Biermann Ch. Handbook of Pulping and Papermaking

Подождите немного. Документ загружается.

674 32. CORRUGATED CONTAINERS

second headbox) which gives a blotchy appear-

ance.

A second method is to use an entirely white

linerboard, which is expensive. A third method is

to coat the corrugated box.

Corrugator

finishing

operations

The continuous board must then be cut into

blanks (individual sheets) that will each make an

entire box. The continuous board first goes

through the slitter/scorer section. Here, the board

is cut and scored in the machine direction. A

series of sheeter knives make cuts in the cross

machine direction to produce the blanks.

Blanks to containers

The blanks go through a series of operations

to become containers. They go through a printing

operation. They must be scored in the cross

machine direction. Slots must be formed by

stamping out portions of the box to define the tabs

for assembly.

Overall considerations

The cost of paper delivered to the box plant

is usually 60—80% of the final cost of a box.

Paper use (especially trim and other waste) must

be minimized to stay competitive with other box

plants. Boxes are constructed to meet require-

ments of the rail (especially Rule 41) or trucking

industries (Rule 222 for common carriers). It

takes about 60—70 tons of paper to produce 1

million square feet of corrugated board. A corru-

gator with a width of 97" and operating at 600

ft/min will use over 3500 tons of paper per month.

About 5—7% of the incoming paper will

become scrap material. This is usually handled by

a pneumatic system that goes to a cyclone separa-

tor. From the separator the waste paper goes to a

baler, where it is apt to be sold to a pulp mill

using waste paper.

Scheduling orders

A wide variety and number of orders may be

processed in a relatively short time. These must

be scheduled to meet deadlines, to obtain the

maximum capacity of

machines,

to make the most

efficient use of the machine width, and other

factors. Also, the proper type and suitable amount

of paper must be on hand (a 5—6 week inventory

of paper is generally suitable). Scheduling orders

efficiently, as is any scheduling operation, can be

an extremely complex task using linear program-

ming and other tools. Computer modeling of the

scheduling process is extraordinarily complicated;

thus,

this position is of key importance to the

smooth operation of the box plant.

Partial rolls of paper that remain after a job

should be measured, recorded in inventory, and

used in the next suitable job so one is not inundat-

ed with partial rolls (called butts) in inventory.

32.3 TEST METHODS

Testing of

corrugated hoard

A variety of tests are used to test corrugated

board, the two paper components of board, or

assembled

boxes.

Some of these, depicted in Fig.

32-3,

are performed between two flat platens.

Various types of compression (or crush) tests are

critical since most corrugated boxes fail in com-

pression when stacked.

For routine quality control, sealed, empty

boxes are tested on relatively large test devices

with parallel platens (moving at 10—13 mm/min

during the test) in the box compression test, BCT,

(Tappi Standard T 804). A number of different

box designs (flute types, liner types and weights,

etc.) can be compared directly provided the boxes

are all the same size. Inner packing may be used

for testing design loads. This test best represents

the performance of most corrugated boxes in

service (i.e., where they are stacked on top of

each other). In reality, however, loading will not

be uniform, humidity conditions will vary, loading

times will be much longer; loads substantially

lower that of the test value can be used in service.

Testing of board

components

The remaining tests described here are done

on small samples on a test device such as that

shown in Fig. 7-16 on page 179. The pin adhe-

sion test (T 821) tests the strength of the bond of

the flute tips of the corrugating medium to one of

the liner boards in a sample of corrugated board.

The sample size is 2 in. by 6 in. for boards with

A or C flutes and 1.25 in. by 4 in. for boards with

B flutes. Low values for this test are caused by

poor adhesive or poor adhesive application (if

failure is at the glue line) or containerboard with

a low internal bond strength (if failure occurs in

the paper).

TEST METHODS 675

The box compression test (BCT) measures the re- The concora medium test, CMT, measures the

sistance

assembled box

compressive force, crushing resistance of a fluted sample of medium.

6 pressure pins

WMMMMln

7 support pins

The pin adhesion test, PAT, measures the resis- The corrugated crush test, CCT, is performed on

tance of

the

flute tips to separation from the liner, medium (in

a jig)

fluted

the

laboratory.

The flat crush test, FCT, measures the resistance The concora liner test, CLT, measures the edge-

of the

flutes

to a crushing force. wise compression in a supported sample of liner.

The edge crush test, ECT, measures the compres- The ring crush test, RCT, measures edgewise

sion strength as in boxes that are stacked. compression in a ring of liner supported by

jig.

