High performance tubular protonic

ceramic fuel cells via highly-scalable

extrusion process Liangzhu Zhu &
Ryan O'Hayre & Neal P. Sullivan
Visit to download the full and correct content document:
https://ebookmass.com/product/high-performance-tubular-protonic-ceramic-fuel-cells-
via-highly-scalable-extrusion-process-liangzhu-zhu-ryan-ohayre-neal-p-sullivan/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Elsevier Weekblad - Week 26 - 2022 Gebruiker

https://ebookmass.com/product/elsevier-weekblad-
week-26-2022-gebruiker/

Solid Oxide Fuel Cells: From Electrolyte-Based to

Electrolyte-Free Devices Bin Zhu

https://ebookmass.com/product/solid-oxide-fuel-cells-from-
electrolyte-based-to-electrolyte-free-devices-bin-zhu/

PEM Fuel Cells Gurbinder Kaur

https://ebookmass.com/product/pem-fuel-cells-gurbinder-kaur/

Jock Seeks Geek: The Holidates Series Book #26 Jill

Brashear

https://ebookmass.com/product/jock-seeks-geek-the-holidates-
series-book-26-jill-brashear/
Proton Exchange Membrane Fuel Cells Inamuddin

https://ebookmass.com/product/proton-exchange-membrane-fuel-
cells-inamuddin/

Nanotechnology in Fuel Cells Huaihe Song

https://ebookmass.com/product/nanotechnology-in-fuel-cells-
huaihe-song/

Ammonia Fuel Cells 1st Edition Ibrahim Dincer

https://ebookmass.com/product/ammonia-fuel-cells-1st-edition-
ibrahim-dincer/

Azure SQL Hyperscale Revealed: High-performance

Scalable Solutions for Critical Data Workloads 1st
Edition Zoran Bara■

https://ebookmass.com/product/azure-sql-hyperscale-revealed-high-
performance-scalable-solutions-for-critical-data-workloads-1st-
edition-zoran-barac-2/

Azure SQL Hyperscale Revealed: High-performance

Scalable Solutions for Critical Data Workloads 1st
Edition Zoran Bara■

https://ebookmass.com/product/azure-sql-hyperscale-revealed-high-
performance-scalable-solutions-for-critical-data-workloads-1st-
edition-zoran-barac/
 international journal of hydrogen energy xxx (xxxx) xxx

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

High performance tubular protonic ceramic fuel

cells via highly-scalable extrusion process

Liangzhu Zhu a,c,*, Ryan O'Hayre a,**, Neal P. Sullivan b,***

a
Metallurgical and Materials Engineering Department, Colorado School of Mines, 80401, USA
b
Mechanical Engineering Department, Colorado School of Mines, 80401, USA
c
Ningbo Institute of Materials Technology & Engineering, Chinese Academy of Sciences, 315201, China

highlights

Up to 66 cm long straight green PCFC anode supports are extruded.

Tubular PCFC cells with dense electrolyte layer are fabricated.

Three emerging PCFC cathode materials, BSCF, BCFZY, and PBSCF are synthesized and compared.

Maximum power density exceeds 500 mW cm2 at 600  C using BCFZY and BSCF cathodes.

A short 2-cell stack delivers 2.3 W power output at 600  C.

