Download Will High Power Impulse Magnetron Sputtering (HIPIMS)

Will High Power Impulse Magnetron Sputtering (HIPIMS) Be the Right Technique
for Nb/Cu Coated 1.5 GHz Superconducting Cavities?
Meng Yin1,2,3, O. Azzolini2, G. Keppel2, C. Pira2, A. Rossi2, V. Palmieri2.
1
Università di Padova, Padova, Italy. 2 INFN, Laboratori Nazionali di Legnaro, Legnaro (Padova), Italy.
3
China Institute of Atomic Energy, Beijing, China.
The main advantage is the control over energy and
direction of the deposition material. In some sense,
HIPIMS can be seen as a method that combines the
advantages of conventional magnetron sputtering and arc
evaporation. It produces highly ionized droplet free
plasma.
The plasma conditions in HIPIMS can be used for a resputtering of the growing film through the bombardment of
the ions of the target material itself. The process result is
dense, droplet free films. From the point of view of
accelerator techniques, coatings produced with HIPIMS
are highly interesting as it has been proven that the film
properties are enhanced.
The main problems with the current version of HIPIMS
are reduction of deposition rate and the transition to an arc
discharge. The deposition rate for the HIPIMS discharge is
expected to be in the range 30-80 % compared to a
conventional magnetron discharge, with the same average
power. The other problem with HIPIMS is the arcing
tendency on the cathode surface, especially for targets with
low melting point (luckily this is not the case of Nb).
INTRODUCTION
In order to save building costs of superconducting
resonant cavities for particle accelerators, the development
of the sputtering technique for the deposition of niobium
(Nb) superconducting films onto copper cavities has
started in the 1980’s at CERN [1]. However high energy
gradients remain unachievable when using thin films. The
main limit of the Nb/Cu cavity is the Q-slope, that makes
the Q factor decreasing at high accelerating fields.
Exploring new sputtering configurations is one of the
principal way to find a solution to the Q-slope problem.
High Power Impulse Magnetron Sputtering (HIPIMS) is
an evolution of the magnetron technique which relies on
100 μs high voltage pulses of the order of 1 kV compared
to the 300 V of the standard DC magnetron process [2].
In this work, an R&D effort has been undertaken to
study the HIPIMS, to improve it and understand the
correlation between the sputtering parameters and the film
morphology, the superconducting properties and the RF
film quality.
HIPIMS
The common way of generating plasma for material
processing by magnetron sputtering is to use two metal
electrodes: a cathode and an anode (usually the grounded
chamber walls) enclosed in an evacuated vacuum chamber.
In order to generate the plasma a source of power is
needed, such as a direct current (DC) power supply.
Sputtering film technique has the following drawbacks:
1. the working gas is trapped in the film; it may cause
intrinsic defects inside of the grain. The impurities of
the working gas are not good for the thin film;
2. the deposition energy is low, which does not help to
avoid columnar grains.
So, we investigate HIPIMS to avoid these drawbacks.
HIPIMS is a method for physical vapor deposition of
thin films which is based on magnetron sputter deposition.
HIPIMS utilizes extremely high power densities of the
order of kWcm−2 in short pulses (impulses) of tens of
microseconds at low duty cycle (on/off time ratio) of
<10 %. A distinguishing feature of HIPIMS is its high
degree of ionization of the sputtered metal and high rate of
molecular gas dissociation [3].
In HIPIMS there are three principal parameters that
drive the process: Voltage (V), Frequency (F) and Pulse
Time (PT), as it can be seen in figure 1.
LNL Annual Report
Fig. 1. HIPIMS cycle.
EXPERIMENTAL SYSTEM
For our experiments we used a deposition system where
the target is a niobium post magnetron with an external
source of magnetic field, instead of a cylindrical
magnetron like in the standard CERN configuration [4]. To
make a deposition in this system requires two power
supplies, one for the cathode and one to feed the coil. The
compact vacuum system is capable of moving inside and
outside a big magnetic coil. The cathode is composed by a
Niobium post magnetron (magnetron diameter 43 mm,
103
Applied, General and Interdisciplinary Physics
increases with Frequency. No correlation with PT.
