by user

Category: Documents





Journal of Nuclear Materials 452 (2014) 51–56
Contents lists available at ScienceDirect
Journal of Nuclear Materials
journal homepage: www.elsevier.com/locate/jnucmat
Damage studies on tungsten due to helium ion irradiation
N.J. Dutta, N. Buzarbaruah, S.R. Mohanty ⇑
Centre of Plasma Physics-Institute for Plasma Physics, Sonapur, Kamrup 782402, India
h i g h l i g h t s
Used plasma focus helium ion source to study radiation induced damage on tungsten.
Surface analyses confirm formation of micro-crack, bubbles, blisters, pinholes, etc.
XRD patterns confirm development of compressive stress due to thermal load.
Reduction in hardness value is observed in the case of exposed sample.
a r t i c l e
i n f o
Article history:
Received 27 January 2014
Accepted 23 April 2014
Available online 5 May 2014
a b s t r a c t
Energetic and high fluence helium ions emitted in a plasma focus device have been used successfully to
study the radiation induced damage on tungsten. The reference and irradiated samples were characterized by optical microscopy, field emission scanning electron microscopy, X-ray diffraction and by hardness testers. The micrographs of the irradiated samples at lower magnification show uniform mesh of
cracks of micrometer width. However at higher magnification, various types of crystalline defects such
as voids, pinholes, bubbles, blisters and microcracks are distinctly noticed. The prominent peaks in
X-ray diffraction spectrum of irradiated samples are seen shifted toward higher Bragg angles, thus indicating accumulation of compressive stress due to the heat load delivered by helium ions. A marginal
reduction in hardness of the irradiated sample is also noticed.
Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction
Materials that are proposed to be used as plasma facing component (PFC) in future fusion devices must work in adverse environment that includes combinations of high temperatures, intense
damaging radiations, heat and mechanical loadings, reactive chemicals, etc. Therefore, the task of constructing and operating fusion
reactors with materials that survive the aforesaid adverse environment and meet the objectives for safety, environment and performance is still very challenging. The development of the new
materials for such reactors would require fundamental understanding of processes like radiation damage, plasma-surface interactions, sputtering, corrosion and erosion [1]. Tungsten has been
considered as one of the most important material as a PFC in the
divertor and baffle regions of a tokamak because of its intrinsic
advantages such as high atomic number, high melting point, high
threshold energy towards sputtering and good thermomechanical
properties [2]. The PFC of fusion reactors are in generally prone
to damage due to its direct contact with the edge plasma and high
particle flux of neutrals, ions and impurity particles possessing
⇑ Corresponding author. Tel.: +91 9435464759.
E-mail address: [email protected] (S.R. Mohanty).
0022-3115/Ó 2014 Elsevier B.V. All rights reserved.
broad energy spectrum. Hence, in recent years, there has been an
increasing interest in testing radiation induced damage on fusion
reactor materials through deployment of small scale laboratory
experiments [3–8]. During the last decade, some notable experiments were carried out to understand the radiation effect on fusion
materials using divertor simulator [3], plasma interaction with surface and components experimental simulator [4], high flux ion
beam test device [5], inertial electrostatic confinement fusion
device [6], plasma focus [7,8], etc. Formation of surface defects
such as bubbles, microcracks, blisters, voids and nanostructures
on tungsten due to irradiation with helium ions having energy of
few eV to keV have been reported in literature [9–12]. Kajita
et al. [13] observed the formation of nanofibers on tungsten surface
when irradiated with 75 eV helium ions. While studying the
response of polycrystalline tungsten to 30 keV helium ions, Zenobia et al. [10] observed the formation of a highly oriented grass like
structure for a certain fluence of helium ions and for a specific substrate temperature. In case of exposure under moderate energy (a
few keV) helium ions on tungsten surface, the changes mainly
occur in terms of hole and bubble formation which critically
depends on the exposure temperature [12]. This reduces the
strength of surface and causes the material to release impurities
in response to transient heat load such as edge localized modes
N.J. Dutta et al. / Journal of Nuclear Materials 452 (2014) 51–56
[14]. A systematic study of helium retentions and surface blistering
in terms of helium dose, temperature, pulse implantation and
tungsten microstructure was conducted by Gilliam et al. [15]. They
reported that maintaining the helium ion flux per fusion event
below a threshold level may reduce the radiation damage effect
and further stressed the necessity to investigate the effect of variable helium ion energy and simultaneous lattice damage on helium
retention and blistering.
