ArticlePDF Available

SPAWAR Systems Center-Pacific Pd:D CoDeposition Research: Overview of Refereed LENR Publications

Authors:

Abstract

Scientists at the US Navy SPAWAR Systems Center-Pacific (SSC-Pacific), and its predecessors, have had extraordinary success in publishing LENR papers in peer-reviewed journals. This success hasn't come easily and is due to several factors. One key reason for this success was the courage of the SSC-Pacific upper management in allowing scientists to conduct research and publish results in a controversial field. The few journal editors, who had the fortitude to consider our work, also contributed to this success. This contrasts with the majority of their peers who, taking the path of least resistance, ignored our work out of hand and returned manuscripts with, 'the subject matter is not in the purview of the journal'. The reviewers also played a role in the successful publication of LENR-related papers. A multitude of reviewers, many outside the LENR field, had to put aside their biases and look objectively at our data. In turn, the reviewers' relentless concerns forced us to tenaciously address their issues. Ultimately, the SSC-Pacific team published 21 refereed papers in seven journals and a book chapter, spanning 19 years beginning in 1989. This paper is a brief synopsis of those publications. 1. S. Szpak, P.A. Mosier-Boss, and J.J. Smith, "On the Behavior of Pd Deposited in the Presence of Evolving Deuterium," J. Electroanal. Chem., 302 (1991) 255-260 This was a preliminary note introducing the Pd/D co-deposition protocol as an alternative experimental approach to initiate LENR. Temperature measurements using thermocouples placed in the cathode and solution show that the cathode was hotter than the solution. This indicates that the observed heat is not due to Joule heating. A ten fold increase in tritium content in the solution was observed. Experiments were conducted with photographic film in close proximity of the cathode. After development, the film showed a grid pattern due to the Ni screen cathode and was attributable to the emission of soft X-rays.
1
Szpak, S., et al.
SPAWAR Systems Center
-
Pacific Pd:D Co
-
Deposition Research: Overview of Refereed LENR
Publications
in
ICCF
-
. 2008. Washington, DC.
SPAWAR Systems Center
-
Pacific Pd:D Co
-
Deposition
Research
:
Overview of
Refereed LENR Publications
S. Szpak
1
, P. A. Mosier
-
Boss
1
, F. Gordon
1
, J. Dea
1
, J. Khim
2
, L.
Forsley
2
1
SPAWAR Systems Center
-
Pacific, San Diego, CA 92152
2
J W K International, Annandale, VA.
Abstract
Scientists at the US Navy SPAWAR Systems Center
-
Pacific (SSC
-
Pacific), and
its predecessors, have had extraordinary success i
n publishing LENR papers in
peer
-
reviewed journals. This success hasn’t come easily and is due to several
factors. One key reason for this success was t
he courage of the SSC
-
Pacific upper
management
in allowing
scientists to conduct research and publish re
sults in a
controversial field.
The
few journal editors
, who
had the fortitude to consider our
work
, also contributed to this success
.
This contrasts with the majority of their
peers who, taking the path of least resistance, ignored our work out of hand an
d
returned manuscripts with, ‘the subject matter is not in the purview of the journal’.
The reviewers also played a role in the successful publication of LENR
-
related
papers. A multitude of reviewers, many outside the LENR field, had to put aside
their bia
ses and look objectively at our data. In turn, the reviewers’ relentless
concerns forced us to tenaciously address their issues. Ultimately, the SSC
-
Pacific
team published 21 refereed papers in seven journals and a book chapter, s
panning
19 years beginnin
g in 1989. This paper is a brief synopsis of those publications.
1. S. Szpak, P.A. Mosier
-
Boss, and J.J. Smith, “On the Behavior of Pd Deposited in the Presence
of Evolving Deuterium,”
J. Electroanal. Chem
.,
302
(1991) 255
-
260
This was a preliminary note
introducing the Pd/D co
-
deposition protocol as an alternative
experimental approach to initiate LENR. Temperature measurements using thermocouples
placed in the cathode and solution show that the cathode was hotter than the solution. This
indicates that th
e observed heat is not due to Joule heating. A ten fold increase in tritium content
in the solution was observed. Experiments were conducted with photographic film in close
proximity of the cathode. After development, the film showed a grid pattern due to
the Ni screen
cathode and was attributable to the emission of soft X
-
rays.