Fig. 32-3. Some common tests performed on corrugating medium, linerboard, corrugated board,

and completed boxes. One test machine is shown in Fig. 7-16 on page 179.

676 32. CORRUGATED CONTAINERS

The flat crush test (T 808) measures the

resistance of the flutes to crushing force from a

sample cut from manufactured corrugated board.

T 808 specifies a circular sample of

or 10 in.^ in

area. Low values for this test might be due to

medium of low strength, crushed flutes, or flutes

that have not been well formed (such as leaning

flutes).

This test is not suitable for double— or

triple—wall boards since the inner liner(s) are

subject to lateral motion during the test, but would

be much less so in products in

use.

This test does

not simulate forces applied to boxes in use, so the

correlation between this test and how a box will

perform in service is not always high. The bend-

ing stiffness of the corrugated board should com-

plement the FCT (Markstrom, 1992).

The edge crush test or edgewise compression

test, ECT (T 811) also uses a sample cut from

manufactured board. The shape may be rectangu-

lar or as that shown in the Fig. 32-3; the latter

shape is designed to give a failure away from the

edge.

The force is applied parallel to the flutes,

just as in a container in service. This test is a

good indicator of the general performance of

box

in transportation. ECT can be related to BCT by

use of the McKee (1963) equation (Markstrom,

1992).

There are many modifications to this test

which are discussed and quantitatively compared

by Markstrom (1992).

The remaining tests described here are

designed to test the properties of the medium or

liner, i.e., they are quality control tests of the raw

materials. The sum of the compression strengths

the

liner and medium layers correlates well with

the compression strength of the corrugated board.

The short—span compression test, SCT, has

been described on page 177. T 826 specifies a

sample 15 mm wide. This test provides similar

information to other compression tests except

sample geometry does not enter into the results.

Geometry is an issue for at least two reasons:

slender samples, such as a straw in the lengthwise

direction, fail by buckling long before the material

fails by compression; samples that fail at the edge

give lower strength results than are truly

representative. For this reason it is replacing

other, older compression tests such as those

described below (Markstrom, 1992).

The

concora medium

test, CMT, or flat crush

test [T 809 specifies a sample 0.5 x 6.0 in. (12.7

X 152.4 mm)] measures the crushing resistance of

a single—face sample prepared in the laboratory.

It tests the medium's resistance to crushing.

The corrugated crush test, CCT, or fluted

edge crush measures the edgewise compression

strength of a piece of corrugated medium. This

provides a measure of the corrugating medium's

contribution to the compression strength of a

corrugated box. It is an alternative to the ring

crush test (T 818). T 824 calls for a sample 0.5

in. by 6 in. with the longer dimension in the

machine direction. The sample is held in a speci-

men holder during the test. The concora liner

test, CLT, is very similar to the CCT, except the

sample is not corrugated.

The ring crush test, RCT, measures the

edgewise compression

a ring of liner or medium

that is supported in the groove of a jig. Two

variations of the test are T 818 (flexing beam) and

T 822 (rigid platen).

32.4 ANNOTATED BIBLIOGRAPHY

Corrugated containers—thermal

resistance,

misc,

Bormett, D.W., Overall effective thermal

resistance of corrugated fiberboard contain-

ers,

USDA For. Ser. Res. Pap. FPL 406

(1981),

12 p. The effective thermal resis-

tance of corrugated fiberboard containers was

determined versus air velocity (0—5.4 m/s)

and board thickness. Differences were noted

for heating versus cooling for boards less

than 20 mm thick. This information is useful

for consideration of how produce and frozen

foods will behave during shipping. The

effective thermal resistance of a corrugated

fiberboard package was found to be about

0.18 K-m^/W (although the outer boundary

air and box—product interface may effective-

ly double or triple this value) for a kraft

linerboard package (0.33 mm thick) to 1.10

for a corrugated fiberboard box having walls

51 mm thick as shown in Table 32-1.

TAPPI Technical Information Sheets (TIS)

with numbers 0303 to 0305, and several 0306

ANNOTATED BIBLIOGRAPHY 677

to 0309 cover various aspects of corrugating

medium and starch adhesive preparation.

DiDominicis, A.J. and G.H. Klein, Corru-

gating, in Pulp and Paper Chemistry and

Chemical Technology, 3rd ed., J.R. Casey,

Ed., 1983, Wiley, New York, pp 2373-2398.

Markstrom, H., Testing Methods and Instru-

ments for Corrugated

Board,

3rd ed.,

Lorentzen & Wettre, Stockholm, 1992, 77p.