article info abstract

Article history: We report an effective method to fabricate long, anode-supported tubular protonic ceramic
Received 10 March 2021 fuel cells (PCFCs) and test cells in single-cell and short-stack mode. Further, we use our
Received in revised form tubular PCFC platform to directly compare three high performance cathodes reported in
31 May 2021 literature: BaCo0$4Fe0$4Zr0$1Y0$1O3-d (BCFZY), Ba0$5Sr0$5Co0$8Fe0$2O3-d (BSCF), and PrBa0.5-
Accepted 4 June 2021 Sr0$5Co1$5Fe0$5O6-d (PBSCF) using indentical preparation methods, which can minimize ef-
Available online xxx fects from variation of materials either due to suppliers or subsequent processing and
testing from different research labs. Using a BCFZY cathode, the maximum power density
Kewwords: of our tubular PCFC reaches 164, 308, and 517 mW cm2 at 500, 550, and 600  C, respec-
PCFC tively. A 2-cell tubular short stack provides a total power of 2.3 W at 600  C with tube di-
Solid oxide fuel cell ameters of 0.82 cm and a total tube active length of 3.2 cm. At 600  C, the maximum power
Tubular density reaches, 534, 517, and 326 mw cm2 for the BSCF, BCFZY, and PBSCF cathodes,
Extrusion respectively. Under the same conditions, the BSCF-based cell shows the lowest total
Cathode resistance mostly due to the lowest ohmic resistance and modest polarization resistance.
The BCFZY-based cell has the lowest polarization resistance but larger ohmic resistance
leading to a slightly higher total resistance than BSCF. The PBSCF cell has an ohmic
resistance close to BSCF but a total polarization resistance much larger than either BSCF or
BCFZY cell which results in the lowest overall performance.
© 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Ningbo Institute of Materials Technology & Engineering, Chinese Academy of Sciences, 315201, China.
** Corresponding author.
*** Corresponding author.
E-mail addresses: zhuliangzhu@nimte.ac.cn (L. Zhu), rohayre@mines.edu (R. O'Hayre), nsulliva@mines.edu (N.P. Sullivan).
https://doi.org/10.1016/j.ijhydene.2021.06.018
0360-3199/© 2021 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
2 international journal of hydrogen energy xxx (xxxx) xxx
Introduction
Protonic ceramic cells (PCCs) including protonic ceramic fuel
cells (PCFCs) and protonic ceramic electrolyser cells (PCECs)
have rapidly attracted attention in the past ten years. PCCs
offer several intriguing advantages over their conventional
oxygen ion conducting solid oxide cells (SOCs) predecessors,
including higher theoretical fuel utilization under fuel cell
mode, higher coking resistance, the ability to produce nearly
dry, pure H2 under electrolysis mode, and the ability to oper-
ate at temperatures typically 200  C lower than conventional
SOCs. These advantages result in lower materials cost, longer
device lifetime, and the potential to directly match their
electrochemical operation to the temperature window of
several key thermochemical processes [1e8].
High protonic conductivity of ~0.03 S cm1 at 600  C has
been measured in both BCZYYb7111 (BaCe0$7Zr0$1Y0.1Yb0.1-
O3ed) and BCZYYb4411 (BaCe0$4Zr0$4Y0.1Yb0.1O3ed) electrolytes
[3,9]. Maximum power densities exceeding 600 mW cm2 at
600  C have also been reported by a few studies [3,5,6]. While
these studies have led to promising performance for PCC vs.
SOC lab-scale button cells at lower temperatures, progress in
large-scale PCC development has lagged. In general, SOC
commercialization has focused on either planar or tubular
geometries [10e16]; the same geometries are also preferred for
PCC scale-up and stacking. However, PCC scale-up is
hampered by several important issues. For example, based on
our experience sintering barium perovskite type proton con-
duction materials, we have found these materials react
considerably with substrates such as low-cost alumina or
zirconia plates, necessitating more costly solutions for vol-
ume manufacturing. Another difficulty in fabricating large
planar PCC stack assembly lies in the thermal expansion
mismatch between the metal alloy interconnects and the PCC
due to thermo-chemical expansion effects, i.e., zero or even
negative expansion can occur around the PCC working tem-
perature (normally 600  C) [17]. This behavior severely chal-
lenges planar PCC stack stability and operation, particularly
during start up and shut down.
Many of the issues challenging planar PCC scale-up and
stacking, however, can be potentially mitigated by pursuing a
tubular PCC approach. Firstly, tubular PCC parts can be hang-
fired, therefore eliminating the need for a setter, which miti-
gates reactivity and results in better quality control in cell
fabrication. Secondly when assembled into stacks, tubular
PCCs are mostly free from metal alloy interconnects and the
geometry minimizes stress concentration risers and the ef-
fects of thermo-chemical expansion. Additionally, the tubular
geometry generally provides higher mechanical strength than
a planar geometry of the same wall thickness. Finally, the
tubular PCC geometry better facilitates integration with
additional catalysts for applications where a second-step
thermochemical reaction is coupled to the electrochemical
cell process, such as catalytic membrane reformer applica-
tions or the in-situ synthesis of ammonia or other value-
added chemicals [18,19]. Therefore, the fabrication of large
tubular PCCs may offer a promising avenue for commerciali-
zation. So far, however, relatively few studies have pursued
the tubular PCC approach.
Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
Different methods have been developed for tubular PCFC
fabrication [20e27], among which the phase inversion process
is used for most current lab-scale tubular PCFC demonstration
efforts. This process in general is suitable for making hollow
structures with finger-like or aligned pores by exploiting self-
assembling “polymer-rich” (solid phase) and “polymer-poor”
phases. The “polymer rich” network phase aggregates the
solid ceramic particles that will be retained after firing, while
the “polymer-poor” phase dissovles in a nonsolvent liquid
phase which is usually water, leaving an aligned pore struc-
ture [22,28e30]. Since the outer surface first sees the non-
solvent liquid during the phase inversion proces, aligned
pores generally form with an orientation perpendicular to
outer surface. Although vertical pores are desired in fuel cell
electrodes to facilitate mass transport, this structure usually
results in poor mechanical strength. In both cases, pore
orientation develops due to either the one-dimensional
chemical gradient of non-solvent concentration (i.e., water
concentration at the outer surface of a casted tube in the
phase inversion process) or the one-dimensional temperature
gradient in the solvent. Such aligned pores can lead to me-
chanical issues due to the highly anisotropic structure and
may ultimately lead to cell failure, thereby diminishing the
advantages associated with improved mass transport. This is
likely one of the reasons that such processes have not been
widely adopted for SOC scale-up/volume production. From a
manufacturability perspective, extrusion may therefore pro-
vide a better route towards large-scale production for both
tubular SOC and tubular PCC fabrication. Unfortunately, there
is very limited work on the extruded PCCs, and when extru-
sion has been used for tubular PCC production, the fabrication
details are often absent [19,26].
In addition to the preference for hard-to-scale phase-
inversion and slip-casting processes, most prior tubular PCFC
studies have deployed earlier-generation LSM (La1-xSrxMnO3-d)
or LSCF (La1-ySryCo1-xFexO3- d) cathodes. Very few studies
have yet sought to implement some of the more recently
advanced PCC-specific cathodes such as BCFZY or PBSCF.
Thus, in addition to developing a scalable extrusion-based
PCC fabrication protocol in this work, we also seek to imple-
ment and compare three of these emerging cathode materials
(BCFZY, PBSCF, and BSCF) using a single long tubular PCC
parent cell to facilitate direct comparison under well-
controlled testing conditions. We also demonstrate the per-
formance of a short-stack 2-tube bundle. The ultimate goal of
this study is to bring the tubular application of PCCs one step
closer to fruition and to provide a rational comparison of
currently developed PCC cathode materials in order to accel-
erate the development of PCC technology.
Materials and methods
Anode support extrudate preparation
Anode support powder was prepared by ball milling NiO and
BaCe0$7Zr0$1Y0.1Yb0.1O3-d (BCZYYb) at a weight ratio of 65:35.
The BCZYYb precursor was directly added as mixture of
BaCO3, CeO2, ZrO2, Y2O3, and Yb2O3 according to the set stoi-
chiometry via in-situ solid state reactive sintering process
 international journal of hydrogen energy xxx (xxxx) xxx
[1,5]. 20 wt% of corn starch and 3 wt% of methyl cellulose were
added as pore former and water soluble binder respectively.
The resulting powder precursor was blended by ball milling
for 15 h. The powder after ball milling was stored for subse-
quent extrusion processes.
Anode support extrusion
To extrude the anode support tubes, the as-prepared anode
support powder was mixed with water and firstly extruded
using a kitchen type pasta extruder (due to lack of laboratory
pug mill). This step generates a ceramic clay with suitable
viscosity for further extrusion. The ceramic clay was then
shaped into rectangular bricks, sealed and aged for 16 h before
final tube extrusion. For the final extrusion, the aged clay was
loaded into the pressing chamber, slightly vacuumed, and
slowly extruded (See Fig. S1 in Supplementary Information
(SI)). The extruded green tubes were firstly dried inside an
alumina tube with an inner diameter slightly larger than the