2. S. Szpak, C.J. Gabriel, and R. J. Nowak, “Electrochemical Charging of Pd Rods,”
J.
Electroanal. Chem
.,
309
(1991) 273
-
292
A model was developed to describe the electrochemical c
harging of palladium rods. This model
coupled the interfacial processes with the transport of interstitials in the electrode interior. It was
shown that boundary conditions arise from the solution of equations governing the elementary
adsorption
-
desorption
and adsorption
-
absorption steps as well as the symmetry of the electrode.
2
3. S. Szpak, P.A. Mosier
-
Boss, S.R. Scharber, and J.J. Smith, “Charging of the Pd/
n
H System:
Role of the Interphase,”
J. Electroanal. Chem
.,
337
(1992) 147
-
163.
Slow scan cyclic vo
ltammetric studies of Au/Pd/
n
H were conducted to examine the dynamics of
transport of electrochemically deuterium/hydrogen across the electrode/electrolyte interphase. It
was found that a coupled, two
-
layer model of the interphase describes the observed be
havior as a
function of scan rate and electrolyte composition. The effect of chemisorbing species, thiourea,
and pH on the transport across the interphase was also investigated.
4. S. Szpak, P.A. Mosier
-
Boss, C.J. Gabriel, and J.J. Smith, “Absorption of De
uterium in
Palladium Rods: Model vs. Experiment,”
J. Electroanal. Chem
.,
365
(1992) 275
-
286.
A model that incorporates variables such as electrochemical rate constants, bulk diffusion
coefficient, and charging current has been developed. Such a model can b
e used to predict the
overpotential, surface coverage, and bulk loading of the electrode during charging. The
computed time dependence of the bulk loading has been compared with published experimental
charging curves. Microscopic examination of a charging
Pd cathode using Nomarski optics has
shown that, even within a single grain, there are preferred sites of absorption.
In
-
situ
XRD
measurements of the charging Pd cathode shows that deuterium preferentially enters the Pd
lattice through the 111 sites. With
additional charging, a broadening and a shift to lower 2
angles was observed which suggested the presence of a supercharged layer.
5. S. Szpak, P.A. Mosier
-
Boss, R.D. Boss, and J.J. Smith, “Comments on the Analysis of Tritium
Content in Electrochemical Ce
lls,”
J. Electroanal. Chem
.,
373
(1994) 1
-
9.
The time dependence of tritium content of an open cell operating galvanostatically with
intermittent sampling has been derived and is given by the following expression:
where
f
is the tr
itium mass fraction, m is the mass of the electrolyte phase, r(i) denoted the rate
of change associated with the cell current, q is the rate at which tritium is added/removed, and s
is the isotopic separation factor. It was concluded that a complete mass b
alance between the
liquid and gas phases was necessary in order to determine that tritium was produced in the Pd/D
system.
6. S. Szpak, P.A. Mosier
-
Boss, and J.J. Smith, “Deuterium Uptake During Pd
-
D Codeposition,”
J. Electroanal. Chem
.,
379
(1994) 121
-
127
.
Deuterium uptake during Pd
-
D co
-
deposition was examined using galvanostatic perturbation
techniques. The resultant potential relaxation curves exhibit four distinct potential
-
time intervals
where the relaxation process is controlled by the interaction be
tween the transport of deuterium
from inside the lattice to the surface to form adsorbed deuterium and the reduction of palladium
from solution.
7. S. Szpak, P.A. Mosier
-
Boss, S.R. Scharber, and J.J. Smith, “Cyclic Voltammetry of Pd + D
Codeposition,”
J. E
lectroanal. Chem
., 3
80
(1995) 1
-
6.
Processes associated with the Pd + D alloy codeposition were examined by cyclic voltammetry.
The dynamics of the interphase region are discussed.
1
1
)
0
(
)
(
)
0
(
1
)
(
)
1
(
)
0
(
)
(
)
0
(
)
0
(
)
(
S
S
m
t
i
r
m
i
r
S
q
m
t
i
r
m
f
t
f
3
8. S. Szpak, P.A. Mosier
-
Boss, and J.J. Smith, “On the Behavior of the Cat
hodically Polarized
Pd/D System: Search for Emanating Radiation,”
Phys. Lett. A
.
210
(1996) 382
-
390.
