In Partial Fulfillment of the Requirements For the Degree of

ELECTRODE STRUCTURE. MEASUREMENT PERFORMED IN VACUUM AT 25°C... 92 FIGURE 2.18.M” CALCULATED AND ANALYZED SPECTRA WITH MATHEMATICS. REVIEW A) IS A MODEL WITH EQUIVALENT LEADS. DIETHYLPHOSPHATOETHYL GROUPS CUT OFF (P- BEA), AND E. 131 FIGURE 4.3.29SI CPMASNMR SPECTRA OF I) PURE-SILICA SAMPLES EXTRACTED BY ACETIC ACID AS.

INTRODUCTION

M OTIVATION

The research efforts highlighted in this thesis were conducted to address some of the aforementioned goals. First, proton-conducting zeolites were investigated with the hope of improving the operating performance of DMFCs and PEMFCs by addressing some of the engineering challenges associated with the separation membrane between the anode and cathode, particularly while maintaining high proton conductivity in elevated temperatures that normally cause dehydration. and decreased proton conductivity, and reduced methanol crossover.

A B RIEF I NTRODUCTION TO F UEL C ELLS AND B ATTERIES

Fuel Cells Overview
Nafion for Fuel Cells
Batteries

Two of the major challenges in operating fuel cells include methanol crossover in the case of DMFCs and reduced proton conductivity caused by separator membrane hydration when operating at elevated temperatures (especially in PEMFCs where catalysts are sensitive to CO impurities in the H2 feed stream when used at low temperatures). Because of their small pores, zeolites could reject methanol at the anode, reducing the rate of chemical diffusion to the cathode.

Figure 1.1. Schematic of hydrogen fuel cell.

Z EOLITES AND Z EOLITE - LIKE M ATERIALS

What are Zeolites and Zeolite-like Materials?
Synthesis Considerations
The Zincosilicates
Organically Functionalized Molecular Sieves (OFMSs)

Cations and hydroxide ions are often introduced in the form of strong bases such as sodium hydroxide. A family of zeolite-like materials discovered in the last decade is that of zincosilicates.

Table 1.1. Structural properties of known zincosilicates.

I ON AND P ROTON C ONDUCTION IN Z EOLITES AND Z EOLITE - LIKE M ATERIALS

Proton Conduction in Acid Functionalized Micro- and Mesoporous Materials
Micro- and Mesoporous Materials in Proton Conducting Membranes
Ion Conduction in Microporous Materials

Li < K, while the activation energy showed a minimum for the sodium-exchanged sample in the material with Si/Al = 1 (Li > Na < K ~ Rb) and a minimum for the potassium-exchanged sample in the material with Si/Al = 5 ( Li > Na > K < Rb). Thus, dielectric loss peaks were attributed to relaxation of the sodium ion in the "side pockets".

R EFERENCES

INSTRUMENTATION

I NTRODUCTION TO I MPEDANCE S PECTROSCOPY

Dielectric Relaxation
Conduction

This is the same equation we arrived at from the circuit model in Figure 2.1 when , , and , so this circuit model is known. A single relaxation is the simplest possible case and in general there is a distribution of relaxation times in solids and the governing equations become more complex. Conductivity in the absence of a concentration gradient, i.e. the movement of charge carriers driven solely by an applied electric field, can be described as 2.6) where drift mobility, C is the concentration of mobile ions, with carrier charge.

Table 2.1. Impedance functions and their transforms.

A PPARATUS FOR M EASURING P ROTON C ONDUCTIVITY

While this apparatus was designed to meet the current and future needs of the research effort, the contact between the lower platinum plate and the small spring-loaded probe was unreliable, and the upper platinum plate did not stay connected to the plastic rod used for application. pressure on the sample. Further, upon heating above room temperature, complete hydration of the sample was lost and the hydration level was unknown. Powdered samples were loaded into the sample holder on top of the lower platinum cylinder and gently compacted.

Figure 2.7. Apparatus constructed for variable temperature measurement of

A PPARATUS FOR V ARIABLE T EMPERATURE /H UMIDITY M EASUREMENTS

Electrical contact to each of the 32 electrodes was achieved by a high-temperature compatible coaxial cable that ran from inside the chamber through a foam plug to the external laboratory environment. A Solartron 1260 was then manually connected to two of the coaxial cables, uniquely addressing an individual sample for measurement. While this arrangement allowed near-simultaneous characterization of 16 samples in a temperature- and humidity-controlled environment, at elevated relative humidity and temperature, the samples became mechanically unstable, and exfoliation of the gold contacts and loss of structural integrity were observed.

