POWER March 2015 - 31

WATER & WASTEWATER
tem or condenser, their presence does not
necessarily require the use of a neutralizing
amine. There are many mixed-metallurgy
units that operate using ammonia and that
carefully control air in-leakage with very low
copper corrosion rates.
The choice of which neutralizing amine to
use (and there are many) should be based on
where and how it is to function. It is critical that
both the basicity (amount of pH rise per ppm of
amine) and volatility of the amine (the ratio of
what goes into the steam versus what remains
in the water) is matched to the application.
The criticism of the general use of amines
in high-pressure utility cycles is centered on
two issues: the degradation of these organic
molecules in the steam cycle (particularly in
the superheater and reheater) and the consequence
of these degradation products-
namely, an increase in the cation conductivity
of the condensate and feedwater.
It has been long known that as neutralizing
amines pass through the steam cycle, they
break down into ammonia and organic acid
byproducts such as acetic acid, formic acid,
and carbon dioxide. The percentage of degradation
is certainly specific to the particular
amine and concentration in the steam, but it is
also unit specific and depends, at a minimum,
on the size and complexity of the superheater
and reheater piping, where it appears most of
the degradation occurs.
Those who advocate for the sole use of
ammonia instead of amines point to the degradation
of these products and see them as
" single-use " chemicals-good for only one
trip around the steam cycle. If all the amine
degrades with one trip through the superheater
and reheater, it cannot be available to
minimize the corrosion of copper condenser
tubes or affect the pH of a steam/water mixture
in the feedwater, and so it would not be
worth the trouble.
However, there are many different factors
that affect amine degradation rates and,
therefore, how beneficial an amine might be
in the system. These include the operating
pressure of the unit, where the copper alloys
are located, and whether the unit even has a
reheater. For example, in the standard tripledrum
HRSG, a significant percentage of the
amine may leave with the LP steam, where it
recycles through the condenser and preheater
sections of the HRSG and never sees the
high-temperature areas. This would significantly
increase its longevity and usefulness.
All these factors need be taken into account
when considering whether an amine would
be beneficial at a particular plant. It would
behoove anyone who is considering trying
an amine to set up to sample and test for the
amine and degradation products around the
cycle and also quantify improvements to iron
March 2015 | POWER
and copper corrosion rates. That will help
them determine, for their particular unit, if
the benefits of amine use outweigh the costs.
The degradation products of any amine
will add to the cation conductivity of the
condensate and feedwater. The longevity and
chemical structure of the amine will affect
the cation conductivity " bump " that the plant
will experience. Degassed cation conductivity
can remove carbon dioxide but generally
not all the other organic acids produced by
amines. So if amines are used, the normal
cation conductivity will need to be adjusted
for the presence of these products.
condensate pump discharge, there is no protection
for the copper alloy condenser tubes
against the combined effect of dissolved oxygen,
carbon dioxide, and ammonia. This is
why it is so critical to minimize air in-leakage
and control feedwater pH.
Many units have been replacing copper
alloy feedwater heaters with carbon steel or
stainless steel tubes over the years. When the
last copper feedwater heater is replaced, the
reducing agent can almost always be eliminated,
regardless of whether the condenser
contains copper alloys or not.
Carbon steel corrosion is inhibited by the
When the last copper feedwater heater is
replaced, the reducing agent can almost
always be eliminated, regardless of whether
the condenser contains copper alloys or not.
Controlling Oxidation Reduction
Potential
It can be generalized that the ability of an
alloy to withstand corrosion is a function of
the stability and tenacity of the oxide layer
that forms on the metal surface. As discussed
above, stainless steel has a very tight and
tenacious layer of chromium oxide that prevents
corrosion of the metal from oxygen and
from the common pH ranges found in feedwater.
Establishing
and maintaining a good oxide
layer on carbon steel is critical to minimizing
FAC. Copper oxides are also protective-as
long as they remain in place.
Particularly in the case of copper alloys,
the oxide layer can be easily disrupted. Research
has shown that one of the most corrosive
times for copper alloys is when they
cycle between a reducing and oxidizing
condition. Therefore, it is imperative that
mixed-metallurgy feedwater systems contain
sufficient reducing agent such as hydrazine
or carbohydrazide to maintain a reducing
condition at all times.
A reducing condition is not the same as
the absence of dissolved oxygen. Regardless
of how well the deaerator is functioning,
if there are copper feedwater heaters in the
system, the continuous addition of a reducing
agent is required to achieve the negative ORP
that is protective of copper alloys.
All volatile reducing agents used in utility
cycles break down at temperatures typically
associated with HP feedwater heaters or the
economizer-and certainly by the time the
water reaches the boiler. Therefore, regardless
of which reducing agent is added to the
www.powermag.com
presence of small amounts of dissolved oxygen.
Research has shown that as little as 5
ppb to 10 ppb of dissolved oxygen significantly
reduces the rate of FAC under feedwater
conditions. This occurs because the
dissolved oxygen present in the low-temperature
feedwater (from the condenser to the
deaerator) forms iron oxides that fill in the
pores of the outer layer of the magnetite, dramatically
improving its stability. Even in the
absence of any measurable dissolved oxygen,
after the deaerator, the ORP remains positive
and increases the stability of the magnetite
layer through the HP feedwater heaters and
economizer.
The formation of these more resilient
protective oxides is the basis of oxygenated
treatment, which is successfully used on all
supercritical plants in North America and
many HP drum units. However, simply discontinuing
the use of a reducing agent should
never be confused with oxygenated treatment,
where pure oxygen is purposefully
injected, the deaerator vents are closed, and
the dissolved oxygen levels in the feedwater
are an order of magnitude higher than in a
conventional feedwater system.
Stable feedwater chemistry in the absence
of a reducing agent continues to strengthen
the passive oxide layer throughout the feedwater
piping over time. Therefore, although
dissolved oxygen levels may temporarily
spike during a startup, it is also unnecessary
to add a reducing agent during layup or for
the subsequent startup. ■
-David Daniels is a POWER contributing
editor and senior principal scientist at
M&M Engineering Associates Inc.
31
http://www.powermag.com
POWER March 2015

Table of Contents for the Digital Edition of POWER March 2015