Pd/D co
-
deposition experiments were conducted inside lead caves while measuring gamma and
X
-
rays, as a function of time, using a HPGe detector with an Al w
indow and a Si(Li) detector
with a Be window. The cathodically polarized Pd/D system was observed to emit X
-
rays with a
broad energy distribution and with an occasional emergence of recognizable peaks attributable to
the Pd K
α
and Pt L lines. The emission of X
-
rays is sporadic and of limited duration.
9. S. Szpak and P.A. Mosier
-
Boss, “On the Behavior of the Cathodically Polarized Pd/D System:
A Response to Vigier’s Comments,”
Phys. Lett. A
.
221
(1996) 141
-
143.
Preliminary res
ults of thermal imaging of the Pd/D cathode prepared using the co
-
deposition
technique are presented. Hot spots are observed that appear/disappear chaotically. With time
these hot spots merge into larger islands that exhibit oscillatory behavior. SEM image
s of a Pd/D
cathode that had melted during electrolysis are shown.
10. S. Szpak, P.A. Mosier
-
Boss, R.D. Boss, and J.J. Smith, “On the Behavior of the Pd/D
System: Evidence for Tritium Production,”
Fus. Technol
.,
33
(1998) 38
-
51.
In these experiments, the D
2
and O
2
gases were recombined in a separate chamber. The tritium
content in the liquid and gas phases were measured daily using a liquid scintillation. The
measured data were analyzed using the mass balance expression that was derived earlier. It was
obse
rved that tritium production occurred in bursts and sporadically. During a burst, the rate of
tritium production was estimated to be 10
3
to 10
4
atoms s
-
1
. Tritium produced during prolonged
electrolysis was transported out of the electrode interior by two d
istinct paths. One path results in
enrichment of tritium in both the electrolyte and gas phases. The second results in enhancement
only in the gas phase.
11. S. Szpak and P.A. Mosier
-
Boss, “On the Release of from cathodically polarized
Palladium Elec
trodes,”
Fus. Technol
.
34
(1998) 273
-
278.
The release paths for tritium produced during electrochemical compression of deuterium in a Pd
lattice were examined. The results indicate that tritium production requires high D/Pd atomic
ratios. This requirement
is met if there are no channels reaching the contact surface. The
electrogenerated tritium is distributed among the voids and bulk material. Gas evolution
promotes a continuous exchange between the atoms residing in the subsurface layer and
with t
hose in the adsorbed state. Atoms in the adsorbed state exchange with the molecules of the
contacting electrolyte phase or gaseous phase, leading to two distinct transfer paths.
12. P.A. Mosier
-
Boss and S. Szpak, “The Pd/
n
H System: Transport Processes and
Development
of Thermal Instabilities
,” Nuovo Cimento Soc. Ital. Fis. A
,
112
(1999) 577
-
587.
The surface temperature distribution of the cathode prepared by Pd/D co
-
deposition on a Ni
screen was measured using an infrared camera. It was observed that, unli
ke joule heating, excess
enthalpy generation occurs in the form of localized events in close proximity to the contact
surface. It was also observed that, the higher the electrolyte temperature, the more frequent the
events. In the limit, these events overl
ap to produce oscillating islands.
13. S. Szpak, P.A. Mosier
-
Boss, and M. H. Miles, “Calorimetry of the Pd + D Codeposition,”
Fus. Technol
.,
36
(1999) 234
-
241.
H
n
1
H
n
1
4
Calorimetric measurements indicate that the excess enthalpy generated in cells using cathodes
p
repared by the co
-
deposition process is, on average, higher than that produced in cells using
solid Pd rods. Infrared imaging of the cathodes prepared by Pd/D co
-
deposition shows that the
heat sources are highly localized. The steepness of the temperature
gradients indicate that the
heat sources are located in close proximity to the electrode
-
solution contact surface.
14. S. Szpak, P.A. Mosier
-
Boss, M. H. Miles, and M. Fleischmann, “Thermal Behavior of
Polarized Pd/D Electrodes Prepared by Co
-
deposition,”
Thermochimica Acta
,
410
(2004) 101
-
107.
