Professional Documents
Culture Documents
by N.V. Mandich
HBM Engmeering Co., Lansing, I/.
Electroplating with pulse current (PC) and pulse-reverse current (PRC) is a technique
where standard, stationary DC current is replaced with its modulated (nonstationary) forms.
Although it has been recognized for a long time electroplating with modulated currents was
mostly dormant and used only occasionally for PRC plating of cyanide copper where better
leveling was recognized early in the 1940s and 1950s.
Lack of appropriate current sources and/or its prohibitive cost hampered the further
interest for this otherwise promising technique for the next 4 decades. Although it was known
that the morphology of gold and gold alloys would be influenced by periodically reversed or
pulsed current. great interest was developed only when it was realized that pulse-plated gold
deposits proved superior to DC-plated deposits for certain electronic applications. The age of
electronics and materials shortages gave new importance to gold as a commodity, and a boom
in gold prices led to a demand for ways of reducing gold consumption.
Pulsing proved to be an answer and this justified the more expensive pulse rectifiers.
After that theoretical contributions and industrial applications followed.
Before the effect of pulsed current on various aspects of metal deposition is elaborated
it is necessary to examine the way in which pulsed current waveforms are characterized. The
advent of modern electronics and microprocessor control has permitted great flexibility of the
programming of the applied current waveform. Trains of pulses can be programmed to give
very complex waveforms. Square waves are the easiest waveforms to produce because this
requires only a switching arrangement rather than a specialized waveform generator. Current
waveforms can be divided into two major groups: These are unipolar pulses, where all the
pulses are in one direction, and bipolar pulses, where anodic and cathodic pulses are mixed.
There are many variants on these, but as the complexity of the waveform increases so does
the number of variables, which makes it more difficult to understand how a particular
waveform affects the electrodeposition process. Consequently. the present discussion is
restricted to the consideration of the simplest case of unipolar and bipolar pulses, i.e., pulse
and pulse reverse.
In order to characterize a direct current it is sufficient to know the current density. The
characterization of a train of current pulses requires three parameters to be known. These are
the cathodic peak pulse current density, j<; the cathodic pulse length, f,.: and the interval
between the pulses, r,,, as shown in Fig. I. In practice average current density, j,,,, is
measured, and it is expressed in the case of PC as follows:
j,, = Cj, X t,)Kt, + t,,) (1)
A quantity frequently encountered is the duty cycle, T, representing the portion of time
in each cycle when the current is on. It is defined by the following equation:
T = t,/(t, + t,) (2)
The product of the duty cycle and the peak pulse current density gives the PC average
current density.
j,, = j, X T (3)
375
Unipolnr
Pulsing
Pulse _ Superimposed
Pulse Duplex
Pulse ‘I--
7-1
Pulsed
Pulse Pulse-on
Pulse
Bipolar
Pulsing
(PRC)
Pulse
Reverse ta Pulse
Reverse
(withofTtime)
& -+i ic
ZC
s: 0
Z&3
7l-T w
Pulsed
Pulse
Reverse Pulse-On
Pulse
Reverse
vu 0
376
The characterization of a PRC waveform requires four parameters to be known and
defined by the following equations:
Unipolar pulsed current justified its practical application mainly through its ability to
affect the mechanisms of electrocrystallization, which in turn. controls the physical and
mechanical properties of the electrodeposited metal. Because the nucleation rate of a growing
electrodeposit is proportional to the applied current density, the use of high current density
pulses can produce deposits with reduced porosity and. in most cases, a finer grain size.
Whether a finer grained deposit is obtained in practice depends upon what happens during t,
when the current i\ interrupted because this can allow the desorption of impurities and
encourage renucleation with the formation of new, smaller crystal grains. The effects obtained
in practice are also dependent on the specific electrochemical system to which a particular set
of the current pulses is applied.
