See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/271911104

Hydrogen Peroxide and Caustic Soda: Dancing with a Dragon while

Bleaching

Article  in  Tappi Journal · July 2013

DOI: 10.32964/TJ12.7.59

CITATIONS READS

12 17,929

3 authors:

Peter W. Hart Carl J. Houtman

Westrock United States Department of Agriculture
171 PUBLICATIONS   1,039 CITATIONS    120 PUBLICATIONS   3,472 CITATIONS   

SEE PROFILE SEE PROFILE

Kolby Hirth
US Forest Service
33 PUBLICATIONS   1,206 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Conductivity impacts on processes View project

Is diffusion inhibition the mechanism of decay resistance in modified wood? View project

All content following this page was uploaded by Peter W. Hart on 08 February 2015.

The user has requested enhancement of the downloaded file.

 PEER-REVIEWED BLEACHING

Hydrogen peroxide and caustic soda:

Dancing with a dragon while bleaching
PETER W. HART, CARL HOUTMAN, and KOLBY HIRTH

ABSTRACT: When hydrogen peroxide is mixed with caustic soda, an auto-accelerating reaction can lead to gen-
eration of significant amounts of heat and oxygen. On the basis of experiments using typical pulp mill process con-
centration and temperatures, a relatively simple kinetic model has been developed. Evaluation of these model
results reveals that hydrogen peroxide-caustic soda systems are extremely sensitive to hydrogen peroxide:caustic
soda ratio, transition metal contamination, and temperature. Small changes in initial conditions can result in a closed
system becoming explosive. Analysis of model results was used to develop guidelines for safer application of hydro-
gen peroxide in a mill setting.
Application: Because peroxide at 10% cannot generate an explosive mixture with sodium hydroxide, mills that
either install new or retrofit bleaching systems should consider installing a dilution system that will take the peroxide
from the storage concentration of 50% to 10% before introduction into the mixer.