outer diameter of the extruded tube at 30  C for about 3 h
followed by additional drying at 80  C for 5 h. This drying
procedure produced straight tubes with good mechanical
strength to facilitate further handling.
Electrolyte and cathode powder preparation
To improve electrolyte density, nanosize BCZYYb powder was
prepared by a sol-gel process using procedures similar to those
reported in previous studies [4,31]. Briefly, citric acid and EDTA
were added into distilled water under stirring. Nitrates of
Ba(NO3)2, Ce(NO3)3$6H2O, ZrO(NO3)2 (35 wt.-% in dilute nitric
acid), Y(NO3)36H2O, and Yb(NO3)3$5H2O were added under
stirring. 1 wt% of NiO was added directly to the sol in the form
of Ni(NO3)2$6H2O as a sintering agent. NH3.H2O was slowly
added at a ratio of 1000 ml NH3$H2O per mole of citric acid or
EDTA, after which the solution became dark red and trans-
parent. The molar ratio of EDTA: citric acid: total metal ions was
1.5 : 1.5: 1, with a pH of ~9. The solution was placed on a 250  C
hot plate under stirring to form a gel. The gel was dried at 150  C
for 24 h to form a black, porous, charcoal-like precursor. The
solid precursor was manually crushed and slightly ground, and
then ball milled with yttria-stabilized zirconia media and iso-
propanol for 24 h. Following milling, the black charcoal slurry
was dried at 90  C for 12 h. The powder was then calcined at
800  C for 10 h, and then wet ball milled in isopropanol with
YSZ media for 72 h to further reduce the particle size. The slurry
was then dried for 24 h at 90  C and stored for electrolyte slurry
preparation. The three cathodes, Ba0$5Sr0$5Co0$8Fe0$2O3-d (BSCF),
BaCo0$4Fe0$4Zr0$1Y0$1O3-d, PrBa0.5Sr0$5Co1$5Fe0$5O6-d (PBSCF),
were synthesized in the same way (without nickel nitrate
addition) but calcined at lower temperature and shorter time
(600  C for 5 h) to minimize grain growth.
Full-cell fabrication
Due to limited furnace size, the dried long green tubes were
cut into shorter tubes. A pair of small holes were made at the
top of each tube to facilitate hang firing using platinum wires
for tube suspension. These shorter green tubes were debin-
dered and firstly sintered at 1200  C for 2 h. To apply the
Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
3
electrolyte layer, the BCZYYb nano-powder was mixed with a
binder (5 wt % V-006 in alpha-terpineol) and a dispersant
(20 wt% Solsperse 28,000) in alpha terpineol to prepare a slurry
using a method reported previously but with modified slurry
composition [4]. The weight ratio was set as electrolyte:
binder: dispersant ¼ 5:1:2. The mixed electrolyte slurry was
then brush-applied to the outer surface of the pre-sintered
anode support tube. After electrolyte application, the tubes
were hang-fired by platinum wires inside open alumina tubes
at 1450  C for 10 h. After sintering, a center region of each tube
was brushed painted with one of the three cathode pastes. All
three cathodes pastes were prepared by mixing 4 g of cathode
powder, 1 g of BCZYYb (synthesized via solid state reaction
and calcined at 1400  C for 10 h), 0.4 g 5 wt % V-006 in alpha-
terpineol, and 1 g of 20 wt% Solsperse 28,000 [4]. The non-
coated region of each tube was covered by tape to ensure an
accurate electrode area. After application of cathode paste,
the tubes were fired at 900  C for 5 h, completing the cell
fabrication process. The tubes had a final outer diameter of
~0.82 cm. For single cells, the cathode length is 0.5 cm making
the active cathode area of ~1.29 cm2. For short stack test, each
tube has a cathode length of 1.6 cm, making the total active
area for the two tubes in series connection 8.24 cm2.
Cell testing assembly
After cathode application and firing, a thin layer of Au paste
was applied on top of the cathode to improve current collec-
tion. Ag wire was then wrapped on top of the Au layer, short
Ag grids were manually painted across the center Ag wire
extending to two edges of cathode area, as shown in Fig. S2a
(See SI). A thin Au paste coating was also to the anode side
(inner anode chamber) and four Ag grids were painted at
about 90 interval using a silver paste (DAD-87). A cylindrical
Ag mesh was gently inserted into the anode chamber with two
wire leads extending out for electrical connection and mea-
surement. The space inside the cylindrical Ag mesh was filled
with porous insulation wool to secure good electrical contact.
Finally, the tubular cell with wire leads was mounted in a
specially designed test fixture (Fig S2a and 2b). The tube ends
were sealed using Ceramabond (552-VFG). A piece of quartz
tube was placed roughly co-axially with the ceramic cell. A
thin air tube was inserted into the center zone of the quartz
tube whose open ends were blocked by insulation wool.
Lastly, a thermocouple was placed beside the tubular cell. This
completed the initial test assembly. The whole assembly was
then loaded to a vertical furnace and heated up for cell testing.
For the short stack demonstration, the cells were connected in
series and tested similarly except no forced air flow was
provided.
Characterization and measurement
A Keithley 2000 multimeter was used to monitor cell voltage.
Solartron 1260 and 1287 were used for electrochemical mea-
surements and analyses. 4-probe and 2-probe measurements
were used for single cell and short stack testing, respectively.
After testing, cell cross section imaging was conducted using a
JEOL 7000 Field-Emission Scanning Electron Microscope
(FESEM). Structural information for the synthesized cathode
4 international journal of hydrogen energy xxx (xxxx) xxx
materials was acquired by a PANalytical PW3040 X-ray
Diffractometer.
Results and discussion
Cell macro- and micro-structure
Fig. 1a shows an example of an extruded NiO/BCZYYb green
tube reaching the maximum extrudable length in this work,
Fig. 1 e Photographs documenting the tubular PCC
fabrication process used in this study. (a) a 66 cm long
green tube after extrusion. (b) a pre-sintered short tube
with applied electrolyte layer on the outer surface. (c) hang-
firing apparatus for half-cell sintering. (d) Sintered half-cell
with electrolyte layer on the outer surface. (For
interpretation of the references to color in this figure
legend, the reader is referred to the Web version of this
article.)
Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
about 66 cm long. The extruded tube is reasonably, although
not perfectly straight, and is sufficiently long for large scale
tubular stack application. Due to limited size of the sintering
furnaces available in our lab, we cut these long tubes into
shorter tube segments for firing and cell testing. Fig. 1b shows
a representative unfired tube segment with applied electrolyte
layer. Fig. 1c shows the hang-firing process used for this study
which eliminates contact of the cell with the substrate.
Finally, Fig. 1d shows a fully-sintered half-cell. The visible
reflection of light serves as the first evidence of a dense elec-
trolyte layer.
Fig. 2aed provide representative SEM cross-section images
from a tested cell at various magnifications for the anode
support/electrolyte/cathode sandwiched structure. The cell
consists of a ~650 mm thick Ni/BCZYYb anode support (after
reduction), a ~15 mm thick BCZYYb electrolyte layer, and a
~15 mm thick BCFZY cathode. The anode support shows evenly
distributed pores of less than 10 mm diameter facilitating mass
transport of fuel (Fig. 2a). As shown in Fig. 2b, the electrolyte
layer is dense with a few isolated pores of submicron diameter
which do not affect the cell open circuit voltage (OCV). The
anode support shows nanosize exsolved Ni particles on the
grains of the percolating electrolyte phase (Fig. 2c). In-situ
nano-catalyst formation (exsolved and catalytically active Ni
nanoparticles that decorate the protonic ceramic phase in the
cermet anode) has been previously reported by several groups
and contributes to high performance as well as high coking
resistance against hydrocarbon fuels [5,32e34]. Even after
testing, the BCFZY cathode remains highly homogeneous and
uniform with fine particle size and porosity (~100 nm), and the
cathode/electrolyte interface remains coherent without signs
of fracture or delamination (Fig. 2d).
Fuel cell performance
We first establish the baseline performance of our tubular cell
using the BCFZY cathode, as we have previous experience
applying and testing this cathode on planar PCC and SOC ar-
chitectures [1,4,5,35]. Fig. 3 shows the performance of the
BCFZY-based cell under dry hydrogen and humidified air from
500 to 600  C. The cell yields OCVs of 1.13, 1.11, and 1.08 V and
maximum power densities (MPDs) of 164, 308, and
517 mW cm2 at 500, 550, and 600  C, respectively. Represen-
tative performances from BSCF cathode at the same temper-
atures are shown in Fig.S3. To the author's knowledge, this
performance represents the highest tubular PCFC perfor-
mance so far reported on highly scalable extrusion process at
the same temperature. The performance however are lower
than that reported by Chen et al. [20,23] via phase inversion
process. Chen et al. applied a 6-coat phase-inversion process
to produce a highly pore-oriented, ultra-thin ~220 mm anode
support tube. While this approach greatly facilitates mass
transport, as previously mentioned, the thin and anisotropic
nature of the support can potentially compromise robustness.
Further, the lengthy multi-step, multi-coat phase inversion
fabrication process may not be ammenable to volume
production.
Electrochemical Impedance Spectroscopy (EIS) analysis of
the baseline BCFZY-based cell under OCV mode (Fig. 4) reveals
significant decreases in both the ohmic resistance (measured
 international journal of hydrogen energy xxx (xxxx) xxx 5