The thermal behavior of Pd/D electrodes, prepared by the co
-
deposition technique, was
examined using a Dewar
-
type electrochemical cell calorimeter. Results indicated that excess
enthalpy is generated during and afte
r the completion of the co
-
deposition process. The rates of
excess enthalpy generated using the co
-
deposition technique were higher than those obtained
using Pd wires or other forms of Pd electrodes. Positive feedback and heat
-
after
-
death effects
were obse
rved. The rates of excess power generation were found to increase with an increase in
both cell current and cell temperature.
15. S. Szpak, P.A. Mosier
-
Boss, C. Young, and F.E. Gordon, “The Effect of an External Electric
Field on Surface Morphology of Co
-
d
eposited Pd/D Films,”
J. Electroanal. Chem
.,
580
(2005)
284
-
290.
After plating out the Pd on a Au foil, the cell current was increased and an external electric field
was applied across the cell. The experiment was terminated after 48 h. The cell was
disass
embled and the cathode was subjected to analysis using an SEM. In the absence of an
external electric field, the Pd deposit exhibits a cauliflower structure. After exposure to an
external electric field, significant changes in the morphology of the Pd/D de
posit were observed.
Fractal features were observed as well as dendritic growths, rods, wires, and craters.
Considerable work is needed to account for the variety of shapes. The process of shape change is
driven by energy transferred from the electrostatic
field and directed by the field.
16. S. Szpak, P.A. Mosier
-
Boss, C. Young, and F.E. Gordon, “Evidence of Nuclear Reactions in
the Pd Lattice,”
Naturwissenschaften
,
92
(2005) 394
-
397.
When a cathode prepared by Pd/D co
-
deposition is subjected to an exter
nal electrostatic field,
SEM analysis of the deposit shows discrete sites exhibiting molten
-
like features. Such features
require substantial energy expenditure in order to form. EDX analysis of these features shows
the presence of new elements (Al, Mg, Ca
, Si, Zn,…) that could not be extracted from cell
components.
17. S. Szpak, P.A. Mosier
-
Boss, and F.E. Gordon, “Further Evidence of Nuclear Reactions in the
Pd/D Lattice: Emission of Charged Particles,”
Naturwissenschaften
,
94
(2007) 511
-
514.
CR
-
39 is a s
olid state nuclear track detector that is used to detect energetic particles such as
alphas, protons, tritons, and helium
-
3. Pd/D co
-
deposition was done, in the presence of an
external electric or magnetic field, with the cathode in direct contact with a C
R
-
39 detector.
Tracks on the CR
-
39 detector were observed where the cathode was in contact with the plastic
indicating that the source of the tracks is the cathode. The features of these tracks (optical
contrast, shape, and bright spot in the center of the
pit) are consistent with those observed for pits
in CR
-
39 that are of a nuclear origin. The emission of the energetic particles is sporadic and
occurs in bursts.
5
18. P.A. Mosier
-
Boss, S. Szpak, F.E. Gordon, and L.P.G. Forsley, “Use of CR
-
39 in Pd/D Co
-
dep
osition Experiments,”
Eur. Phys. J. Appl. Phys.,
40
(2007) 293
-
303.
A series of control experiments were conducted. It was shown that the tracks observed in CR
-
39
detectors subjected to Pd/D co
-
deposition were not due to radioactive contamination of the ce
ll
components. No tracks were observed when Cu was electrochemically plated on the surface of
the CR
-
39 detectors. This indicates that the pits cannot be attributed to chemical attack of the
surface of the CR
-
39 by either D
2
, O
2
, or Cl
2
present in the elec
trolyte. Nor can the pits be
attributed to the metal dendrites piercing into the surface of the detectors. Additional experiments
showed that LiCl is not essential for the production of pits and that the density of pits
significantly decreases when light w
ater is substituted for D
2
O. Quantitative analysis using an
automated scanner shows that there are three populations of tracks (0.1
-
0.5 μm, 0.9
-
4.0 μm, and
4.1
-
12 μm) and that the pits can be either perfectly circular or elliptical in shape.
19. P.A. Mosie
r
-
Boss, S. Szpak, F.E. Gordon, and L.P.G. Forsley,
Detection of Energetic
Particles and Neutrons Emitted during Pd:D Co
-
deposition
,”
Low Energy Nuclear Reactions
Source Book
, American Chemical Society , Chapter 14, pp 311
-
334. (2008).
Co
-
deposition proced
ures and control experiments specifically identified the conditions under
which nuclear particles were observed, and ruled out chemical means of mimicking nuclear
tracks. The nuclear tracks are quantitatively examined and are consistent with neutron knock
-
ons.