H ydrogen peroxide is an environmentally compatible ready source of oxygen; thus, typical extinguishing technol-
bleaching chemical that has been employed in kraft ogy that uses exclusion of oxygen as a mode of action (e.g.,
mills for many years. As part of a bleaching sequence, foams or dry powders) is much less effective.
hydrogen peroxide brightens and provides brightness Runaway decomposition of hydrogen peroxide has result-
stabilization, if used as the last stage in a multistage bleach ed in catastrophic events in several mills over the last 15 years.
plant [1–4]. Because hydrogen peroxide bleaching does Two bleach plants experienced pump explosions when hy-
not delignify fibers, it does not produce as many soluble drogen peroxide came into contact with caustic soda (NaOH)
by-products as other popular bleaching chemicals; thus, it in the absence of pulp and resulted in serious injuries [12,13].
can reduce organic load to wastewater treatment plants. Both mills had been using 50%-concentration hydrogen per-
Hydrogen peroxide is delivered to pulp mills via tank oxide without incident for several years.
truck or railcar. Truck deliveries are near 50% con- Other incidents, one in Europe and one in North
centration [5]. Railcar deliveries may be either at America, resulted from rapid decomposition with-
TAPPI
50% or 70% concentration. If a mill receives hy- in a storage tank when an organic chelant was
PEERS 2012
drogen peroxide deliveries at 70% concentra- mistakenly mixed with hydrogen peroxide
tion, it is routinely diluted to 50% concentration Honorable [8,14]. In those cases, the explosions were like-
as it is unloaded. Thus, industrial-strength hy- Mention ly caused by a combination of alkaline-catalyzed
drogen peroxide employed in kraft pulp bleach- decomposition and rapid oxidation of an organic
ing is generally near 50% concentration. chemical.
Under proper storage conditions, hydrogen perox- Alkali-catalyzed decomposition of hydrogen per-
ide decomposes at a rate of about 1% per year [6]. oxide follows second-order reaction kinetics, as
Traces of transition metals (e.g., iron, copper, and shown in Eq. (1) [15]. The rate of decomposition
manganese) increase hydrogen peroxide decompo- depends on the concentration of both hydrogen
sition [7], and suppliers add stabilizers to sequester these metals. peroxide (HOOH) and its conjugate base (HOO-):
Fifty-percent-concentration hydrogen peroxide solutions repre-
sent major safety hazards, such as skin burns; eye injuries; po- dP/dt = –2k (HOOH) (HOO–) (1)
tential for rapid pressurization within tanks, pipes, pumps, and/
or storage vessels; potential for fire due to formation of oxygen In Eq. (1), P is total hydrogen peroxide concentration (i.e.,
during decomposition; hydrogen peroxide-organic vapor phase P = [HOOH] + [HOO-]) and t is time. Note that the 2 is due to
explosions; and runaway decomposition [8–11]. the consumption of two moles of hydrogen peroxide for every
During startup conditions, fires resulting from the reaction mole of reaction. Makkonen reported an apparent rate of
of hydrogen peroxide with partially dried pulp on standpipe 8 × 10–3 L/mol/s at 45°C [14]. The apparent rate follows the Ar-
walls have been reported. Industrial-strength hydrogen per- rhenius theory with an activation energy of 54.8 kJ/mol [15].
oxide fires are extremely dangerous because they provide a Because the decomposition from alkaline catalysis cannot be
JULY 2013 | VOL. 12 NO. 7 | TAPPI JOURNAL 59
BLEACHING
completely separated from decomposition due to trace transi- Hg Cell Membrane ET