While promoting the publisher's test instru-

ments, it is a useful, relatively unbiased

source for the theory of the test methods. It

is usually available on a complimentary basis.

McKee, R.C., J.W. Gander, and J.R.

Wachuta, Compressive strength formula for

corrugated boxes, Paperboard Packaging

48:19,

149-159(1963).

Table 32-1. OveraU effective thermal resistance

(values listed are R or K-m^/W) of a cubical

corrugated container. From USDA For. Ser.

Res.

Paper FPL 406 (1981).

Board

caliper,

1 mm

0.00

0.33

4.23

12.7

25.4

50.8

Air velocity, m/s

0.0

0.33

0.42

0.45

0.61

0.80

1.10

1.8

0.14

0.19

0.31

0.50

0.73

1.04

5.4

0.13

0.18

0.28

0.50

0.70

1.04

MISCELLANEOUS TOPICS

33.1 VISCOSITY AND SURFACE TENSION

OF WATER FROM 0 TO lOO^'C

CRC Handbook of Chemistry and Physics

(67th ed.) gives the viscosity of water from 0° to

100°C based on a contribution from the National

Bureau of Standards that is not subject to copy-

right. The viscosity of water is given in Fig. 33-1

and is calculated from 0°—20°C [Hardy, R. C,

and R. L. Cottington, /. Res, NBS 42:573(1949)]

as:

The viscosity of water from 20° to 100°C is

calculated according to Swindells, J.F., NBS,

unpublished data as:

1.3272(20-T)

- 0.001053(r-20)^

Tly=l. 002X10 ^^105

The surface tension of water against air is

and the dielectric constant of water (e) (Lange's

Handbook of Chemistry and Physics, 13th ed.,

page 10-99) are also given in Fig. 33-1.

^ =10 998.333+8.1855(r-20) +0.00585(r-20)^

-1.30233)

Temperature, Celsius

Fig. 33-1. Viscosity, surface tension, and dielectric constant of water from 0 to

100**

678

VAPOR PRESSURE OF WATER 679

33.2 VAPOR PRESSURE OF WATER FROM 0 TO 280^C

The absolute vapor pressure of water (psia) is shown from 0—100°C (Fig. 33-2) and 100—180°C

(Fig. 33-3). The following equation is useful for calculating the pressure to within 0.05 psi of published

data for 100-~180°C (Handbook of

Chemistry

and Physics, 40th ed., pp. 2328—2330, CRC Co.).

psia = -6.487 + 0.1229 T -h 0.000599566 T" - 0.0000171538 T^ + 0.0000002004133 T^

32F 50F 68F 86F 104F 122F 140F 158F 176F

194F 212F

800

20 30

40 50 60

Temperature, Celsius

Fig. 33-2. Vapor pressure of water from 0 to 100°C.

212F 248F 284F 320F 356F 392F 428F 464F 500F 536F

•1000

240

260

•900

-800

•700

•400

•300

•200

280

100

•0

-600 .^

•500

160 180 200 220

Temperature, Celsius

Fig. 33-3. Water vapor pressure as a function of temperature from 100 to 280°C with expanded

scale for pressures below 100 psia.

680 33. MISCELLANEOUS TOPICS

33.3 PROPERTIES OF AIR AND WATER

The properties of air are particularly impor-

tant for combustion processes. The composition

of air is given in Table 33-1. General properties

of air are given in Table 33-2. The importance of

water to the industry also goes without saying; the

properties on water are summarized in Table 33-3.

33.4 WATER CONDITIONING

BETZ® (Trevose, Pennsylvania) has pub-

lished BETZ Handbook of Industrial Water Condi-

tioning, which is a very good source of informa-

tion on this topic. Much of the information in the

sections on water treatment is a condensation of

this extensive work.

Table 33-1. Content of dry, atmospheric air.

Component Content,

Volume

Nj 78.08

O2 20.95

Ar 0.94

CO2

0.033

Others

0.003

Mass

75.52

23.14

1.29

0.05

Table 33-2. Properties of dry air at 25°C.

Property

Average molecular weight

Speed of sound

Specific heat

Value

28.964 g/mol

340 m/s

0.240

cal/g/K

Table 33-3. Properties of water at 25°C.

Property Value

Specific heat

Heat of fusion (freezing, 0°C)

Heat of vaporization

Density

Index of refraction (sodium light)

0.998

cal/g/K

79.72 cal/g

540 cal/g

0.997

g/cm^

1.3329

Water treatment methods discussed in this

section are often termed external treatments since

they are designed for large quantities of water not

designated for use inside boilers or for use in

other specific applications.