On the background of the investigations reported above, we
have studied here the effect of energetic helium ions on tungsten
using an ingenious plasma focus (PF) device. The low energy PF
device is a simple and cost effective pulsed coaxial plasma accelerator that makes use of the self-generated magnetic field for compressing the plasma to a very high density (1025–1026 m3) and
high temperature (>1 keV). Being a rich source of charged particles
and electromagnetic (EM) radiations, it has been successfully used
in numerous applications namely as X-ray source of microlithography [16] and microscopy [17], electron source for microlithography [18], X-ray interstitial radio surgery [19], EUV source for EUV
lithography [20] and as ion source for processing of materials
[21] and thin film deposition [22]. The uniqueness of a PF device
is that it produces pulsed ion beams with highly charged states
having wide range of energies and it does not require any external
confinement system or extraction grid. Generally, this device emits
ions of energies in the ranges from a few keV to a few MeV depending upon its operating energy [23]. Using such a PF device, Bostick
et al. [24] studied blister formation on various metallic plates (Al,
Cu and Si) by using deuteron beams liberated from a PF device
and obtained an information about the fine structure of the deuterium beam by making statistical analyses of blister parameters. On
the other hand, Pimenov et al. [25] investigated the damage
induced on tungsten plates by using axially emitted high energy
ion beam and dense plasma stream at different capacitor banks
energy. Due to the anisotropic nature of ion emission from a PF
device, it would be better to study the ion irradiation on materials
not only at axial position but also at different angular positions [7].
In this paper, therefore, we have investigated the main features
of damages that occurred on tungsten material when irradiated by
wide spectrum helium ion pulses that emanated from our PF
device. Ex situ characterization of irradiated as well as reference
materials has been carried out using optical microscopy, field
emission scanning electron microscopy, X-ray diffraction and
hardness tester.
2. Experimental procedures
The helium ion source employed in the present investigation is
basically a 2.2 kJ Mather geometry PF device which is powered by a
low inductance 7.1 lF high energy storage capacitor charged up to
25 kV. A schematic drawing of the device is given in Fig. 1. The
storage capacitor is connected by a low inductance spark gap to
the co-axial and concentric electrode system which is enclosed in
a stainless steel chamber having helium gas filling at pressures of
several mTorr. The inductance of the capacitor along with the
external connections and the switch are about 130 nH. The details
of the device can be seen elsewhere [26,27]. The electrical energy
of the capacitor bank is first transferred to the electrodes and, as
a result, the gas breakdown occurs across the electrodes forming
the current sheaths that propagate along the electrodes by J B
Lorentz force. When the current sheaths reach at the open end of
the electrodes, the sheaths undergo radial pinching forming a
highly dense plasma column. This plasma column is short lived
and unstable. Moreover, due to instabilities it breaks up and emits
X-rays, energetic ions and electrons, and neutrons [26]. The
ions emitted from the plasma column are basically instability
Fig. 1. Schematic of experimental set up.
accelerated ions having energy range of tens of keV to MeV and
moves mostly towards the top of the chamber. The ion density distribution, fluences were measured at a height of 6 cm from top of
the anode using ion collector [28]. It is well-known that operating
gas type does not affect many of the key characteristics such as
plasma density, plasma temperature and ion spectra of an optimized PF device [29] and therefore, for a single PF shot, we have
considered the expected ion fluence and flux at the location of
sample as 1018 m2 and 1025 m2 s1, respectively.
The thermal evolution of samples during and after ion irradiation is studied mainly theoretically and reported in literatures
[30,31]. Since in this experiment, the ion beams interacted with
the sample for a few hundred of nsec, therefore, it is observed that
the sample temperature variation is transient in nature with extremely high temperature rise rate (40 K/nsec) followed by fast
cooling [31]. Exact measurement of thermal evolution of sample
during ion bombardment requires a detector or sensor having time
resolution superior to 200 nsec which is impossible from technological point of view. Therefore, we have made use of a thermocouple and found that the bulk temperature of samples varies from 70
to 100 °C depending upon ion pulses.