tion metals [16], the interpretation of these data is difficult. As
examples, Makkonen measured slower decomposition rates NaCl <10 ppm <30 ppm 1.0 wt%
for reactions stabilized with magnesium or silicate [15], and
NaClO3 0.5 ppm 3 ppm 0.15%
Galbacs and Csanyi reported rates as low as 10–6 L/mol/s using
extreme methods to minimize inorganic contamination [16]. Na2CO3 0.02 wt% 0.03 wt% 0.1 wt%
The net reaction for hydroxide ion–catalyzed decomposi-
Na2SO4 10 ppm 15 ppm 0.01 wt%
tions produces two moles of water and one mole of oxygen for
every two moles of hydrogen peroxide, shown as Eqs. (2)-(4): Iron <1 ppm 1–5ppm 5–10 ppm

I. Typical concentrations of contaminants in commercial caustic

HOOH + HOO- → H2O + O2 + OH- (2) soda [23].
OH- + HOOH → HOO- + H2O (3)
2 HOOH → 2H2O + O2 (4) that we extrapolated our kinetic model well beyond the ex-
perimental range of temperatures and concentrations and
The estimated activation energy for the rate-limiting reac- thus, errors could be substantial [11, 20]. The goal of the work
tion (Eq. [2]) is 110 kJ/mol [16]. described here is to conduct experiments with chemical con-
Because bleaching efficiency is important for process eco- centrations and temperatures similar to bleach plant operat-
nomics, many studies of transition metal–catalyzed hydrogen ing or startup conditions and then develop models without
peroxide decomposition have been published [17–19]. Unfor- excessive extrapolation.
tunately, a simple set of reactions cannot be written for iron-
catalyzed decomposition. Mechanisms involving various EXPERIMENTAL
chain reactions, radicals, reduced iron intermediates, and iron The experiments described here were conducted with ex-
complexes have been written for the iron-catalyzed decom- treme caution. In all cases, the hydrogen peroxide and sodium
position. The precise details are still the subject of active de- hydroxide were mixed in open vessels surrounded by running
bate. The net reaction for iron-catalyzed decomposition, how- water. The operator wore a face shield, rubber gloves, and a
ever, is likely the same as the alkaline decomposition (Eq. [4]), full apron. The injection of the sodium hydroxide was accom-
but the activation energy is likely significantly lower, about plished by a syringe connected to 2 m of Teflon tubing, which
54 kJ/mol [15]. allowed the operator to be away from the vessel during the
Our previous kinetic model, developed from Makkonen’s mixing. If one is considering repeating these experiments,
data [15], indicated that runaway decomposition may be the similar precautions are strongly recommended.
cause of the explosions at two chemical pulp mills [20]. It also For all experiments, commercial hydrogen peroxide (Hy-
provides a reason why chemical-pulp bleach plants tend to drite Chemical; Brookfield, WI, USA) was used without fur-
experience more catastrophic events than mechanical pulp ther purification. Using iodometric titration, a concentration
mills. One prediction of this model was a delay between mix- of 48% hydrogen peroxide [21] was determined.
ing and onset of a rapid decomposition. Figure 1 shows a Three grades of commercial caustic soda were obtained
prediction of pressure resulting from mixing 25 L of 50 wt% from PPG Industries (Lake Charles, LA, USA): mercury (Hg)
hydrogen peroxide with 25 wt% sodium hydroxide in a vessel cell, membrane, and ET (diaphragm). They represent prod-
with a restricted vent. A limitation of this previous work was ucts from three manufacturing methods [22]. Each of these
types of caustic soda likely has different concentrations of
contaminants, as shown in Table I [23].