Triple tracks are presented as evidence of
12
C(n,n´)3
α indicative of DT fusion.
20. P.A. Mosier
-
Boss, S. Szpak, F.E. Gordon, and L.P.G. Forsley,
Triple Tracks in CR
-
39 as
the Result of Pd/D Co
-
deposition: Evidence of Energetic Neutrons
,”
Naturwissen
schaften
, in
press. (2008).
Triple tracks have been observed in CR
-
39 detectors used in Pd/D co
-
deposition experiments.
Microscopic examination of the bottom of the triple track pit shows that the three lobes of the
track are splitting apart from a center
point. The presence of
three α
-
particle tracks outgoing
from a single point is diagnostic of the
12
C(n,n´)3
α carbon break up reaction and
is easily
differentiated from other neutron interactions occurring within the CR
-
39 detector
. The presence
of triple tracks suggests that DT
reactions that produce
9.6 MeV neutrons are occurring inside
the Pd lattice.
21. P.A. Mosier
-
Boss, S. Szpak, F.E. Gordon, and L.P.G. Forsley,
“Use of CR
-
39 in Pd/D Co
-
deposition Experiments: A Response to Kowalski,”
Eur. Phys. J. Appl. Phys.,
41
(2008)
in press.
Earlier we reported that the pits generated in CR
-
39 detectors during Pd/D co
-
deposition
experiments are consistent with those observed for pits that are of a nuclear origin. Recently, that
interpretation has been challenged. In this communicatio
n, additional experimental data and
further analysis of our earlier results are provided that support our original conclusions.
Technical Report
Full-text available
The Pd/D co-deposition process has been shown to provide a reproducible means of manufacturing Pd-D nano-alloys that induce low energy nuclear reactions (LENRs). Cyclic voltammetry and galvanostatic pulsing experiments indicate that, by using the co-deposition technique, a high degree of deuterium loading (with an atomic ratio D/Pd>1) is obtained within seconds. These experiments also indicate the existence of a D+ species within the Pd lattice. Because an ever expanding electrode surface is created, non-steady state conditions are assured, the cell geometry is simplified because there is no longer a need for a uniform current distribution on the cathode, and long charging times are eliminated.
Conference Paper
Full-text available
One of the characteristics of the cold fusion phenomenon (CFP), i.e. low energy nuclear reactions in solids with high density hydrogen isotopes at near room-temperature in ambient radiations, is the localization of nuclear reactions relevant to the phenomenon. The earliest data sets showing the localization of nuclear reactions were given by Iyengar and Srinivasan (1989, 1990) for product tritium localized in spots (fraction of a millimeter or less in size) at surfaces of a Ti sample. Morrey et al. (1990) reported 42He in surface layers of Pd with a width of about 25 μm. Okamoto et al. (1994) observed nuclear transmutation of minor elements in surface layers of Pd samples, especially 2713Al into 2814Si, in a layer of about 2 μm. Miley et al. (1996) observed various transmutation products, especially transmutation of A28Ni isotopes, in the thin Ni layer of width about 0.1 μm in a protium system. Mizuno et al. (1996) also observed similar nuclear transmutations in PdDx and AuHx systems. Iwamura et al. (2002, 2005) observed transmutations of Sr, Cs and Ba into Mo, Pr and Sm, respectively, in/on a Pd surface layer of thickness about 100 A. Furthermore, the second product confirmed its spatial localization with a diametrical size of about 100 μm. Thus, we know that the reaction products of the CFP are localized in regions at surface layers about a few μm in width of massive CF materials and also in thin CF materials with layer structures of widths less than one μm. The confinement of nuclear reactions in surface/boundary layers of a few μm might be a characteristic of the CFP in protium and deuterium systems. This characteristic of the CFP is explained by the structure of a CF material composed of (1) a host material composed of a metal and a hydrogen isotope and (2) a guest material including agents for the CFP on the surface of the former. The localization of nuclear reactions in the guest material in spot-like regions is another characteristic showing nuclear reactions between the agent (such specific nuclei as alien nuclei (Li, Sr, Cs) or the irregularity of atomic arrays). These characteristics of localized nuclear reactions are investigated quantum mechanically using properties of nucleus and proton/deuteron wavefunctions in the CF material.