Figure 2.9. Sample holder for pelletized sample used in temperature/humidity chamber

A PPARATUS FOR M EASURING I ON C ONDUCTIVITY

Electrical connection to the lower contact is made by the indicated Pt wire attached by welding. A Pt wire running through the inside of an alumina tube wire welded to the Pt rod is used to contact the top of the sample. A slide bush (not visible) charges the visible spring to apply gentle force to the sample, through the Pt rod used to make electrical contact with the sample.

Figure 2.11. Vacuum chamber used for dehydration and characterization of pelletized samples

M EASUREMENT WITH I NTERDIGITATED E LECTRODE S TRUCTURES

S YSTEM Q UALIFICATION AND A NALYSIS

Finally, Figure 2.17 shows M” data for Na-X powder drop casting from an ethanol suspension onto an interdigital electrode (IE) structure (Novocontrol). This is shown in Figure 2.18, along with the circuit model used to generate the appropriate equations (by taking the Laplace transform of the appropriate transfer function). Recall that the M" spectra for a printed pellet with sputtered contacts look like Figure 2.18b), while spectra of Na-X powder held between two Pt plates, Na-X powder droplet cast on IE electrodes and printed Na-X grains without sputtered electrodes are more similar to Figure 2.18a).

Figure 2.15. Imaginary part of the modulus as a function of frequency for a hydraulically pressed pellet of Na-X with sputtered gold contacts

R EFERENCES

PROTON CONDUCTIVITY IN SULFONIC ACID-FUNCTIONALIZED ZEOLITE

A BSTRACT

I NTRODUCTION

One of the aims of this study is to test the hypothesis of cooperativity between sulfonic acid sites and framework acid sites created by the aluminum framework by molecular sieve synthesis with a *BEA framework that contained both sulfonic acid groups and different concentrations of aluminum framework sites by either direct synthesis , as in the case of nanocrystalline beta, or by post-synthetic insertion of aluminum into the hydroxyl nests formed after the removal of framework zinc in CIT-6 (a zinc silicate analog of zeolite beta).25 As shown below, the framework aluminum is not maintained in the formation of sulfonic acid sites . Thus, attention is paid to the effects of the hydroxyl group in the materials by choosing the method of synthesis and post-synthesis "annealing" of framework silanols. Additionally, ammonium and proton-exchanged nanocrystalline zeolite beta are investigated in an attempt to disentangle the factors contributing to the observed proton conductivity and to better design future materials.

E XPERIMENTAL

The samples were rinsed until the pH of the effluent was the same as that of the rinse water (and the measured conductivity values did not change with further rinsing). All samples were thoroughly washed and allowed to equilibrate with water before measurement. The samples were compressed to 80 in-oz (~2000 psi) and held at this pressure throughout the measurement.

Figure 3.1. PETMS was incorporated into the pores of zeolite beta, or grafted onto the pore walls of MCM-41, and sulfonated using oleum

R ESULTS AND D ISCUSSION

Mass loss below 200 °C is attributed to water and water bound to the sulfonic acid groups. The incorporation of phenylsulfonic acid groups into hydrophilic nanocrystalline zeolite beta (S-PE-BEA-X) further increases the proton conductivity to ~5*10-3 S/cm. Furthermore, phenylsulfonic acid-functionalized MCM-41 (S-PE-MCM-41) exhibits a conductivity value approximately that of S-PE-BEA-X.

Figure 3.3. X-ray diffraction data from i) as-synthesized samples. a. Pure silica

C ONCLUSION

However, in attempts to study cooperativity, phenylsulfonic acid-containing pure silica beta (S-PE-PE-BEA, synthesized from fluoride-containing media) proton conductivities are observed to be an order of magnitude lower than phenylsulfonic acid-containing beta synthesized from TEAOH-containing gels , even when their acid charges are similar. This may be due to the hydrophobicity/hydrophilicity of the pores, as the formation of a hydrogen-bonded water network is important for rapid proton hopping through the Grotthüss mechanism. The acid load by TGA and titration is slightly higher for S-PE-MCM-41, possibly leading to the slightly higher conductivity compared to the S-PE-BEA-X samples. exchanged form) and that the conductivity observed in these samples does not arise from a cooperative effect between the bound organic sulfonic acid groups and the aluminum acid sites.