Mercury-cell caustic soda

To produce Hg-cell caustic soda, mercury is used as the cath-
ode for the electrolysis of sodium chloride (NaCl) to produce
sodium amalgam that is reacted with deionized water to pro-
duce a sodium hydroxide solution. A 50 wt% NaOH solution
is produced directly without any additional evaporation. The
Hg-cell process produces the highest-purity caustic soda com-
mercially available.

Membrane-grade caustic soda

1. Dynamic model prediction of alkaline decomposition of The membrane-cell process produces medium-purity caustic
hydrogen peroxide in a vented mixing vessel. Conditions were soda. It uses a selective membrane during the electrolysis of
25 L of 50 wt% hydrogen peroxide with the addition of 25 wt% sodium chloride. Sodium ions diffuse across the membrane
caustic soda [20]. to the cathode while keeping chlorine gas and salt brine in
60 TAPPI JOURNAL | VOL. 12 NO. 7 | JULY 2013
 BLEACHING

Limit of used under standard conditions and with argon humidifica-

µg/g Hg Cell Membrane ET
Quantitation tion. Table II shows only iron and zinc present above the meth-
Co nd nd nd 0.04 od’s detection limit of quantitation. For most of the elements,
Cr nd nd nd 0.02
the concentration is below the detection limit of the method.
Using a colorimetric analysis [24], the iron composition was
Cu nd nd nd 0.05
also determined to be 0.2, 0.5, and 2.9 ppm for Hg cell, mem-
Fe 0.20 0.67 7.0 0.04
brane, and ET caustic soda, respectively. The deionized water
Mn nd nd nd 0.04 employed for this work was also analyzed for iron and deter-
Ni nd nd nd 0.01 mined to be below 0.1 ppm, the detection limit of the method.
Zn nd 0.16 nd 0.08
nd = not detected Visual demonstration
For the purposes of illustrating the decomposition of hydro-
II. Trace metal concentration for the caustic soda samples used
gen peroxide, 200 mL of commercial 48 wt% hydrogen per-
in this study.
oxide was added to a 2-L stainless-steel beaker. The beaker
another compartment. Sodium metal is reacted with purified was placed on a glass stand inside a much larger steel tank
water to produce caustic soda. The solution produced by the with running water diluting any overflow before going to a
membrane cell process is 33–35 wt% NaOH. The concentra- drain. A separate clamp was used to secure a K-type thermo-
tion is raised to 50 wt% solution by evaporation. The ion-se- couple and Teflon tubing connected to a syringe in the beaker.
lective membranes result in relatively low impurity levels. At time 0 s, 30 mL of 25 wt% Hg-cell caustic soda was injected
However, small quantities of salts do migrate across the mem- into the hydrogen peroxide. Time and temperature were dis-
brane and are concentrated during evaporation to a level of played using an in-house written data acquisition program
75–100 ppm depending upon membrane condition and type. driving a National Instruments (Austin, TX, USA) data acquisi-
tion board.
Diaphragm (ET)-grade caustic soda Figure 2 shows progress of the hydrogen peroxide-caus-
The least-pure form of commercially available caustic soda is tic soda reaction via a series of still frames taken over 268 s.
diaphragm- or ET-process soda. This process
uses asbestos or asbestos substitutes to sepa-
rate the coproducts caustic soda and chlorine
during the electrolysis of sodium chloride.
The diaphragm cell produces weak cell liquor
that contains 12–14 wt% NaOH and roughly
the same concentration NaCl salt. The cell li-
quor is subsequently evaporated in a multief-
fect evaporation process to a final concentra-
tion of 50 wt% NaOH. Excess salt is crystallized
and filtered through the evaporation process
for subsequent reuse or recycle. This process
produces the lowest-quality electrochemical
caustic soda solutions.
Table II shows the specific analysis re-
sults for trace metal contamination of the
caustic soda samples employed in this work
using ICP-OES (inductively coupled plasma-
optical emission spectrometry). Samples of
all three caustic grades were surveyed for the
presence of metal contamination and were
analyzed quantitatively for several metals.
Caustic samples were prepared by neutraliza-
tion with hydrochloric acid and dilution to
1:25. Standards were prepared by serial dilu-
2. Sequence of photos taken during hydrogen peroxide decomposition. The
tions of 20 ppm stock solutions with equimo-
reaction time and temperature are displayed on the computer screen. Mercury-
lar NaCl. A Horiba Ultima II spectrometer (Ir- cell grade caustic soda, 30 mL, was injected into 200 mL of hydrogen peroxide
vine, CA, USA), equipped with the whole at 0 s. The initial mixture temperature was 45°C. The temperature slowly rises
spectrum WIN-Image survey-mode software over 4 min and then rapidly accelerates over 5 s. The rapid release of energy is
and the Analyst quantitative software, was demonstrated by the large steam plume.
JULY 2013 | VOL. 12 NO. 7 | TAPPI JOURNAL 61
BLEACHING

The initial temperature of the combined reaction mixture was

45°C. After 89 s, the mixture temperature had risen to 47.6°C
(frame I). The temperature continued to slowly rise, reaching
55.8°C after 217 s (frame II). At 256 s (frame III), the reaction
appears to accelerate and rapid oxygen gas evolution turns
the reaction mixture from a clear liquid into a frothy mixture.
After 2 s more (frame IV), the mixture is overflowing the re-
action vessel and the temperature continues to rise. Two sec-
onds after that (frame V), the gas evolution and heating is so
rapid that a steam plume can be seen. Finally, (frame VI) after
268 s of reaction, the temperature has reached 96.9°C and the
reaction is beginning to subside. In summary, after a slow
temperature rise over 4 min, the reaction dramatically accel-
erates over less than 5 s and generates large amounts of steam 3. Temperature versus time for a mixtures of 35g of 25% caustic
and oxygen. Clearly, if this reaction occurred in a closed ves- soda and 200g of 48% hydrogen peroxide.
sel, an explosion would have been likely.