Water must be properly conditioned to pre-

vent dangerous corrosion and scale formation in

boiler tubes that carry water, to be used in cooling

systems, and if it will enter or be reused within

the mill. Wastewater treatment has already been

addressed in Chapter 11.

Contaminants

in water

The theoretical basis of corrosion was pre-

sented in Section

13.11.

A review of this section

will help explain why certain impurities of water

cause corrosion. The solubility products of some

compounds were presented in Section 13.8 and

help explain why certain contaminants tend to

cause scale formation. Section 14.7, a discussion

on coordinate chemistry, is relevant to this section.

Contaminants in water may be classified as

dissolved solids, dissolved gases, and suspended

solids (which make water appear turbid or murky).

In some cases the distinction is not clear since

hydrogen sulfide may occur as the dissolved solid

of the metallic salt. Suspended solids, which can

easily cause deposits, are removed by filtration or

settling, often after clarification treatments.

Any ionic material in water increases corro-

sion by increasing the conductivity of water.

Insoluble ions, especially most calcium salts that

decrease in solubility with increasing temperature,

will contribute to scale. Water hardness is a mea-

sure of the calcium and magnesium content.

Other culprits or synergistic elements include

aluminum (carried over from poor clarifier opera-

tions),

magnesium, silicates, and sulfates. In the

Southern U.S., barium introduced through the

water and wood supply contributes to scale forma-

tion. Acids and bases are ionic in nature and

contribute to corrosion. These materials can be

removed by ion exchange or other demineral-

ization treatments. Chelating agents such as

EDTA bind with and solubilize the worst scale-

forming metal cations.

Dissolved gases such as oxygen and carbon

dioxide increase corrosion. These can be removed

by deaeration. Corrosion inhibitors decrease their

WATER CONDITIONING 681

damage. Oxygen increases pitting and other forms

of corrosion. Carbon dioxide causes corrosion in

steam condensate systems.

A variety of materials cause fairly specific

problems. Ferrous or ferric salts will interfere

with wet end chemistry and decrease the bright-

ness of paper. Ammonia complexes with copper

and zinc alloys, making them soluble. Work in

my laboratory shows that relatively small amounts

of fluoride can decrease rosin sizing with alum.

Aeration

Aeration of water is used to remove dissolved

gases by "dilution" or flushing them away. Since

air contains large amounts of nitrogen and oxygen,

these gases will remain in the water after treat-

ment. Aeration decreases the amount of free CO2

and NH3. Carbonates and ammonium salts,

however, will remain depending upon the pH of

the water. Aeration also precipitates iron as ferric

hydroxide and manganese as manganese dioxide.

Chlorination

Chlorination of water is used to kill microor-

ganisms with adverse health effects. It is used to

produce drinking water. Some mills chlorinate

their water supply in order to decrease the growth

of slime fungi and bacteria. The active species is

hypochlorous acid. The chemistry of chlorine is

covered in Section 17.2. Chlorine is occasionally

used as an oxidant to remove iron and manganese

from water. Chloride, however, causes scale and

corrosion in boilers.

Clarification

Clarification of fresh water is used to prepare

water for domestic and industrial use. Effluent

may also be treated by this method. This method

is used to remove suspended solids by coagulation.

Inorganic salts of aluminum, iron, magnesium, or

silicates can be used. Bentonite (aluminum sili-

cate) is also used. The mechanism involves

colloidal chemistry, which is discussed in detail in

Chapter 21.

Polyelectrolytes, water—soluble polymers

with ionized groups for charge interactions, are

also used. The charged groups may be anionic,

cationic, both cationic and anionic, or only slightly

ionic.

The sludge obtained with polyelectrolytes

tends to be of much lower volume than those

obtained with alum treatment. Low molecular

weight cationic polyelectrolytes (such as

polyamines below 500,000) are used for primary

coagulation. High molecular weight ones (above

20,000,000) are used for flocculation by bridging

particles together. After flocculation, the floes

must be separated from the water by conventional

clarification units.

Filtration

Filtration is used after clarification of water

for drinking or boiler purposes. Water passes

downward through a sand and/or anthracite bed

through a total depth of

to 80 cm (15 to 30 in.)

A multilayer gravel bed supports the filtering

media and allows for collection of the filtered

water. These may be gravity or pressure filters.

Do not use cold water to backwash a hot pro-

cess filter since thermal compression and expan-

sion will cause leaks to develop at the flanges and

oxygen in the cold water will cause corrosion.