R EFERENCES

PROTON CONDUCTIVITY OF ACID FUNCTIONALIZED ZEOLITE BETA, MCM-

A BSTRACT

I NTRODUCTION

Here, samples of zeolite beta, MCM-41, and MCM-48 are compared to address the effects of acid strength, pore size, and dimensionality on proton conductivity. Organically functionalized beta zeolite (denoted BEA), MCM-41, and MCM-48 containing phenyl sulfonic acid, propyl sulfonic acid, ethyl phosphoric acid, or ethyl carboxylic acid were prepared to test the effects of pore structure and acid strength on proton conductivity. Four organic silanes are incorporated into pure beta silica zeolite by direct synthesis and grafted onto calcined MCM-41 and MCM-48 surfaces.

E XPERIMENTAL

The mixture was quenched by addition to 500 ml of cold water and the samples were further washed with at least 2 liters of water (4 x 500 ml). Samples were quenched in 500 ml of cold water and washed with at least 2 liters of additional water (4 x 500 ml). Organic silanes were grafted onto the surfaces of MCM-41 and MCM-48 using the following procedure.

Figure 4.1. Illustration of the acid groups incorporated into zeolite beta, MCM-41, and MCM-48

R ESULTS AND D ISCUSSION

Acid-functionalized BEA materials (S-PE-BEA, S-MP-BEA, P-BEA, C-BEA) exhibit the lowest proton conductivity for each functionality compared to MCM-41 and MCM-48 with the same organic functionality. This observation may arise due to the 3-dimensional interconnected pore structure of MCM-48 compared to the 1-dimensional pore structure of MCM-41. Both S-PE-MCM-48 and S-PE-BEA-100 have interconnected 3-dimensional pore structures and exhibit almost identical acid loadings by TGA and titration.

Figure 4.2. X-ray data of i) pure-silica zeolite beta (BEA) containing a. no organic

C ONCLUSION

R EFERENCES

CONDUCTIVITY OF MONO- AND DIVALENT CATIONS IN THE MICROPOROUS

A BSTRACT

I NTRODUCTION

A method for measuring ionic conductivity is impedance spectroscopy, where a sinusoidal alternating voltage is applied across the sample of interest, and the current flowing through the sample is measured as a function of the sinusoidal frequency. A typical model for ion transport through a grain of powder material is the parallel combination of a resistor and capacitor, which in the frequency domain (after application of the Laplace transform) forms a semicircle in the complex (-Z” v. Z') track down level as a function of frequency. The conductivity of the sample is then calculated by geometrically scaling the low frequency section of the semicircle with the Z' (real) axis.

Figure 5.1. Schematic representation of anionic sites arising from framework

E XPERIMENTAL

VPI-9 Synthesis and Ion Exchange
Synthesis of Dense Zincosilicates
Powder X-ray Diffraction
Energy Dispersive Spectroscopy (EDS)
Impedance Spectroscopy

For each subsequent exchange cycle, the samples were decanted and 20 ml of fresh solutions were added. The samples were then heated to a temperature above their melting temperature (see Table 5.1), held there for 12 h, cooled slowly to 650 °C, and then allowed to cool to room temperature. Powder samples were prepared for impedance spectroscopic (IS) characterization by axial hydraulic pressing between 13 mm tungsten carbide dies.

Table 5.1. Synthesis, Crystallization, and Sintering Temperatures for Dense Zincosilicates

R ESULTS AND D ISCUSSION

Ion Exchange of VPI-9 and Zeolite X
Powder X-ray Diffraction Patterns of M-VPI-9
Impedance Spectroscopy
Ionic Conductivity and Activation Energy in M + -VPI-9
Ionic Conductivity and Activation Energy in M 2+ -VPI-9
Powder X-ray Diffraction and EDS of M 2 ZnSi X O 2(X+1)
Ionic Conductivity in M 2 ZnSi X O 2(x+1) and Comparison to Microporous VPI-9

Activation energy for each sample is extracted from the slope of the corresponding line, and these values are reported in Figure 5.5b. The M2ZnSi5O12 series was used to study the effect of the alkali cation at constant zincosilicate composition, and the K2ZnSixO2(x+1) series to investigate the effect of Si/Zn ratio (M2ZnSi4O10 materials are not known for all three cations of interest, K, Rb and Cs). Volatilization of the alkali metal is not ruled out, but is not believed to be significant, as ion-exchanged VPI-9 and zeolite X samples show a similar deficiency by EDS, and powder X-ray diffraction indicates the appropriate phases (with the exception of K2ZnSi2O6 for which the phase is unknown).