Article
Full-text available
Earlier we reported, in this journal, that the pits generated in CR-39 detectors during Pd/D co-deposition experiments are consistent with those observed for pits that are of a nuclear origin. Recently, that interpretation has been challenged. In this communication, additional experimental data and further analysis of our earlier results are provided that support our original conclusions.
Article
Full-text available
Evidence for tritium production in the Pd/D system under cathodic polarization is presented. A comparison of the observed distribution and that calculated, based upon the conservation of mass, leads to the conclusion that tritium is produced sporadically at an estimated rate of ca 10 3-104 atoms per second. The results of several runs are interpreted by employing the concept of an electrode/elec- trolyte interphase and the accepted kinetics of hydrogen evolution. Observation of burst-like events followed by longer periods of inactivity yield poor reproducibility when distributions are averaged over the total time of electrolysis.
Article
Full-text available
Surface temperature distribution associated with excess enthalpy production during the codeposition process is presented. The interpretation is sought via the multilayer concept of the electrode/elctrolyte interphase. The effect of gas evolution on activities within the interphase is considered.
Article
Expressions are given for the tritium distribution between gas and liquid phases arising from prolonged electrolysis of D2O on Pd electrodes saturated with absorbed deuterium. Errors associated with commonly employed experimental procedures and their interpretation using a simple model for the distribution between the gas and liquid phases are evaluated. With regard to tritium production, an unambiguous resolution requires a complete mass balance. Conclusions based on partial data, such as the measurement of the tritium content in a single cell component (e.g. electrode, electrolyte or “electrolytic” gas phase), may be misleading, particularly if production rates are either low or intermittent.
Article
Processes associated with the Pd+D alloy codeposition are examined by cyclic voltammetry. The voltammograms cover the potential range: +0.3 to −1.3 V (measured against an Ag/AgCl/KCl (sat) reference) and indicate that the partial current due to the Pd2+ ion reduction is diffusion limited at slow scan rates. Except for the significant increase in cathodic currents due to D2O reduction at ca. −0.25 V which occurs on a freshly generated Pd surface, the shapes of the voltammograms marginally differ from those recorded in the absence of Pd2+ ions in the electrolyte phase. A discussion of the dynamics of the interphase is presented.
Article
The dynamics of transport of electrochemically generated deuterium across the electrode/electrolyte interphase was examined by slow scan (10 mV s−1) voltammetry. The investigation covers the potential range −1.2 to +0.4 V measured vs. an Ag/AgCl reference. It was found that a coupled, two-layer model of the interphase describes the observed behavior as a function of scan rate and electrolyte composition. The effect of chemisorbing species, e.g. CN− ions, as well as reactive species, e.g. SC(NH2)2, on the transport across the interphase is also discussed. Results are contrasted with those obtained for light water.
Article
Electrodes prepared by PdD codeposition exhibit highly expanded surfaces which achieve high degrees of DPd loading within seconds. In this communication, morphology of the Pd electrode, the structure of the interphase, and selected thermal effects are discussed.
Article
The polarized PdD electrode undergoes significant morphological changes when exposed to an external electric field. These changes range from minor, e.g. re-orientation and/or separation of weakly connected globules, through forms that result from a combined action of the field as well as that connected with the evolution of gaseous deuterium, to shapes that require substantial energy expenditure.
Article
Thermal behavior of polarized Pd/D electrode, prepared by the co-deposition technique, serving as a cathode in the Dewar-type electrochemical cell/calorimeter is examined. It is shown that: (i) excess enthalpy is generated during and after the completion of the co-deposition process; (ii) rates of excess enthalpy generation are somewhat higher than when Pd wires or other forms of Pd electrodes are used; (iii) positive feedback and heat-after-death effects were observed; and (iv) rates of excess power generation were found to increase with an increase in both cell current and cell temperature, the latter being higher.
Article
The electrochemical charging of Pd rods by deuterium involves a complex coupling of electrochemical, interfacial and transport processes. In order to predict the overpotential, surface coverage and bulk loading of the electrode during charging, a model has been developed that incorporates the essential features of these processes and involves variables such as the electrochemical rate constants, the bulk diffusion coefficient and the charging current. Features of the computed time dependence of the bulk loading are then compared with published experimental charging curves. New microscopic observations and X-ray diffraction data provide further evidence for the details of the charging process.