A common example of different grain growth caused by the desorption of impurities is
observed when current pulses are applied to an acid gold-cobalt alloy plating system. In this
case deposits by pulse current have a very low carbon and nitrogen content and the grain size
is either decreased or increased depending on duty cycle. There is also a significant increase
in the ductility of these deposits and an increase in electrical conductivity and density, ah well
as a decrease in porosity. Decrease in porosity in practice means that a thinner gold deposit
can be applied and still pass porosity tests, resulting in a significant cost saving. Another
rather unexpected metal where pulse current plating has found its application is nickel, which
was of practical interest due to the ability of pulsed current to control the levels of stress and
ductility in nickel deposits. This has home obvious applications in the electroforming industry
where stress control is of paramount importance.
The rate of metal deposition is dictated by the average current density and is equivalent
to the applied current density in DC plating. Now, the first major limitation of pulse plating
from a practical viewpoint becomes evident. In order to produce the same average deposition
rate as for DC, as duty cycle is reduced. the peak pulse current density needs to be increased.
For example, at a duty cycle of IO%, a peak pulsed current density of 40 A/dm’ is required
for an average current density of 4 A/dm’. In practical applications this would seldom be
viable due to limitations of rectifier capacity. As duty cycle is increased it begin\ to approach
DC; so a compromise situation needs to be achieved. If sufficient spare rectifier capacity is
available a duty cycle of 33 to 50% is probably the minimum practical value. For applications
using high current, such as chromium platin,, c a value of 50 to 7.5% would be the lowest usable
duty cycle.
After considering the practical limitations of duty cycle it is necessary to consider the
effects of frequency, which is defined as the reciprocal of the cycle time, t:
f = I/(t, + ‘J = 1/t (6)
At high f, the double layer does not have time to charge fully during the I, or fully
discharge during the t, time. This has a smoothing effect on the applied waveform, which
377
begins to approach DC current, and this limits the maximum useful frequency to around 500
Hz for most applications; however, higher frequencies can he used where very high pulse
curl-ent densities are employed hecause the double layer charge and discharge times become
shorter as peak pulse currenl density is increased.
Some manufacturer\ of pulse-plating equipment advocate the use of very high frequen-
cies, but the practical maximum frequency that can be applied to a plating electrolyte is
limited by the capacitance of the double layer at the interface between the plating electrolyte
and the article being plated.
Unsubstantiated claims have often been made that pulse plating can improve the
maximum rate of deposition obtainable from plating baths by scvcral orders of magnitude.
The theory behind this reasoning is that as frequency is increased. the diffusion layer become,
thinner: however. the diffusion layer, which is encountered with pulse plating. also pulses at
the high frequency applied to a plating system. The diffusion layer does not have time to
disperse completely between pulses. and the total thickness of the diffusion layer approaches
that obtained when plating with DC cut-rent. Consequently, the use of pulse current hur \ery
little effect on the limiting current density. Metal ions cannot be discharged fasler than they
are supplied to the cathode surface. The rate of supply of metal ions depends only on
hydrodynamic and concentration factors and is little influenced by the applied waveform.
The major factor\ governing the supply of metal ions to the cathode surface are
hydrodynamic factors: rate of agitation, viscosity, diffusion coefficient, and the metal ion
concentration. The rate of consumption of metal ions depends only on the average current
density and ij not influenced to any extent by the shape of the applied waveform. Although
pulse plating cannot increase the theoretical limiting current densily of a plating electrolyte it
must he taken into account that the maximum practical current density at which a plating
electrolyte is usually operated is only 20 to 30% of the theoretical limiting current density.
Thih i\ because mass transport effects can produce burning at higher current densities areas.
Obviously, there is room for improvement. and higher practical plating rates can sometimes
be achieved with pulse current due to improved deposit properties. This effect is attributed to
the influence of pulhed current on the electrocry~talli~ation rather than on an enhancement of
the rate of mass transport.
