.10 S/cm at 6% MOF

loading, a value far superior to that of either the MOF-free polyphenylene oxide

polymer or Nafion 117 under the same conditions. A maximum value of 0.25 S/cm

was obtained at about 90°C. The proton conduction was attributed to increased acidity of the water molecule by coordination to an Fe(III) cation (Wu et al. 2013).

While integrating MOFs into the polymer electrolyte is one approach to taking

advantage of the special properties they provide, there are advantages to using the

MOF alone as the electrolyte membrane. For example, use of a metal–organic framework as the electrolyte membrane in a fuel cell should allow for an increase in operating temperature beyond that seen for typical organic polymers. Furthermore, MOFs

are not necessarily limited to using water as the proton carrier (use of water limits

the operating temperature of the PEM fuel cell). To that end, researchers examined

the anhydrous proton-conducting capabilities of β-PCMOF2, the trisodium salt of

2,4,6-trihydroxy-1,3,5-benzenetrisulfonate. By entrapping 1H-1,2,4-triazole (Figure

6.23) in the channels of β-PCMOF2 to act as an organic proton carrier, researchers

were able to demonstrate anhydrous proton conductivity of 5 × 10 –4 S/cm at 150°C

(Hurd et  al. 2009). Incorporation of an MEA fabricated out of this β-PCMOF2triazolium conglomerate and using it in an H2/air fuel cell gave a Voc value of 1.18 V

at 100°C. The Voc decreased, unfortunately, with increasing temperature, presumably

due to fuel crossover. Thus, while this performance is still well below that shown by

a hydrated Nafion membrane, this approach is an attractive alternative to hydrated

fuel cells and their limitations.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

6 Polymer Electrolyte Membrane Fuel Cell

Tải bản đầy đủ ngay(0 tr)

×