SUPERCONDUCTIVE PROPERTIES – RRR. The
superconducting properties of the samples are not good,
yet. Both the Tc and RRR are lower than needed; also our
standard (coated in DC mode) presents RRR lower than
10, and Tc less than Tc of bulk Nb. This means that the
problem is caused by the sputtering system and not by the
technique. Despite this, we have found some interesting
correlation between the superconductive films RRR and
HIPIMS process V and F.
RRR reaches a maximum for 800 V, then decreases at
very low values for 1000 V. RRR decreases linearly with
Frequency. This behavior is inverse with respect to grain
dimension. We have not found a clear relationship between
pulse time and RRR.
wings diameter 65 mm) fixed in the centre of the cavity
through a water cooled steel tube. For the deposition we
used quartz samples fixed in a sample holder with the same
shape of the cavity. In this first experiment we have
deposited Nb thin film only in the centre of the cavity (the
iris). The ultimate pressure is in the order of 10-9 mbar.
Sputtering pressure is 1⋅10-2 mbar in Ar. The sketch and
the photo of the used set-up are shown in figure 2.
CONCLUSIONS
a)
At the end, we will try to answer to the difficult question
presented in the title of this report.
We did not arrive to high values of RRR for our
samples: maximum values were around 9 (normally,
cavities have RRR higher than 20). But also in DC mode
we obtained the same results in the same system, so this
implies that it is not a problem of technique, but there is
something wrong in our deposition system. The electron
bombardment, the uncooled cathode and the high
temperature reached by the cavity during the process are
the principal candidates to explain the film pollution.
On the other hand, we found that HIMPIMS parameters
(in particular voltage and frequency) influence the
microstructure and the RRR of the coated films. This
means that if we can find the right deposition parameters
we would obtain a high quality superconductive film.
At the moment the goal is to identify and solve the
problem of pollution of the film, in order to obtain high
RRR values. Only after having obtained high purity films,
HIPIMS can be studied deeply and used for deposition of
the entire surface of the cavity. After that, the second stage
of optimization of the deposition will be possible in order
to deposit a copper cavity and measure the features.
b)
Fig. 2. 3D image (a) and photo (b) of the deposition system.
EXPERIMENTAL SYSTEM
The Nb films was characterized by X-ray diffraction
(XRD), Profilometer, measure of Residual Resistivity
Ratio (RRR) and Critical Temperature (Tc).
We correlate deposition rate, film microstructure and
superconductive properties, to the three principal
parameters of HIPIMS process: Voltage (V), Frequency
(F) and Pulse Time (PT).
DEPOSITION RATE. The results of our experiments is
that the deposition rate increases with the increasing of
values of each of these three parameters: voltage,
frequency and pulse time increase. V, F and PT.
In particular the relationship with F and PT are linear,
whereas increase in deposition rate with V goes
exponentially.
RETICULAR PARAMETER. For voltage lower than
800 V, the reticular parameters (a) is lower than bulk Nb
reticular parameter. When voltage is 1000 V, the reticular
parameters (a) is higher than bulk Nb reticular parameter.
This means that for high sputtering potential Nb films pass
from a compressive stress to a tensile stress.
GRAIN DIMENSION. A maximum in grain dimension
when Voltage is 800 V. Grain dimension also exponentially
LNL Annual Report
[1]
[2]
[3]
[4]
104
C. Benvenuti et al., Physica, C 316 (1999) 153-188.
V. Kouznetsov, Swedish Patent 525231, (2005).
V. Kouznetsov, Patent WO 2005/005684 A1, (2005).
C. Benvenuti et al., Physica, C 351 (2001) 421.
Applied, General and Interdisciplinary Physics