Pulsed laser deposition (PLD)
is currently the favored method for scaled up production
of High Temperature Superconductor (HTS) tapes. Common
PLD equipment consists of high pulse energy lasers
running at low repetition rates and more or less static
targets. A common belief among developers of HTS tapes
is that PLD can never be cost-effective for commercial
HTS tape manufacture because of the expense of
purchasing and operating the excimer lasers used in the
process. However, the latest advances in high-power
industrial excimer laser design, combined with High-Rate
PLD, are reducing the manufacturing cost of HTS tape
production.
Excimer laser PLD is known
to be an excellent technique for manufacturing of
complex HTS films with high critical current density (Jc)
larger than 20MA/cm2. The most common
wavelengths for this kind of PLD are 248 and 308nm.
Research-grade excimer lasers are economic tools that
offer a choice of high pulse energy at moderate
repetition rates. Since PLD requires a target energy of
around 2J/cm2 to ablate YBCO, the preferred
tool for PLD research is a laser that features high
pulse energy. Such lasers may deliver stabilized pulse
energies up to 600mJ at 248nm, while offering a maximum
pulse repetition rate of only 50Hz. Resulting deposition
rates are on the order of less than 2nm-m2/h.
However, this modest pulse deposition rate is generally
not an issue, since throughput is not critical in a
research setting.
The situation is completely different for volume
production of HTS tape where deposition rates of greater
than 30nm-m2/h are required. In order to
maximize throughput, volume production demands lasers
that offer both high pulse energy and high pulse
repetition rate.
In 2001, Hans Christen, at the
Materials Science and Technology Division & Center
for Nanophase Materials Sciences at Oak Ridge National
Lab, together with engineers at Coherent, performed
detailed calculations of the costs for these pulses and
translated these costs into a laser cost per length of
tape. The calculations were based on the results of
several independent studies that both showed that an
excimer such as the laser mentioned above can produce
1cm wide tape with a 1 micron YBCO layer at the
approximate rate of 37m per kilowatt hour of laser
power. Christen estimated the laser running costs at
$228 per kW-hr and the cost of servicing personnel at
between $4 and $8 per kW-hr. In addition, capital costs
for the machines per kW-hr is somewhere between $20 and
$100 depending on the lifetime (depreciation period) of
the laser.
This gives us a cost range
of between $6.81 and $9.08 per meter. Assuming that
these wires can be produced with critical transport
currents (Ic) of 400A/cm-width, laser costs
would amount to between $17 and $22.70 per kA-m.
These costs are being
aggressively reduced by laser design innovations that
extend the lifetimes of all consumables, which include
electricity, gas, replacement optics, replacement
thryatrons (used to switch the high laser voltage), and
replacement laser tubes.
By improving clean room
manufacturing processes, Coherent has improved laser
tube life. With our new LAMBDA SX series of high power
industrial excimer lasers, laser tube lifetimes of up to
6 billion pulses, and gas lifetimes of up to two weeks
in two day shift operation, have been achieved. Another
important advance is the elimination of the thyratron
high voltage switch. As noted by Christen, the typical
lifetime for this high voltage switch is around 2
billion pulses.
This translates into 10
thyratron replacements per year at 20 billion
pulses/year. In one of our new lasers, this technology
has been replaced with a maintenance free
all-solid-state switch with a lifetime of several years.
This eliminates the significant cost of thyratrons, as
well as the downtime associated with their replacement.
By extending component lifetimes and increasing output
power in these ways, the $228 per kW-hr estimated by
Christen in 2001 can be lowered to $110 per kW-hr today.
Other advances in laser
technology have been targeted at improving the laser
performance, and hence PLD yield. These include real
time pulse-to pulse active stabilization which has cut
pulse-to-pulse variation from typically 0.9% to 0.5% (1
sigma) for 308nm. Additional innovations include
proprietary techniques, such as our Powerlok and Timelok
technologies, which ensure this stable laser performance
is maintained even in burst operation and on fast moving
targets. This is important because the Ic of
a length of HTS tape will only be as high as the lowest
value. In theory, a bad tape section can be identified
and removed, but joining coated conductors is extremely
complex and needs to be avoided in a commercial setting.
Today’s high power excimers
deliver up to 1050mJ/pulse (stabilized at 308nm) with a
maximum pulse repetition rate of 300Hz. These lasers,
such as the
LAMBDA SX 315C, are operated already in the
demanding 24 hours/day, 7 days per week production of
low-temperature polycrystalline silicon (LTPS) for LCD
and OLED displays. This corresponds to 22 to 25 million
pulses per day, or 7.5 to 8.5 billion pulses per year,
depending on idle times and system utilization.
Given such improvements, we
estimate the laser costs of 400A/cm-width HTS tape today
are between 9.05 and 14.72 cents per kA-m.
Improved processing
techniques specific to PLD for YBCO coated conductor
manufacture are further reducing costs. Alexander
Usoskin, at
European High Temperature Superconductors, has
developed high throughput PLD (HR-PLD) device for YBCO
film manufacturing. The set up consists of a scanning
mirror, a tape transportation mechanism, and a movable
target where the laser beam is scanned across.
Furthermore, the transportation mechanism employs a
Quasi-Equilibrium-Heating (QEH) method which provides a
temperature control of +/-2°C for the substrate tapes.
This equipment was used to
coat YBCO on stainless-steel tapes with a bi-axially
textured yttria-stabilized ZrO2 buffer layer.
Compared to conventional PLD, a very homogeneous YBCO
microstructure was achieved. The good film morphology
results in a high Jc of 2.3MA/cm2
and a deposition rate of 40nm-m2/hr using a
150W LAMBDA excimer laser. Further development of the
150W configuration has subsequently resulted in a 70nm-m2/hr
achievement. Hence, with a 315W laser such as the LAMBDA
SX 315C the throughput of HR-PLD can be much above this
latter value.
PLD currently represents the
most effective method for production of HTS tapes with
high current densities. The development of 315W excimer
lasers and HR-PLD equipment makes PLD increasingly
economical. In addition, design improvements of
industrial excimer lasers have lowered the total cost of
ownership of excimer lasers significantly. Furthermore,
specific process setups have been reported that show
very high efficiency in terms of the HTS film
production, and that will lead to high throughput and
low cost.