High purity (99.95 wt.%) polycrystalline tungsten plates were
procured from Goodfellow, United Kingdom. The concentrations
of impurities in tungsten material as determined by the manufacturer are given in Table 1. The plates were cut into square sizes
(typically 5 5 and 10 10 mm2) with the help of a precession
cutter and mechanically polished in a Buehler make polisher [7]
using different grits (180, 240, 360, 420, 600 and 1200) of abrasive
paper. Samples were then polished with a nylon cloth for getting a
mirror finish surface. The polished tungsten samples were then
introduced into the PF chamber and mounted axially and also in
other angular positions with the help of a movable sample holder
placed at a distance 6 cm from the top of the anode. The samples
were irradiated at room temperature to single and multi PF shots
(i.e., pulses of helium ions) keeping helium filling gas pressure at
0.5 Torr inside the chamber. Inclination of samples with respect
to ion trajectories was kept perpendicular in our case. This fact
has been ascertained by observing the mostly circular ion tracks
N.J. Dutta et al. / Journal of Nuclear Materials 452 (2014) 51–56
Table 1
Concentrations of impurities in tungsten material.
Content (ppm)
recorded on nuclear track detectors while investigating the spatial
distribution of ions of source [28].
The irradiated and reference tungsten samples were characterized by Olympus BX51M optical microscopy (OM), Sigma Zeiss
field emission scanning electron microscopy (FESEM) for ascertaining the changes in their surface morphology. Similarly, structural
analyses and hardness measurements of the reference and the irradiated samples were carried out by employing Bruker D8 X-ray diffraction (XRD) system and Buehler Micromet 2101 microhardness
tester, respectively.
3. Results and discussion
The energetic ions from the device were allowed to impinge on
the tungsten samples which were kept at different angular positions as mentioned earlier. After getting irradiated by helium ions,
a clear difference in surface morphology is observed visually. The
smooth, shiny and bright surface of reference sample got transformed into uneven surface with dull appearance and reflecting
yellowish gray color. When viewed under the optical microscope
to investigate the exact surface morphological changes caused
due to helium ion irradiation, a significant difference has been
marked on the surface morphology. As evident from the micrograph shown in Fig. 2a, the reference sample is quite smooth without any major surface imperfections. On the contrary, the
irradiated samples reveal a uniform network or mesh of cracks that
appeared in the samples that were kept at axis or nearby to it. The
cracks are of micron size width (1–4 lm) having primary as well
as secondary branches as illustrated in Fig. 2b and c. Moreover,
interconnected cracks or loop like patterns are mainly observed
either in the samples kept at axis and close to it or the samples
irradiated by larger number of PF shots. In some irradiated samples, bubble formation is seen near the cracks which are mainly
due to diffusion of helium ions in tungsten and subsequent trapping in bubbles.
In order to study the detailed damage pattern at higher magnification, both the irradiated and reference samples were viewed
under FESEM. While observing the surface texture of reference
sample one can notice distinct tiny microstructures arranged in
clusters that are similar to dendrite type of structure as shown in
Fig. 3a. These clusters get smeared out due to ion bombardment
as evident from the micrograph depicted in Fig. 3b. Moreover, at
certain places small cracks appeared in reference sample which
are produced during the mechanical polishing of the sample. But
these crack patterns are quite different from the patterns that are
observed after ion irradiation as shown in Fig. 3c. In our study,
microcracks having width of a few hundred nanometer to micrometer are noticed in irradiated samples. Such type of cracks on tungsten surface is also reported earlier [7,8]. Shirokova et al. [8]
observed similar type of microcracks patterns on tungsten and in
doped tungsten samples by bombarding these samples with deuterium plasma. Microcracks patterns that are quite similar to the pattern obtained in the present study are also noticed on tungsten
surface by Bhuyan et al. [7] when such samples were irradiated
with proton beams. A look at the exposed samples (Figs. 4 and 5)
at higher magnification clearly indicates various crystalline defects
such as voids, pinholes, bubbles, blisters and microcracks which
arise due to multiple heat load on tungsten surface by helium
ion pulses. It is reported elsewhere that 500 eV helium ions can
Fig. 2. Optical micrographs of samples: (a) reference, (b) irradiated to 5 pulses of
helium ions and (c) irradiated to 10 pulses of helium ions.
displace atoms from tungsten crystal [2]. Since in the present study
tungsten is bombarded with helium ions of much higher energy,
the vacancies and interstitials are created in the crystal structure.