Calorimeter experiments
Experiments similar to the visual experiments were conduct-
ed with commercial-grade 48 wt % hydrogen peroxide, but
the stainless steel beaker was replaced with a glass vacuum
dewar to reduce the heat loss and reactions caused by the
stainless-steel container. A Teflon-coated, motor-driven stirrer
operating at 100 rpm was used to ensure complete and rapid
mixing. The three grades of caustic soda were used after dilu-
tion to 25% concentration with deionized water. For each ex-
periment a measured amount of hydrogen peroxide, approx-
imately 200 g, was mixed with various amounts of diluted
caustic soda, 35–85 g. Temperature data were collected at
approximately 8 Hz with the same computer system described
previously.
4. Initial rate at 50°C for various mixtures of hydrogen peroxide
RESULTS and caustic soda.
Calorimetric experiments were performed at various ratios of
hydrogen peroxide-to-caustic soda concentrations. Typical The results of the model can be used to estimate the initial
data are shown in Fig. 3. While the concentrations were reaction rate at 50°C due to both the alkali- and iron-catalyzed
similar, the temperature rises appear to occur at different reactions. The differences among the various grades of caustic
rates, with the caustic containing the highest iron concentra- soda become apparent. Inspection of Fig. 4 shows that for
tion (ET) being the fastest. Hg-cell caustic soda, approximately 70% of the initial rate is
The results from 21 similar trials were used to estimate the due to the alkali-catalyzed mechanism. For ET-grade caustic
rate constants for alkali- and iron-catalyzed decomposition. soda, the iron-catalyzed pathway dominates.
The rate equations were integrated using a fifth-order Runge- In a triangle diagram, Figs. 5–7, all possible compositions
Kutta method implemented in Microsoft Excel. Two parallel of a three-component mixture can be represented. Each ver-
reactions with different activation energies were assumed, and tex represents a different pure component. For example, pro-
the Solver function in Excel was used to optimize the kinetic gressing left to right along the base represents a progression
parameter values to minimize the squared error between the from pure water to pure hydrogen peroxide, with the mid-
model and the data for all the experiments at the same time. A point representing a 50:50 mixture. Because the kinetic pa-
single set of kinetic parameters was able to fit the data set span- rameters for the two reactions were estimated from the calo-
ning all 21 experiments, a total of 7500 temperature measure- rimetric experiments and the iron concentration is known, it
ments, with an average error of 0.5°C. The best fit of the data was possible to determine the time for a reaction mixture,
was obtained when the activation energies of the alkali- and with a particular composition, to reach 100°C. The results of
iron-catalyzed decomposition were set to 108 and 51.7 kJ/mol, this calculation, assuming an initial temperature of 50°C, are
respectively. The model also accounted for the heat carried represented as time contours on triangle diagrams (Figs. 5–7).
away by the evaporation of water from the mixture. For example, Fig. 5 shows that compositions enclosed by the
62 TAPPI JOURNAL | VOL. 12 NO. 7 | JULY 2013
 BLEACHING

5. Reaction model results for mercury-cell grade caustic soda. 7. Reaction model results for ET-grade caustic soda. For a
For a mixture starting at 50°C, the contours represent the time mixture starting at 50°C, the contours represent the time
required to reach 100°C. The blue line represents the operating required to reach 100°C. The blue line represents the operating
line for typical mill operation. The pink line is our proposed line for typical mill operation. The pink line is our proposed
safer operating line. Graph axes are explained in the text. safer operating line. Graph axes are explained in the text.

caustic soda, the reaction mixture could become dangerous

in less than 2 min. At higher temperatures, the reaction condi-
tions become dangerous even more quickly.
When these three triangle diagrams are compared, the im-
portance of impurities (soluble iron) in caustic soda becomes
clear. As iron concentration goes up, the reaction contour
lines move toward the water apex, suggesting that the danger
of explosion covers a wider range of mixture compositions.

