VOLUME 83, NUMBER 21 PHYSICAL REVIEW LETTERS 22 NOVEMBER 1999
Multiphase Foamlike Structure of Exploding Wire Cores
S. A. Pikuz,* T. A. Shelkovenko,* D. B. Sinars, J. B. Greenly, Y. S. Dimant, and D. A. Hammer
†
Laboratory of Plasma Studies, Cornell University, Ithaca, New York 14853
(Received 22 April 1999)
X-ray backlighter images (radiographs) of current-induced explosions of 7.5 25 mm diam metal
wires show for the first time mm scale, time-resolved details of a persistent foamlike liquid-vapor
structure of the expanded wire core. Experiments with refractory and highly resistive metals, with
current rising to 2–5 kA per wire in 350 ns, show that a substantial portion of the wire material is
not vaporized but remains in a condensed state. As the current damps out, the remaining liquid phase
material coalesces into separate droplets.
PACS numbers: 52.80.Qj, 51.50. + v, 68.35.Rh, 83.70.Hq
Recent experiments on the Z accelerator [1] have gener-
ated extremely high-power x-ray pulses from the implosion
of multiwire array Z-pinch loads. In order to understand
those experiments, several groups are attempting to de-
velop the capability to model the whole process starting
from the cold wire state [2 – 4]. Until now, however, there
has been very little experimental information on the ini-
tial stages of wire explosion. The existence of long-lived
dense cores within lower density coronal plasmas during
nanosecond explosions of single wires or multiwire arrays
is well established [5]. However, the physical state of the
core material and the processes occurring there have been
only a matter of conjecture. Using the technique of high
resolution x-ray backlighting (radiography) [6], we have
found complex small-scale (typical sizes of 2 30 mm)
structure of the exploding wire cores of both single-wire
and multiwire loads subjected to 2–5 kA per 7.5 25 mm
diam wire for 1 ms. The radiographs show features im-
plying that the core has become a foamlike, liquid-vapor
mixture that is a result of explosive volume boiling. Most
of the energy which drives these wire explosions is deliv-
ered during the initial 50 ns, when the current is only a
few hundred amperes per wire. (The existence of a mul-
tiphase initial stage in current-driven wire explosions, but
with a very different morphology, was described with pho-
tographs of microsecond discharges by Chace in 1959 [7].)
The persistence of liquid-phase material is demon-
strated by the eventual formation of columns of droplets
along the original wire position after the discharge current
decays to zero in W, Mo, NiCr, and Ti wire explosions.
By contrast, Al, Au, Cu, and Ag wires, all high conduc-
tivity, relatively low melting point materials, form fully
vaporized, more uniform expanding columns of wire ma-
terial, evidently because the energy deposited in them dur-
ing the critical first 50 ns of the current pulse is sufficient
to cause the multiphase condition to be short lived. Our
results demonstrate that it is essential in modeling wire ex-
plosions to include a whole range of material properties,
as well as the formation of plasma around the wire from
desorbed gases and vaporized wire material. The current
waveform applied to each of these wires is similar to that
at the very beginning of the pulse on the Z accelerator.
The main diagnostic technique in the experiments re-
ported here is pulsed point-projection radiography of the
exploding wires. The x-ray sources are very short-time
(,0.5 ns) x-ray bursts from collapsing 13 25 mm diam
Mo-wire X pinches (Fig. 1), the characteristics of which
are described in detail elsewhere [6]. In brief, Mo-wire X
pinches are found experimentally to produce submicron,
bright x-ray sources in the 2.5 – 10 keV x-ray energy range
when the radiation is filtered by a 12.5 mm Ti foil; the
spectrum consists of free-bound continuum and several L-
shell lines. Radiographs are obtained of 1 to 4 wires ex-
ploded in parallel by a damped sinusoidal pulse of current
which reaches a peak amplitude of up to 4.5 kA in about
350 ns (the auxiliary pulser in Fig. 1) and has an e-fold
damping time of about 4 ms. The Mo-wire X pinches,
usually two in parallel, separated by 2 cm, are driven us-
ing the 400 – 450 kA peak current XP pulser [8]. The setup
gives radiographs at two times during each pulse, with the
interval between the radiographs variable from near 0 to
about 20 ns by changing the X -pinch wire size. A filtered
film stack gives up to three radiographs of the wire(s) being
tested for each X -pinch x-ray burst. The spatial resolution
is 1 3 mm, depending upon the X -pinch quality and the
FIG. 1. Schematic view of the experimental setup showing
the x-ray backlighter source positions relative to the exploding
wire array being radiographed.
0031-9007 99 83(21) 4313(4)$15.00 © 1999 The American Physical Society 4313