Fig. 2 e Cross-sectional microstructure of a tested cell. (a) and (b) low and high magnification images showing anode
support/electrolyte/cathode sandwiched structure after reduction and cell testing. (c) Exsolution of Ni particles in the
electrolyte phase on the anode side was observed. (d) The cathode (BCFZY, in this example) is porous and homogeneous
with <100 nm particle size.

Fig. 4 e Electrochemical Impedance Spectroscopy (EIS)

analysis of a tubular cell with BCFZY cathode at different
temperatures.

Fig. 3 e Polarization and power density curves of a single

tubular cell with BCFZY cathode at different temperatures.
via the high frequency intercept of the EIS spectrum) and
polarization resistance (measured by the width of the
impedance arc) as the temperature increases from 500 to
600  C. At 600  C, the cell performance is predominantly
limited by ohmic resistance. Further improvements should
therefore seek to decrease the total ohmic resistance, focusing
particularly on enhacing electronic conduction in cathode,
decreasing ionic resistance in the electrolyte layer (e.g. via a
thinner electrolyte), and mitigating contact and interfacial
resistance at the electrolyte/electrode interfaces (e.g. via
introduction of functional layers or interfacial contact layers).
We subsequently tested the performance of a BCFZY-based
2-cell tubular short stack using series cell connection (Fig. 5).
The two-tube bundle produces a 2.0 V OCV and reaches a total
maximum power of 2.3 W at 600  C. The lower average OCV
produced by the two-tube bundle (~1 V) as compared to the
individual single cells (~1.08 V) in Fig. 3 indicates possible
leakage in the stack assembly. The maximum power density
of the twodtube stack, normalized to the total electrode area,
reaches ~280 mW cm2. This is significantly lower than that of
the single cell (517 mW cm2), which we again attribute to
challenges with gas manifolding and cell sealing as well as a

Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
6 international journal of hydrogen energy xxx (xxxx) xxx

switch from a 4 probe measurement configuration for single-

tube testing to a 2 probe measurement for the two-tube
testing, which prevents elimination of the electrical wire
lead and contact resistances.

Comparison of BCFZY, BSCF and PBSCF cathodes

An important objective of this study is to compare recently-

Fig. 5 e IV and power characteristics of the 2-cell short
tubular stack with BCFZY cathode at 600  C.
reported high-performance PCFC cathodes (BCFZY [1], BSCF
[6], and PBSCF [3]). We utilize a single tubular cell ancestor to
provide a consistent and reliable comparison. Here we use the
stoichiometric cathode compositions exactly as reported by
these previous studies but otherwise use identical materials
synthesis processes, cathode application and testing proced-
ures to minimize any impact from experimental or processing
variations. Fig. 6 compares the XRD pattern of the three syn-
thesized cathode powders. All three powders are synthesized
by the same sol-gel approach and each show a similar

Fig. 6 e Comparison of the three cathode powders. (a) XRD patterns of the three synthesized cathode powders with a
reference comparison to the perovskite-type SOFC cathode La0.6 Sr0·4Co0·9Fe0·1O3-d (LSCF): pdf# 01-089-5720. For this XRD
comparision, the cathode powders were first further calcined at 900  C for 5 h to mimic the real conditions after cathode
application on cells. (b) A photograph taken under the same background exposure shows the PBSCF is the blackest, BCFZY is
less dark (approximately dark-brown). The color of the BSCF powder is between PBSCF and BCFZY, but closer to PBSCF. The
cathodes shown in the photograph were only calcined at 600  C. (For interpretation of the references to color in this figure
legend, the reader is referred to the Web version of this article.)

Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
 international journal of hydrogen energy xxx (xxxx) xxx 7

perovskite-type pattern (here referenced to La0.6 Sr0.4 Co0$9Fe0.1,

LSCF, pdf# 01-089-5720). Relative to the LSCF reference pattern,
the diffraction peaks of the three cathode materials increas-
ingly shift toward lower angles from PBSCF to BSCF to BCFZY,
reflecting the increasing lattice parameter. Although PBSCF has
been generally reported as a double perovskite and refined to
the orthorhombic structure (pmmm, space group# 59) [36,37],
the absence of doublets here and also as seen in other reported
double perovskite structure cathode materials such as Pr0.5-
Ba0$5CoO3-d (PCO) suggests that its structure may be highly
sensitive to the materials preparation method [38]. For direct
comparison, we firstly treated PBSCF also as cubic perovskite
(pm-3m, space group #221). Lattice refinement to cubic perov-
skite yields a lattice constant a of 3.881, 4.005, and 4.115
A for
PBSCF, BSCF and BCFZY, respectively. The refined cell param-
eters for both BSCF and BCFZY are close to those reported in the
literature [1,38]. It is worth mentioning that if the PBSCF is Fig. 7 e IV and power density performance comparison of
treated as double perovskite and refined to orthorhombic small, nominally identical tubular cells (all cut from a
structure instead, then a and b remain ~3.881
A while c becomes single parent tubular half-cell) with either BCFZY, PBSCF or
~2a, i.e., ~7.762

A. These values are close to those obtained by BSCF cathode at 600  C.
Jiang et al. (3.851, 3.854, and 7.715

A for a, b, and c, respectively)
by refining to the double perovskite structure [37]. These results
shows signs of activation polarization limitation (reflected in
also suggest that it is likely possible to synthesize many other
its concave shape) in the low to mid-range current density
similar cathode materials by moderate substation of the A site
region. The BSCF-based cell also shows slight hints of such
and B site elements with other metal elements in the cubic
effects, but the BCFZY-based cell produces a consistently
perovskite structure, leveraging identically the same synthesis
convex j-V curve shape and instead shows signs of possible
procedure used in this study.
mass transport limitations in the higher current density re-
Fig. 6b compares the visual color of the three cathodes.
gion which likely contribute to its slightly lower MPD.
After calcination, PBSCF and BSCF show a pure black color,
EIS analysis under OCV mode (Fig. 8) is consistent with the
similar to black charcoal. BCFZY, while still very dark, is not
observations extracted from the j-V curve measurement. The
pure black (note the color difference is not quite differentiable
BSCF-based cell has the lowest total cell area specific resis-
when captured by a camera as compared with observation by
tance (ASR) therefore achieving the highest MPD, while the
naked eye). It is known that the apparent color of materials
PBSCF-based cell has the highest ASR, leading to the lowest
like semiconductors can usually be explained by band gap,
MPD. Based on the low and high frequency intercepts, it is
free electron concentrations, surface smoothness, etc. [39]
clear that the PBSCF cell is largely limited by the electrode
The difference in the apparent color seen here may suggest
polarization resistance as measured by the size of the
slightly different intrinsic physical properties in these three
extended impedance arc, and reflecting the contributions
cathodes.
from both activation and mass transport effects. Due to the
Fig. 7 compares the cell performance under dry H2/
concave shape of the j-V curve, it may be further suggested
humidfied air at 600  C for the three cathodes. MPDs of 534,
that the activation polarization resistance dominates the
517, and 326 mW cm2 are obtained for BSCF, BCFZY, and
PBSCF, respectively. Although the three cells came from a
single tube and were fabricated and tested identically, they
show slightly varying OCVs. We attribute this phenomena to
slight leaks or sealing differences during cell assembly. Such
small leaks may cloud results. However, we note that the
material with the lowest OCV demonstrates the highest power
density, implying that gas leakage is not responsible for the
observed performance differences. It is not surprising that
BSCF shows the highest MPD performance among the three
cathodes, although it is generally considered as only having
mixed oxygen ion and electronic conduction and was origi-
nally developed for conventional SOFC applications. Indeed,
to our knowledge the highest performance ever reported for a
PCFC came from a planar cell that employed a BSCF cathode
[6]. On the other hand, the PBSCF cells shows noticeably lower
performance although it has also previously been reported to
produce high performance in PCFC button cells, albiet under
different synthesis and testing conditions than those used Fig. 8 e Comparision of EIS results of cells with BCFZY,
here [3]. Interestingly, the j-V curve for the PBSCF-based cell PBSCF and BSCF cathode at 600  C.

Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
8 international journal of hydrogen energy xxx (xxxx) xxx
behavior of the PBSCF-based cell. In contrast, the BCFZY-
based cell shows the lowest total polarization resistance
(smallest arc) but has a much larger ohmic resistance as seen
by its significantly higher high-frequency impedance inter-
cept than either the BCFZY or PBSCF cells. Thus, ohmic
resistance dominates the behavior of the BCFZY-based cell, in
agreement with its j-V curve and consistent with findings
from previous studies [1].
This study, based on controlled tests and parallel com-
parision from nominally identical tube segments, suggests
that the BCFZY-based cell may, in part, be limited by ohmic
conduction within the cathode itself. In other words, further
improvement could likely be obtained by considering doping
strategies to increase the electronic conductivity of BCFZY
and/or blending with a highly conductive second phase [40].
We hypothesize that the relatively lower electronic conduc-
tivity of BCFZY vs. BSCF and PBSCF is likely due to the
appreciable amount of Zr and Y doping which are introduced
to increase its oxygen ion and proton conduction [1,41,42]. We
note the inherent tradeoffs in cathode design, however, as
these large, fixed-valent cations also enhance the structural
and chemical stability of BCFZY (especially vs. BSCF), illus-
trating the potentially complex and multi-variate challenges
of cathode optimization.
Summary
We established an effective and facile method to fabricate
long green tubes for protonic ceramic cells (PCCs) that yields
good power density performance (>500 mW cm2) under H2/
air fuel cell testing. The process is well positioned for large-
scale PCC applications. We also used the tubular architec-
ture to evaluate three state-of-the-art high-performance PCFC
cathodes (BCFZY, BSCF, and PBSCF). Specific summary from
this study include:
1) Up to 66 cm long straight green anode supports can be
successfully extruded using the common NiO/BCZYYb
protonic ceramic materials set. Hang-firing leads to high
quality anode supported tubular PCFCs.
2) Using the BCFZY cathode, the maximum power density
reached 164, 308, and 517 mW cm2 at 500, 550, and 600  C,
respectively. A short 2-cell stack delivered 2.3 W power
output with a total active tube length of 3.2 cm and a
diameter of 0.82 cm at 600  C.
3) At 600  C, segments cut from a common parent tube coated
with BSCF, BCFZY, and PBSCF-based cathodes attained
maximum power densities of 534, 517, and 326 mw cm2,
respectively.
4) The BSCF-based cell showed the lowest ohmic resistance
and modest polarization resistance, leading to the lowest
overall total resistance and the highest MPD. The BCFZY-
based cell has the lowest total polarization resistance but
much larger ohmic resistance, leading to a slightly higher
total resistance and slightly lower MPD than BSCF. The
PBSCF-based cell has an ohmic resistance close to BSCF but
a total polarization resistance about twice that of the BSCF
cell and four times of the BCFZY cell which result in the
lowest performance and the lowest MPD.
Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
Acknowledgements
The information, data, and work presented herein was funded
in part by the Advanced Research Projects Agency-Energy
(ARPA-E), U.S. Department of Energy, under Award Number
DE-AR0000808 and by the Army Research Office under grant
number W911NF-17-1-0051. The views and opinions of au-
thors expressed herein do not necessarily state or reflect those
of the United States Government or any agency thereof.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ijhydene.2021.06.018.
references
[1] Duan C, Tong J, Shang M, Nikodemski S, Sanders M, Ricote S,
et al. Readily processed protonic ceramic fuel cells with high
performance at low temperatures. Science 2015;349:1321.
[2] Babilo P, Uda T, Haile SM. Processing of yttrium-doped
barium zirconate for high proton conductivity. J Mater Res
2007;22:1322e30.
[3] Choi S, Kucharczyk CJ, Liang Y, Zhang X, Takeuchi I, Ji H-I,
et al. Exceptional power density and stability at intermediate
temperatures in protonic ceramic fuel cells. Nature Energy
2018;3:202e10.
[4] Duan C, Kee R, Zhu H, Sullivan N, Zhu L, Bian L, et al. Highly
efficient reversible protonic ceramic electrochemical cells for
power generation and fuel production. Nature Energy
2019;4:230e40.
[5] Duan C, Kee RJ, Zhu H, Karakaya C, Chen Y, Ricote S, et al.
Highly durable, coking and sulfur tolerant, fuel-flexible
protonic ceramic fuel cells. Nature 2018;557:217e22.
[6] An H, Lee H-W, Kim B-K, Son J-W, Yoon KJ, Kim H, et al. A
5  5 cm2 protonic ceramic fuel cell with a power density of
1.3 W cme2 at 600  C. Nature Energy 2018;3:870e5.
[7] Chen Y, deGlee B, Tang Y, Wang Z, Zhao B, Wei Y, et al. A
robust fuel cell operated on nearly dry methane at 500  C
enabled by synergistic thermal catalysis and electrocatalysis.
Nature Energy 2018;3:1042e50.
[8] Babilo P, Haile SM. Enhanced sintering of yttrium-doped
barium zirconate by addition of ZnO. J Am Ceram Soc
2005;88:2362e8.
[9] Yang L, Wang S, Blinn K, Liu M, Liu Z, Cheng Z, et al. Enhanced
sulfur and coking tolerance of a mixed ion conductor for
SOFCs: BaZr0.1Ce0.7Y0.2exYbxO3ed. Science 2009;326:126.
[10] Lawlor V. Review of the micro-tubular solid oxide fuel cell
(Part II: cell design issues and research activities). J Power
Sources 2013;240:421e41.
[11] Lawlor V, Griesser S, Buchinger G, Olabi AG, Cordiner S,
Meissner D. Review of the micro-tubular solid oxide fuel cell.
Part I. stack design issues and research activities. J Power
Sources 2009;193:387e99.
 international journal of hydrogen energy xxx (xxxx) xxx
[12] Mahato N, Banerjee A, Gupta A, Omar S, Balani K. Progress in
material selection for solid oxide fuel cell technology: a
review. Prog Mater Sci 2015;72:141e337.
[13] Minh NQ, Mizusaki J, Singhal SC. Advances in solid oxide fuel