Recommendations for safer practice

While installation costs tend to drive designs of hydrogen per-
oxide bleaching systems to high concentrations, building sys-
tems that are inherently safe is a better choice. By inspection
of Figs. 5–7, operating lines can be chosen that do not cross
into the regions of dangerous compositions. For example, the
dashed pink line in each of the three triangle diagrams (Figs.
6. Reaction model results for membrane-grade caustic soda. 5–7) represents the operating line for a mill employing a 10
For a mixture starting at 50°C, the contours represent the time wt% hydrogen peroxide and 25 wt% caustic soda in the bleach
required to reach 100°C. The blue line represents the operating plant. Notice that even a system using ET-grade caustic soda
line for typical mill operation. The pink line is our proposed
is safe for all compositions. Employee safety is an added ben-
safer operating line. Graph axes are explained in the text.
efit of diluting 50% hydrogen peroxide to 10%, before it enters
the mill. This lower concentration is less likely to cause per-
red contour, high hydrogen peroxide concentrations, will boil manent damage to skin or eyes.
in less than a minute and as more water is added (i.e., moving It was determined that using higher-impurity, iron-
toward the water vertex), the time to boiling goes up. containing caustic soda will render the reaction system more
Turning to bleach plant operation, all mixtures of 25 wt% dangerous. If a bleaching system employs Hg-cell grade
caustic soda and 50 wt% hydrogen peroxide will be some- caustic soda, it is possible to safely apply hydrogen peroxide
where on the thick blue lines on these triangle diagrams. up to 18 % without experiencing auto-accelerating
These lines represent the operating line when there is no pulp decomposition reactions.
flowing in the system. Notably in all three cases, the operat- Finally, if a peroxide system must be operated with 50%
ing line enters the dangerous concentration range (i.e., time hydrogen peroxide, reliable control systems must be installed
≤ 1 h) suggesting that a potential for an auto-accelerating reac- to stop the flow of hydrogen peroxide and caustic soda when
tion exists, if the stock flow were to stop. For the ET-grade pulp flow stops. Furthermore, shut-down sequences should
JULY 2013 | VOL. 12 NO. 7 | TAPPI JOURNAL 63
BLEACHING

3. Suess, H.U. (Ed.) in Pulp Bleaching Today, Walter de Gruyter, New

include methods for flushing concentrated hydrogen peroxide York, 2010, pp. 201–226.
from collection points like the pumps that exploded [13,14].
4. Brogdon, B.N., and Hart, P.W. in Bleaching of Pulp (P.W. Hart and
A.W. Rudie, Eds.), 5th edn., TAPPI PRESS, Atlanta, GA, USA, 2012,
CONCLUSIONS Chap. 10.
Hydrogen peroxide decomposition reactions are extremely 5. Hart, P.W. and Rudie, A.W., TAPPI Eng., Pulping Environ. Conf., TAPPI
energy intensive and potentially dangerous. There have been PRESS, 2007, Session 9-1.
catastrophic events at pulp mills that illustrate the danger. 6. Schumb, W.C., Satterfield, C.N., and Wentworth, R.L., “Hydrogen
This work evaluated the hydrogen peroxide decomposition Peroxide,” ACS Monograph No. 128, Reinhold Publishing, New
kinetics at typical bleach plant concentrations and York, 1955, p. 526.
temperatures. Solutions of all three grades of commercially 7. Raines, J.C., Schmidt, J.P., Burelach, J.P., et al., J. Therm. Anal.
available caustic soda—ET, membrane, and Hg cell—were Calorim. 85(1): 2006(53).
employed in these evaluations. Because these three grades of 8. Mackenzie, J., Chem. Eng. 97(6): 1990(84).
caustic soda differ in their iron levels, the impact of iron-
9. Anderson, R., Reid, D., Hart, P., et al., “Hydrogen peroxide (H2O2)
catalyzed decomposition was also quantified. This modeling safe storage and handling –TIP 0606-24,” Technical Information
effort clearly shows that by diluting hydrogen peroxide, an Paper, TAPPI PRESS, Atlanta, 2010.
inherently safe bleaching system can be designed, and 10. Solvay Chemicals, “Safety and handling technical data sheet for
current bleaching systems should be reevaluated to prevent hydrogen peroxide,” Brussels, 2006.
any further explosions. TJ 11. Hart, P.W. and Rudie, A.W., Chem. Eng. Educ. 45(1): 2011(15).
12. Hoekstra, G., “Explosion rocks mill, two workers injured by blast,”
ACKNOWLEDGEMENTS
Prince George Citizen, Prince George, BC, Canada, Feb. 8, 2007.
The authors acknowledge the contributions of Brian Brogdon
13. “Ruptured pump may figure into Evadale mill explosion,” Beaumont
and an anonymous peer reviewer for improving this manu-
Enterprise, Beaumont, TX, USA, Aug. 18, 2001.
script with their constructive comments.
14. Stora Enso, Environmental Statement 2009, Skoghall mill, Helsinki,
Finland, June 28, 2010.
LITERATURE CITED
15. Makkonen, H.P., “Decomposition of hydrogen-peroxide in
1. Davies, D., Dietz, T., and Suess, H.U., Pulp Pap. Can. 110(8): dilute alkaline aqueous solutions,” Ph.D. thesis, University of
T142(2009). Washington, Seattle, WA, USA, 1971.
2. Parthasarathy, V.R. and Brooks, G., TAPPI Eng., Pulping Environ. Conf., 16. Galbacs, Z.M. and Csany, L.J., J. Chem. Soc., Dalton Trans.
TAPPI PRESS, Atlanta, GA, USA, 2006, Session 18-1. 2353(1983).