Successive exposure of the samples to helium ion pulses results
in the migration of interstitials that leave vacancies behind. Subsequently, these vacancies coalesce and form small voids which
N.J. Dutta et al. / Journal of Nuclear Materials 452 (2014) 51–56
Fig. 3. FESEM images of (a) reference sample, (b) irradiated sample showing
smearing of clusters and (c) irradiated sample indicating cracks.
Fig. 4. FESEM image of irradiated sample with magnified view of blisters.
afterward allow some helium ions to go into the lattice resulting in
the formation of helium bubbles. It is observed that the number of
radiation induced defects is comparable to the number of helium
ions implanted into the surface [32]. As the estimated ion density
per ion pulse impinging on the sample is of the order of 1019–1020/
m3 [23,28], a high density of defects namely pinholes and voids
have been noticed in samples which were irradiated to single ion
pulse. Other defects such as voids, pinholes, cracks bubbles and
blisters. are observed in samples irradiated by multi ion pulses (5
and 10 pulses). Therefore, the experiment reasonably establishes
that ion fluences on tungsten are primary controlling factor for
the defect formation. Moreover, the exposed sample temperature
is a key factor in controlling parameters like the helium diffusion
rate, thermal vacancy density and thermal migration for the bubble formation and growth [32]. It may be noted that in the process
of PF ion bombardment no separate heating arrangement is needed
since the heat deposited by the single ion pulse is enough to raise
the temperature of near surface layer of sample up to a few hundred degrees centigrade [33]. Nevertheless, in the case of the samples irradiated by 5 and 10 pulses of helium ions, it is expected that
due to the multi shot exposure the near surface layer of sample
must have attained much higher temperature of the order of thousand degrees centigrade. Owing to successive ion pulses bombardment, the migration of vacancies and helium bubbles occurs along
thermal gradient which results in the growth of bubbles and inter
bubble fracture. It is understood that formation of some micropath takes place along the inter bubble cracks which collectively
coalescence resulting in the formation of surface blisters as shown
in window of Fig. 4 [34]. The formation of subsurface cracks can be
interpreted from the inter bubble fracture model proposed by
Evans [34] according to which the bubbles grow due to increase
in the internal pressure, which decreases their spacing. Eventually
several adjacent bubbles will create a local stress that is sufficient
to form subsurface cracks and these may lead to the formation of
surface cracks as shown in Fig. 4. The formation of holes and bubbles decreases drastically for the samples placed at off axis which is
mainly due to lesser heat load deposited by ion beams [7]. As illustrated in Fig. 5, growth of nanoflakes (having 28 nm diameter and
191 nm length) and nanoglobules structures have been also
observed inside some pinholes and microcracks, respectively. The
formation of these nanoflakes and nanoglobules structures may
be due to the reduction in surface heat deposition around the wall
of pinholes and microcracks by ion pulses. It is widely reported
that the bombardment of low energy helium ions on tungsten
surface favors the formation nanofibers [13].
In order to ascertain the detailed structural changes that have
occurred due to the helium ion irradiation, both reference and irradiated tungsten samples were analyzed using XRD. The diffraction
patterns were recorded from 5° to 90° at a scan rate of 0.0199° and
the result is shown in Fig. 6a. Sharp peaks at 2h equals to 40.21°,
58.22°, 73.17° and 86.94° appear in the spectrum of the reference
sample that are recognized as (1 1 0), (2 0 0), (2 1 1) and (2 2 0)
planes of the tungsten. These peaks correspond to the body centered cubic crystal structure of the tungsten. Another feeble peak
centered at 25.54° is also noticed in the spectrum of the reference
sample. The peak is seen to become prominent when the sample
was irradiated to helium ion pulses (Fig. 6b). Moreover, the four
sharp peaks that appeared in reference sample also appear in the
spectrum of the irradiated sample but these peaks get shifted
towards higher Bragg angles. On the contrary, the feeble peak
observed in the irradiated sample gets shifted towards lower Bragg
angles in comparison to the reference sample. The feeble peak at
(1 1 2) plane [35] may be attributed to the formation of tungsten
oxide (WO3) in both the samples and the intensity is seen to have
increased in the irradiated sample due to the inclusion of more
impurity ions (namely oxygen coming out from the chamber) into
tungsten sample. The possible explanation of shifting of the peaks
towards the higher Bragg angles is due to build up of ion irradiation compressive internal stress in the exposed samples [36]. We
N.J. Dutta et al. / Journal of Nuclear Materials 452 (2014) 51–56
Fig. 5. FESEM image of irradiated sample with magnified view of nanoflakes and nanoglobules structures.