cells: review of progress through three decades of the
international symposia on solid oxide fuel cells. ECS
Transactions 2017;78:63e73.
[14] Singhal SC. Solid oxide fuel cells: status, challenges and
opportunities. Ind Ceram 2008;28:53e9.
[15] Timurkutluk B, Timurkutluk C, Mat MD, Kaplan Y. A review
on cell/stack designs for high performance solid oxide fuel
cells. Renew Sustain Energy Rev 2016;56:1101e21.
[16] Zhu JH, Ghezel-Ayagh H. Cathode-side electrical contact and
contact materials for solid oxide fuel cell stacking: a review.
Int J Hydrogen Energy 2017;42:24278e300.
[17] Loken A, Ricote S, Wachowski S. Thermal and chemical
expansion in proton ceramic electrolytes and compatible
electrodes. Crystals 2018;8. 365 (68 pp.).
[18] Malerød-Fjeld H, Clark D, Yuste-Tirados I, Zano  n R, Catala
 n-
Martinez D, Beeaff D, et al. Thermo-electrochemical
production of compressed hydrogen from methane with
near-zero energy loss. Nature Energy 2017;2:923e31.
[19] Vøllestad E, Strandbakke R, Tarach M, Catala  n-Martı́nez D,
Fontaine M-L, Beeaff D, et al. Mixed proton and electron
conducting double perovskite anodes for stable and
efficient tubular proton ceramic electrolysers. Nat Mater
2019;18:752e9.
[20] Chen C, Dong Y, Li L, Wang Z, Liu M, Rainwater BH, et al.
Electrochemical properties of micro-tubular intermediate
temperature solid oxide fuel cell with novel asymmetric
structure based on BaZr0.1Ce0.7Y0.1Yb0.1O3d proton
conducting electrolyte. Int J Hydrogen Energy 2019;44:16887e97.
[21] Amiri T, Singh K, Sandhu NK, Hanifi AR, Etsell TH, Luo J-L,
et al. High performance tubular solid oxide fuel cell based on
Ba0.5Sr0.5Ce0.6Zr0.2Gd0.1Y0.1O3-dProton conducting
electrolyte. J Electrochem Soc 2018;165:F764e9.
[22] Beibei H, Dong D, Yihan L, Ling Z, Jigui C. Fabrication and
evaluation of stable micro tubular solid oxide fuel cells with
BZCY-BZY bi-layer proton conducting electrolytes. Int J
Hydrogen Energy 2014;39:19087e92.
[23] Chen C, Dong Y, Li L, Wang Z, Liu M, Rainwater BH, et al. High
performance of anode supported BaZr0.1Ce0.7Y0.1Yb0.1O3-
d proton-conducting electrolyte micro-tubular cells with
asymmetric structure for IT-SOFCs. J Electroanal Chem
2019;844:49e57.
[24] Hanifi AR, Sandhu NK, Etsell TH, Jing-Li L, Sarkar P.
Fabrication and characterization of a tubular ceramic fuel
cell based on BaZr0.1Ce0.7Y0.1Yb0.1O3- proton conducting
electrolyte. J Power Sources 2017;341:264e9.
[25] Hanifi AR, Sandhu NK, Etsell TH, Sarkar P. Development of a
novel proton conducting fuel cell based on a Ni-YSZ support.
J Am Ceram Soc 2017;100:4983e7.
[26] Robinson S, Manerbino A, Grover Coors W, Sullivan NP.
Fabrication and performance of tubular, electrode-supported
BaCe 0.2Zr0.7Y0.1O3- fuel cells. Fuel Cell 2013;13:584e91.
Please cite this article as: Zhu L et al., High performance tubular protonic ceramic fuel cells via highly-scalable extrusion process,
International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2021.06.018
9
[27] Zhao F, Jin C, Yang C, Wang S, Chen F. Fabrication and
characterization of anode-supported micro-tubular solid
oxide fuel cell based on BaZr0.1Ce0.7Y 0.1Yb0.1O3-
electrolyte. J Power Sources 2011;196:688e91.
[28] Purkait MK, Sinha MK, Mondal P, Singh R. Chapter 1 -
introduction to membranes. Interface Science and
Technology 2018;2.
[29] Ismail AF, Khulbe KC, Matsuura T. Chapter 2 - RO membrane
preparation. Reverse Osmosis 2019:25e56.
[30] Hołda AK, Vankelecom IFJ. Understanding and guiding the
phase inversion process for synthesis of solvent resistant
nanofiltration membranes. J Appl Polym Sci 2015;132.
[31] Shang M, Tong J, O'Hayre R. A novel wet-chemistry method
for the synthesis of multicomponent nanoparticles: a case
study of BaCe0.7Zr0.1Y0.1Yb0.1O3d. Mater Lett
2013;92:382e5.
[32] Neagu D, Oh T-S, Miller DN, Me nard H, Bukhari SM,
Gamble SR, et al. Nano-socketed nickel particles with
enhanced coking resistance grown in situ by redox
exsolution. Nat Commun 2015;6:8120.
[33] Liu Y, Jia L, Chi B, Pu J, Li J. In situ exsolved Ni-decorated
Ba(Ce0.9Y0.1)0.8Ni0.2O3d perovskite as carbon-resistant
composite anode for hydrocarbon-fueled solid oxide fuel
cells. ACS Omega 2019;4:21494e9.
[34] Shi N, Xie Y, Yang Y, Xue S, Li X, Zhu K, et al. Review of
anodic reactions in hydrocarbon fueled solid oxide fuel cells
and strategies to improve anode performance and stability.
Mater for Renew and Sustain Energy 2020;9:6.
[35] Duan C, Hook D, Chen Y, Tong J, O'Hayre R. Zr and Y co-
doped perovskite as a stable, high performance cathode for
solid oxide fuel cells operating below 500  C. Energy Environ
Sci 2017;10:176e82.
[36] Zhao B, Zhang L, Zhen D, Yoo S, Ding Y, Chen D, et al. A
tailored double perovskite nanofiber catalyst enables
ultrafast oxygen evolution. Nat Commun 2017;8:14586.
[37] Jiang L, Wei T, Zeng R, Zhang W-X, Huang Y-H. Thermal and
electrochemical properties of PrBa0.5Sr0.5Co2xFexO5þd (x
¼ 0.5, 1.0, 1.5) cathode materials for solid-oxide fuel cells. J
Power Sources 2013;232:279e85.
[38] Mohamed R, Cheng X, Fabbri E, Levecque P, Ko € tz R, Conrad O,
et al. Electrocatalysis of perovskites: the influence of carbon
on the oxygen evolution activity. J Electrochem Soc
2015;162:F579e86.
[39] Fulay P, Lee J-K. Electronic, magnetic, and optical materials.
2nd ed. Taylor & Francis; 2016.
[40] Zhu L, Zhang L, Virkar AV. A parametric model for solid
oxide fuel cells based on measurements made on cell
materials and components. J Power Sources 2015;291:138e55.
[41] Jung J-I, Misture ST, Edwards DD. The electronic conductivity
of Ba0.5Sr0.5CoxFe1xO3d(BSCF: x ¼ 0 ~ 1.0) under different
oxygen partial pressures. J Electroceram 2010;24:261e9.
[42] Choi S, Yoo S, Kim J, Park S, Jun A, Sengodan S, et al. Highly
efficient and robust cathode materials for low-temperature
solid oxide fuel cells: PrBa0.5Sr0.5Co2xFexO5þd. Sci Rep
2013;3:2426.
Another random document with
no related content on Scribd:
 Municipal Edison. Municipal
Bernstein.