ABOUT THE AUTHORS

Feeding 50% peroxide into a pump, which is used as
a pulp mixer, can be an inherently hazardous opera-
tion. The cause of a mill explosion has been traced
to peroxide collecting in a pump when pulp flow
was interrupted. We wanted to conduct experiments
with hydrogen peroxide and sodium hydroxide
using commercial concentrations to better under-
stand the rates of reaction and the nature of the
hazard.
Mixtures of hydrogen peroxide and sodium hy- Hart Houtman Hirth
droxide can rapidly generate heat and oxygen. We
age concentration of 50% to 10% before introduction
had to exercise great care in designing an experi-
into the mixer. As shown in this study, peroxide at
mental system that allowed us to follow the reac-
10% cannot generate an explosive mixture with
tions safely. During the course of the study, we real-
sodium hydroxide.
ized that the iron contamination of the sodium
hydroxide was affecting the rate of reaction. A fur-
Hart is manager, new technology, MWV Corp., Atlanta,
ther difficultly was accurately measuring the iron
GA, USA. Houtman is a research chemical engineer
concentration, which was 0.5–5 ppm, in 50% sodium and Hirth is an analytical chemist, USDA Forest
hydroxide. Products Laboratory, Madison, WI, USA. Email Hart
As mills either install new or refit existing bleach- [email protected].
ing systems, they should consider installing a dilu-
tion system that will take the peroxide from the stor-

64 TAPPI JOURNAL | VOL. 12 NO. 7 | JULY 2013

 BLEACHING
17. Lapiette, L., Berry, R., and Bouchard, J., Holzforschung 57(6):
627(2003).
18. Wuorimaa, A., Jokela, R., and Aksela, R., Nord. Pulp Pap. Res. J.
21(4): 2006(435).
19. Popov, E., Välisaari, J., Vuorenpalo, V-M., et al., Holzforschung 61(6):
543(2007).
20. Hart, P.W. and Rudie, A.W., Pulp Pap. Can. 111(4): T30(2010).
21. Harris, D.A., Quantitative Chemical Analysis, 4th edn., W.H. Freeman,
New York, 1995, p. 439.
22. Curlin, L.C. and Bommaraju, T.V. in Kirk-Othmer Encyclopedia of
Chemical Technology, 4th Edn., Vol. 1, John Wiley & Sons, Hoboken,
NJ, USA, 2004, pp. 477–521.
23. Olin Chlor-Alkali Company, “Caustic soda production.” Available
[Online] at http://www.olinchloralkali.com/Library/Literature/
OverviewOfProcess.aspx.
24. Skoog, D.A. and West, D.M., Fundamentals of Analytical Chemistry,
2nd. Edn., Holt, Rinehart, and Winston, New York, 1969, p. 689.

TAPPI Standards and TIPs online. On demand.

Peer Reviewed. Always Current.

Subject Areas:
Containers • Testing Practices • Structural Materials • Nonfibrous Materials
Paper and Paperboard • Fibrous Materials and Pulp
Applications:
Quality Control • Dispute Resolution • Repeatable Results
Baseline Data Verification • Knowledge Base Research

Access To Knowledge. Anytime. Anywhere.

Order your subscription - Members: $250, List Price: $485
Only $125 when purchased with Membership or Renewal

Work Smarter. Join TAPPI. www.tappi.org/standardsonline

JULY 2013 | VOL. 12 NO. 7 | TAPPI JOURNAL 65

View publication stats