The cell geometry dictates primary current density distribution; therefore, it is no1
affected by electrochemical parameters and would he unaffected by the applied current
waveform. The primary current distribution is modified in plating solutions by the secondary
current distribution, which arises due to the effects of activation (chemical) overpotential. In
general the greater the rate of change of potential with increasing current density the more the
overall current density distribution tends toward a secondary distribution and. excluding mash
transport effects. produces a more even metal dibtrihution. As current density is increased the
electrode resistance decreases and the rate of change of potential with current density also
becomes less. Thus, as current density is increased the current distribution tends toward a
primary distribution and the throwing power usually deteriorates. In electrolytes where metal
ions are not strongly complexed and cathodic efficiency is high a low currem density usually
produces a better throwing power than a high current density. With any form of modulated
current, whatever the shape of the waveform and whether unipolar or bipolar pulses are used
in order to maintain the same rate of deposition as for DC current, the cathodic current density
during the cathodic pulse must he higher than the equivalent DC current density. This means
378
that the overall current density distribution tends toward a primary distribution when pulse or
PRC plating are employed. Consequently. the deposit distribution in general would be
expected to be less uniform using pulsed current than with direct current; however, in
electrolytes whcrc the cathodic efficiency is less than lOO%, the use of pulsed current may
change the relative efficiencies of metal deposition and hydrogen evolution and may under a
particular set of conditions produce an increase in throwing power.
The application of pulse current leads to the improvement in the quality of a number of
industrial electrodeposits. By using pulse currem in comparison with the deposits obtained by
DC one can achieve smoother deposits of copper, nickel, and Tint from sulfate solution and
gold and copper from cyanide solutions. with the average current densities and plating times
being kept equal. Also, an equal or hetter quality of the deposits of copper, gold. and rinc is
obtained at higher current densities under pulse current than under DC conditions. A decrease
of grain size of the pulse-plated deposits is generally found to lead to an increased coverage
of the substrate with the same quality of the deposited metal, resulting in decreased porosity
and surface resistance. and increased density of the metals. It can be expected that this
increase in compactness is associated with II decrease in internal stresses and increased
ductility and change of hardness of metal deposits.
In situations where hydrogen is codeposited (current efficiency less than l(M)%) the
current distribution can be better in pulse currem regimes as compared with DC current.
PULSE-REVERSE PLATING
379
current are significantly increased. At the same time a high metal-to-current efficiency must
be maintained and the protective properties of the electrodeposits must satisfy the require-
ments. As was noted earlier, until relatively recently, these conditions were met mainly by
changing DC variables: the metal concentration, temperature and agitation of the bath, as well
as by inlroducing various additives.
Despite the fact that PRC deposition occurs with the incomplete use of the whole
quantity of electricity applied on the electrodes, this method of metal deposition often permits
acceleration of the electroplating rates. Faster deposition can be accomplished by using a
higher working current densities than compared with a DC current. Initially, this is caused by
the prevention of depletion of metal ions in the cathode double layer. to which the anodic
dissolution of the deposit contributes. The upper limit of the working current densities during
the metal deposition is a function of the parameters of the reversed current t,> and i;, and also
of the length of the total period. The limiting cathodic current density usually increases with
an increase of the ratio t,/t, and with a decrease in total time.
The a higher working current density during the PRC deposition as compared with DC
plating is a result of the periodic depolarization of the electrodes. Such an effect of the current
reversal on the electrode processes, as well as the electropolishing action of the anodic current
on the deposit, permits ensuing electrocrystallization on the cathode surface in the preferred
direction (smaller crystal grains) under a higher working (practical) current density. During
the cathodic period this leads to grain refinement usually associated with unipolar, pulse
current pulse plating. which is now further supported by the disturbance of growth steps by
the periodic inversion of the current.
In a number of DC electroplating processes for example, during the electrodeposition of
certain me(als from solutions of their simple salts, the upper limit of the working current
density can be, under certain conditions, held close in magnitude to the theoretical limiting
current. Often, however, under real, actual plaling conditions, the working cu]Tcnt density of
the cathode is limited to the admissible upper limit of the anodic current density. Exceeding
this limit can cause anode paasivation when the electroplating bath is operating in a DC mode
(e.g.. nickel, brass, and cyanide and acid copper baths). Now, currem reversal can prevent
paasivation of the anodes during the period of high cathodic current and sometimes
contributes to a significant increase of the upper limit of the anode working currcnI density
and, thus, in turn, increases the maximum working cathode current density.
380