have also calculated the average grain size of both the samples by
using Scherrer’s formula as given below,
b2h cos h
where D is the grain size, k is wavelength of X-ray, b2h is FWHM of
the plane, and h is Bragg’s angle. It is found that the average grain
size of the sample decrease from 31.26 nm to 27.50 nm due to the
bombardment of helium ions.
Thermal load due to helium ion exposure of tungsten sample is
seen to change the hardness of sample surface. In order to determine the surface hardness, we employed Vickers hardness tester
with 500 g load to estimate hardness values at five different positions on reference as well as on the exposed samples. The microhardness values so obtained are given in Table 2. A marginal
reduction in hardness values is observed at certain positions in
case of the irradiated sample than the reference one. This is
because of the formation of microcracks, pinholes and other
defects in the irradiated samples.
4. Conclusions
Fig. 6. XRD patterns of reference and irradiated sample having 2h value (a) 5–90°
and (b) 5–30°.
Table 2
Microhardness of samples.
Hardness, HV0.5
Hardness measured at five different places along the samples.
From the above investigations it may be concluded that the
helium ions emanated from our PF device can be successfully utilized to study the ion induced damage to the PFC material namely
tungsten. The experiment conclusively establishes the difference in
surface morphology of the irradiated and reference samples when
exposed with helium ions. Optical micrographs of the irradiated
sample exhibit a uniform mesh of cracks having width around 1–
4 lm that spreads throughout the exposed surface. FESEM pictures
of irradiated sample clearly indicate the different types of crystalline defects such as voids, pinholes, microcracks, bubbles and blisters which seems to have appeared due to ion bombardment.
Moreover, formation of nanoflakes and nanoglobules structures
has also been observed at certain places. The development of crystalline defects decreases significantly in the case of the samples
irradiated at off axis region because of lesser heat load deposition
by helium ion pulses. In the XRD spectrum of irradiated samples,
the major peaks exhibit a shift towards the higher Bragg angles
indicating build up of compressive stress due to heat load. A marginal reduction in hardness value of irradiated sample is noticed
which may be due to formation of crystalline defects.
N.J. Dutta et al. / Journal of Nuclear Materials 452 (2014) 51–56
To summarize, the experiment has provided adequate firsthand
information as to how tungsten would suffer changes in its surface
morphology if used as PFC in next generation fusion reactor.
The authors are grateful to the Director, Institute for Plasma
Research, Gandhinagar, India and Centre Director, CPP-IPR, Sonapur, India for supporting to carry out the present work. The authors
are also thankful to Mr. M. Bora of IIT, Guwahati, India and Dr. N.
Adhikary of IASST, Guwahati, India for their help in characterization of materials. The technical assistance of M.K.D. Sarma is
greatly appreciated.
G. Federici et al., Nucl. Fusion 41 (2001) 1967–2137.
O. El-Atwani, M. Efe, B. Heim, J.P. Allain, J. Nucl. Mater. 434 (2013) 170–177.
M. Yamagiwa et al., J. Nucl. Mater. 417 (2013) 499–503.
S. Kajita, N. Yoshida, R. Yoshihara, N. Ohno, M. Yamagiwa, J. Nucl. Mater. 418
(2011) 152–158.
Y. Ueda, H. Kikuchi, T. Shimada, A. Sagara, B. Kyoh, M. Nishikawa, Fusion Eng.
Des. 61–62 (2002) 255–261.
B.B. Cipiti, G.L. Kulcinski, J. Nucl. Mater. 347 (2005) 298–306.
M. Bhuyan, S.R. Mohanty, C.V.S. Rao, P.A. Rayjada, P.M. Raole, Appl. Surf. Sci.
264 (2013) 674–680.