Various Series Bases in Use, 1892.

The above six bases have been superseded by the “Large Edison,” now
called the Mogul Screw base.
 THE EDISON “MUNICIPAL” STREET
LIGHTING SYSTEM

Edison “Municipal” System, 1885.

High voltage direct current was generated,
several circuits operating in multiple, three
ampere lamps burning in series on each circuit.
Photograph courtesy of Association of Edison
Illuminating Companies.

The arc lamp could not practically be made in a unit smaller than
the so-called “1200 candlepower” (6.6 ampere) or “half” size, which
really gave about 350 spherical candlepower. A demand therefore
arose for a small street lighting unit, and Edison designed his
“Municipal” street lighting system to fill this requirement. His
experience in the making of dynamos enabled him to make a direct
current bipolar constant potential machine that would deliver 1000
volts which later was increased to 1200 volts. They were first made
in two sizes having an output of 12 and 30 amperes respectively.
Incandescent lamps were made for 3 amperes in several sizes from
16 to 50 candlepower. These lamps were burned in series on the
1200-volt direct current system. Thus the 12-ampere machine had a
capacity for four series circuits, each taking 3 amperes, the series
circuits being connected in multiple across the 1200 volts. The
number of lamps on each series circuit depended upon their size, as
the voltage of each lamp was different for each size, being about 1½
volts per cp.
A popular size was the 32-candlepower unit, which therefore
required about 45 volts and hence at 3 amperes consumed about
135 watts. Allowing 5 per cent loss in the wires of each circuit, there
was therefore 1140 of the 1200 volts left for the lamps. Hence about
25 32-candlepower or 50 16-candlepower lamps could be put on
each series circuit. Different sizes of lamps could also be put on the
same circuit, the number depending upon the aggregate voltage of
the lamps.

Edison Municipal Lamp, 1885.

Inside the base was an arrangement by which
the lamp was automatically short circuited when
it burned out.
 A device was put in the base of each lamp to short circuit the
lamp when it burned out so as to prevent all the other lamps on that
circuit from going out. This device consisted of a piece of wire put
inside the lamp bulb between the two ends of the filament.
Connected to this wire was a very thin wire inside the base which
held a piece of metal compressed against a spring. The spring was
connected to one terminal of the base. Should the lamp burn out,
current would jump from the filament to the wire in the bulb, and the
current then flowed through the thin wire to the other terminal of the
base. The thin wire was melted by the current, and the spring
pushed the piece of metal up short circuiting the terminals of the
base. This scheme was later simplified by omitting the wire, spring,
etc., and substituting a piece of metal which was prevented from
short circuiting the terminals of the base by a thin piece of paper.
When the lamp burned out the entire 1200 volts was impressed
across this piece of paper, puncturing it and so short circuiting the
base terminals. Should one or more lamps go out on a circuit, the
increase in current above the normal 3 amperes was prevented by
an adjustable resistance, or an extra lot of lamps which could be
turned on one at a time, connected to each circuit and located in the
power station under the control of the operator. This system
disappeared from use with the advent of the constant current
transformer.
 THE SHUNT BOX SYSTEM FOR
SERIES INCANDESCENT LAMPS

Shunt Box System, 1887.

Lamps were burned in series on a high voltage
alternating current, and when a lamp burned
out all the current then went through its “shunt
box,” a reactance coil in multiple with each
lamp.

Soon after the commercial development of the alternating

current constant potential system, a scheme was developed to
permit the use of lamps in series on such circuits without the
necessity for short circuiting a lamp should it burn out. A reactance,
called a “shunt box” and consisting of a coil of wire wound on an iron
core, was connected across each lamp. The shunt box consumed
but little current while the lamp was burning. Should one lamp go
out, the entire current would flow through its shunt box and so
maintain the current approximately constant. It had the difficulty,
however, that if several lamps went out, the current would be
materially increased tending to burn out the remaining lamps on the
circuit. This system also disappeared from use with the development
of the constant current transformer.
 THE ENCLOSED ARC LAMP
Up to 1893 the carbons of an arc lamp operated in the open air,
so that they were rapidly consumed, lasting from eight to sixteen
hours depending on their length and thickness. Louis B. Marks, an
American, found that by placing a tight fitting globe about the arc, the
life of the carbons was increased ten to twelve times. This was due
to the restricted amount of oxygen of the air in the presence of the
hot carbon tips and thus retarded their consumption. The amount of
light was somewhat decreased, but this was more than offset by the
lesser expense of trimming which also justified the use of more
expensive better quality carbons. Satisfactory operation required that
the arc voltage be increased to about 80 volts.
 Enclosed Arc Lamp, 1893.
Enclosing the arc lengthened the life of the
carbons, thereby greatly reducing the cost of
maintenance.

This lamp rapidly displaced the series open arc. An enclosed arc
lamp for use on 110-volt constant potential circuits was also
developed. A resistance was put in series with the arc for use on
110-volt direct current circuits, to act as a ballast in order to prevent
the arc from taking too much current and also to use up the
difference between the arc voltage (80) and the line voltage (110).
On alternating current, a reactance was used in place of the
resistance.
The efficiencies in lumens per watt of these arcs (with clear
glassware), all of which have now disappeared from the market,
were about as follows:
6.6 ampere 510 watt direct current (D.C.) series arc, 8¼ l-p-w.
5.0 ampere 550 watt direct current multiple (110-volt) arc, 4½ l-p-w.
7.5 ampere 540 watt alternating current (A.C.) multiple (110-volt) arc,
4¼ l-p-w.
 Open Flame Arc Lamp, 1898.
Certain salts impregnated in the carbons
produced a brilliantly luminous flame in the arc
stream which enormously increased the
efficiency of the lamp.
 Enclosed Flame Arc Lamp, 1908.
By condensing the smoke from the arc in a
cooling chamber it was practical to enclose the
flame arc, thereby increasing the life of the
carbons.