V. Shirokova et al., J. Nucl. Mater. 435 (2013) 181–188.
N. Yoshida, H. Iwakiri, K. Tokunaga, T. Baba, J. Nucl. Mater. 337–339 (2005)
J. Zenobia, L.M. Garrison, G.L. Kulcinsik, J. Nucl. Mater. 425 (2012) 83–92.
A. Bailini et al., Appl. Surf. Sci. 253 (2007) 8130–8135.
H. Iwakiri, K. Yasunaga, K. Morishita, N. Yoshida, J. Nucl. Mater. 283–287
(2000) 1134–1138.
S. Kajita, W. Sakaguchi, N. Ohno, N. Yoshida, T. Saeki, Nucl. Fusion 49 (2009)
S. Kajita, N. Ohno, W. Sakaguchi, M. Takagi, Plasma Fusion Res. 4 (2009) 004.
S.B. Gilliam et al., J. Nucl. Mater. 347 (2005) 289–297.
R. Petr et al., Rev. Sci. Instrum. 75 (2004) 2551–2559.
R. Feder, J.S. Pearlman, J.C. Riordan, J.L. Costa, J. Microsc. 135 (1984) 347–351.
P. Lee, X. Feng, G.X. Zhang, M.H. Liu, S. Lee, Plasma Source Sci. Technol. 6
(1997) 343–348.
A. Tartari, A.D. Re, F. Mezzetti, E. Angeli, P.D. Chiara, Nucl. Instrum. Methods
Phys. Res., B 213 (2004) 607–610.
S.R. Mohanty et al., Rev. Sci. Instrum. 77 (2006) 043506.
M. Sadiq, S. Ahmad, M. Shafiq, M. Zakaullah, Nucl. Instrum. Methods Phys. Res.,
B 252 (2006) 219–224;
B.B. Nayak, B.S. Acharya, S.R. Mohanty, T.K. Borthakur, H. Bhuyan, Surf. Coat.
Technol. 145 (2001) 8–15.
R.S. Rawat, T. Zhang, K.S. Thomas Gan, P. Lee, R.V. Ramanujan, Appl. Surf. Sci.
253 (2006) 1611–1615;
I.A. Khan, S. Jabbar, T. Hussain, M. Hassan, R. Ahmad, M. Zakaullah, R.S. Rawat,
Nucl. Instrum. Methods Phys. Res., B 268 (2010) 2228–2234.
G. Gerdin, W. Stygar, F.J. Venneri, J. Appl. Phys. 52 (1981) 3269–3275;
S.R. Mohanty, H. Bhuyan, N.K. Neog, R.K. Rout, E. Hotta, Jpn. J. Appl. Phys. 44
(2005) 5199–5205.
W.H. Bostick, V. Nardi, W. Prior, J. Choi, P.J. Fillingham, J. Nucl. Mater. 63 (1976)
V.N. Pimenov et al., Nukleonika 53 (2008) 111–121.
H. Bhuyan, S.R. Mohanty, N.K. Neog, S. Bujarbarua, R.K. Rout, Meas. Sci.
Technol. 14 (2003) 1769–1776.
H. Bhuyan, S.R. Mohanty, N.K. Neog, S. Bujarbarua, R.K. Rout, J. Appl. Phys. 95
(2004) 2975–2981.
M. Bhuyan, N.K. Neog, S.R. Mohanty, C.V.S. Rao, P.M. Raole, Phys. Plasmas 18
(2011). 033101–8.
R.S. Rawat, IEEE Trans. Plasma Sci. 41 (2013) 701–713;
M. Hassan et al., Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/
A. Lepone, H. Kelly, D. Lamas, A. Marquez, Appl. Surf. Sci. 143 (1999) 124–134.
G. Sanchez, J. Feugeas, J. Phys. D: Appl. Phys. 30 (1997) 927–936.
Y. Minyou, Plasma Sci. Technol. 7 (2005) 2828–2834.
J. Feugeas et al., Surf. Coat. Technol. 204 (2010) 1193–1199.
J.H. Evans, J. Nucl. Mater. 76–77 (1978) 228–234.
H.K. Kim, S.H. Huh, J.W. Park, J.W. Jeong, G.H. Lee, Chem. Phys. Lett. 354 (2002)
R.S. Rawat, P. Arun, A.G. Vedeshwar, P. Lee, S. Lee, J. Appl. Phys. 95 (2004)
Fly UP