The reason for the big difference in efficiency between the series
and multiple direct-current arc is that in the latter a large amount of
electrical energy (watts) is lost in the ballast resistance. While there
is a considerable difference between the inherent efficiencies of the
D. C. and A. C. arcs themselves, this difference is reduced in the
multiple D. C. and A. C. arc lamps as more watts are lost in the
resistance ballast of the multiple D. C. lamp than are lost in the
reactance ballast of the multiple A. C. lamp.
This reactance gives the A. C. lamp what is called a “power-
factor.” The product of the amperes (7.5) times the volts (110) does
not give the true wattage (540) of the lamp, so that the actual volt-
amperes (825) has to be multiplied by a power factor, in this case
about 65 per cent, to obtain the actual power (watts) consumed. The
reason is that the instantaneous varying values of the alternating
current and pressure, if multiplied and averaged throughout the
complete alternating cycle, do not equal the average amperes
(measured by an ammeter) multiplied by the average voltage
(measured by a volt-meter). That is, the maximum value of the
current flowing (amperes) does not occur at the same instant that the
maximum pressure (voltage) is on the circuit.
 THE FLAME ARC LAMP
About 1844 Bunsen investigated the effect of introducing various
chemicals in the carbon arc. Nothing was done, however, until
Bremer, a German, experimented with various salts impregnated in
the carbon electrodes. In 1898 he produced the so-called flame arc,
which consisted of carbons impregnated with calcium fluoride. This
gave a brilliant yellow light most of which came from the arc flame,
and practically none from the carbon tips. The arc operated in the
open air and produced smoke which condensed into a white powder.
The two carbons were inclined downward at about a 30-degree
angle with each other, and were of small diameter but long, 18 to 30
inches, having a life of about 12 to 15 hours. The tips of the carbons
projected through an inverted earthenware cup, called the
“economizer,” the white powder condensing on this and acting not
only as an excellent reflector but making a dead air space above the
arc. The arc was maintained at the tips of the carbons by an electro-
magnet whose magnetic field “blew” the arc down.
Two flame arc lamps were burned in series on 110-volt circuits.
They consumed 550 watts each, giving an efficiency of about 35
lumens per watt on direct current. On alternating current the
efficiency was about 30 l-p-w. By use of barium salts impregnated in
the carbons, a white light was obtained, giving an efficiency of about
18 l-p-w on direct current and about 15½ on alternating current.
These figures cover lamps equipped with clear glassware. Using
strontium salts in the carbons, a red light was obtained at a
considerably lower efficiency, such arcs on account of their color
being used only to a limited extent for advertising purposes.
 Constant Current Transformer, 1900.
This converted alternating current of constant
voltage into constant current, for use on series
circuits.

These arcs were remarkably efficient but their maintenance

expense was high. Later, about 1908, enclosed flame arcs with
vertical carbons were made which increased the life of the carbons,
the smoke being condensed in cooling chambers. However, their
maintenance expense was still high. They have now disappeared
from the market, having been displaced by the very efficient gas-
filled tungsten filament incandescent lamp which appeared in 1913.
 THE CONSTANT CURRENT
TRANSFORMER FOR SERIES
CIRCUITS
About 1900 the constant current transformer was developed by
Elihu Thomson. This transforms current taken from a constant
potential alternating current circuit into a constant alternating current
for series circuits, whose voltage varies with the load on the circuit.
The transformer has two separate coils; the primary being stationary
and connected to the constant potential circuit and the secondary
being movable and connected to the series circuit. The weight of the
secondary coil is slightly underbalanced by a counter weight. Current
flowing in the primary induces current in the secondary, the two coils
repelling each other. The strength of the repelling force depends
upon the amount of current flowing in the two coils. The core of the
transformer is so designed that the central part, which the two coils
surround, is magnetically more powerful close to the primary coil
than it is further away.
When the two coils are close together a higher voltage is
induced in the secondary than if the later were further away from the
primary coil. In starting, the two coils are pulled apart by hand to
prevent too large a current flowing in the series circuit. The
secondary coil is allowed to gradually fall and will come to rest at a
point where the voltage induced in it produces the normal current in
the series circuit, the repelling force between the two coils holding
the secondary at this point. Should the load in the series circuit
change for any reason, the current in the series circuit would also
change, thus changing the force repelling the two coils. The
secondary would therefore move until the current in the series circuit
again becomes normal. The action is therefore automatic, and the
actual current in the series circuit can be adjusted within limits to the
desired amount, by varying the counterweight. A dash pot is used to
prevent the secondary coil from oscillating (pumping) too much.
In the constant current transformer, the series circuit is insulated
from the constant potential circuit. This has many advantages. A
similar device, called an automatic regulating reactance was
developed which is slightly less expensive, but it does not have the
advantage of insulating the two circuits from each other.
 ENCLOSED SERIES ALTERNATING
CURRENT ARC LAMPS
The simplicity of the constant current transformer soon drove the
constant direct-current dynamo from the market. An enclosed arc
lamp was therefore developed for use on alternating constant
current. Two sizes of lamps were made; one for 6.6 amperes
consuming 450 watts and having an efficiency of about 4½ lumens
per watt, and the other 7.5 amperes, 480 watts and 5 l-p-w (clear
glassware). These lamps soon superseded the direct current series
arcs. They have now been superseded by the more efficient
magnetite arc and tungsten filament incandescent lamps.
SERIES INCANDESCENT LAMPS ON
CONSTANT CURRENT
TRANSFORMERS
Series incandescent lamps were made for use on constant
current transformers superseding the “Municipal” and “Shunt Box”
systems. The large Edison, now called the Mogul Screw base, was
adopted and the short circuiting film cut-out was removed from the
base and placed between prongs attached to the socket.
 Holder. Socket.

Holder and
socket.

Series Incandescent Lamp Socket with Film Cutout,

1900.
The “Large Edison,” now called Mogul Screw, base was standardized and
the short circuiting device put on the socket terminals.
 The transformers made for the two sizes of arc lamps, produced
6.6 and 7.5 amperes and incandescent lamps, in various sizes from
16 to 50 cp, were made for these currents so that the incandescent
lamps could be operated on the same circuit with the arc lamps. The
carbon series incandescent lamp, however, was more efficient if
made for lower currents, so 3½-, 4- and 5½-ampere constant current
transformers were made for incandescent lamps designed for these
amperes. Later, however, with the advent of the tungsten filament,
the 6.6-ampere series tungsten lamp was made the standard, as it
was slightly more efficient than the lower current lamps, and was
made in sizes from 32 to 400 cp. When the more efficient gas-filled
tungsten lamps were developed, the sizes were further increased;
the standard 6.6-ampere lamps now made are from 60 to 2500 cp.
 THE NERNST LAMP
Dr. Walther Nernst, of Germany, investigating the rare earths
used in the Welsbach mantle, developed an electric lamp having a
burner, or “glower” as it was called, consisting of a mixture of these
oxides. The main ingredient was zirconia, and the glower operated in
the open air. It is a non-conductor when cold, so had to be heated
before current would flow through it. This was accomplished by an
electric heating coil, made of platinum wire, located just above the
glower. As the glower became heated and current flowed through it,
the heater was automatically disconnected by an electro-magnet cut-
out.
 Nernst Lamp, 1900.
The burners consisted mainly of zirconium
oxide which had to be heated before current
could go through them.

The resistance of the glower decreases with increase in current,

so a steadying resistance was put in series with it. This consisted of
an iron wire mounted in a bulb filled with hydrogen gas and was
called a “ballast.” Iron has the property of increasing in resistance
with increase in current flowing through it, this increase being very
marked between certain temperatures at which the ballast was
operated. The lamp was put on the American market in 1900 for use
on 220-volt alternating current circuits. The glower consumed 0.4
ampere. One, two, three, four and six glower lamps were made,
consuming 88, 196, 274, 392 and 528 watts respectively. As most of
the light is thrown downward, their light output was generally given in
mean lower hemispherical candlepower. The multiple glower lamps
were more efficient than the single glower, owing to the heat radiated