Precipitation

softening

Water hardness is a measure of

the

amount of

calcium and magnesium present and is reported as

parts per million (as calcium carbonate). Moder-

ate to high hardness water of 150—500 ppm can

be partially softened by precipitation of the offend-

ing species. Calcium and magnesium associated

with bicarbonate anions are termed carbonate or

temporary

hardness since they can be removed by

heating. Calcium and magnesium associated with

chloride or sulfate are termed permanent or

noncarbonate hardness. Lime or sodium

aluminate may be added to raw water to precipi-

tate noncarbonate salts of calcium and magnesium.

This can be done under a variety of conditions

depending upon the quality of the raw water and

the purpose of the treatment.

Cold lime softening uses calcium hydroxide

to precipitate magnesium ions as magnesium

hydroxide. Sodium carbonate is used to precipi-

tate calcium ions as calcium hydroxide. Residual

calcium is about 35 ppm. Since calcium carbonate

and magnesium hydroxide have lower solubility at

high temperatures, hot lime softening can decrease

magnesium to 2 ppm and calcium to 25 ppm.

Ion exchange

Ion exchange resins are used to soften water

by replacing the cations with sodium ions (and

682 33. MISCELLANEOUS TOPICS

possibly the anions with chloride ions) of sodium

chloride. They may also be used to demineralize

water where the cations are replaced by

H"*^

ions

and the anions are replaced by OH" ions.

In 1905 the German chemist Gans used

sodium aluminosilicate materials {zeolites) to

soften water. These materials can absorb calcium

or magnesium cations and liberate sodium ions,

the basis of water softening. The zeolite could be

regenerated by running concentrated NaCl solution

through it to liberate CaClj and MgClj. While

aluminosilicates are seldom used anymore, the

term zeolite softener is occasionally used for any

cationic exchange resin.

In 1944, cationic exchange resins were devel-

oped based on polystyrene crosslinked with about

6—8% divinylbenzene. This resin had a much

higher capacity than previous resins, meaning that

the same volume of resin could exchange many

times more ions than previous materials. In 1948,

the anion exchange analog was developed. Using

these two resins in sequence allows the complete

demineralization of water, instead of merely ex-

changing cations for sodium ions. The

macroreticular anion resins that have discrete

pores are used first in the sequence to remove

organic materials from the raw water that would

otherwise cause fouling of the standard gelular

resins.

Many modifications of the original poly-

styrene divinylbenzene have been made to accom-

plish a wide variety of separations (such as the

determination of molecular weight distributions of

pol5aners described in Section 18.4).

Figure 33-4 shows the structure of poly-

styrene divinylbenzene resin. The functional

groups used in the major types of ion exchange

resins are also showed. In cationic exchange

resins,

the counter anion is tied to the resin and

not free to migrate. Cations in solution can

exchange with the resin cations until all the resin

cations have been exchanged. At this point, a

solution containing the desired cation in relatively

HCzzzCHo

HClZZlCHo

DIVINYL-

BENZENE

MONOMER

FOR CROSS-

UNKING

OH—

CH2-

CH—

CH2-

CH—

CH2-

CH—

CH2-

CH—

CH2-

CK—

CH2-

CH—

CH2-

6 6 6 6 6 6 6

CH—

CH2-

CH—

CH2-

CH—

CH2- CK—

CH2-

CH—

CH2-

CH—

CH2-

CH—

CH2—

6 6 6

CH—CH2-CH—CH2-CH—CH2-CH—CH2-CH—CH2-CH—CH2-CH—CH2—

D] O O

-CH—CH2^

POLYSTYRENE DIVINYLBENZENE BACKBONE

STRONG CATION (SO)

SULFONIC

STRONG ACID

CATION

EXCHANGE

RESIN

R-C00~

K*"

WEAK CATION (WC)

CARBOXYUC

WEAK ACID

CATION

EXCHANGE

RESIN

R N—R'

(R* may be H)

WEAK ANION (WB)

AMMONIUM

WEAK BASE

ANION

EXCHANGE

RESIN

"HOH

CH,

I «J n

-N-CH-,

'3 J

0H~

TYPE I STRONG ANION (S

QUATERNARY

AMMONIUM

STRONG BASE

ANION EXCHANGE

RESIN

?"3

-N-CH^

OH"

CH2CH2OH

TYPE II STRONG ANION (SB)

QUATERNARY

AMMONIUM

STRONG BASE

ANION EXCHANGE

RESIN

Fig. 33-4. Structures of polystyrene diyinylbenzene ion exchange resins.

WATER CONDITIONING 683

Table 33-4. Boiler feedwater guidelines.

Drum

pressure

psig

0-300

<450

<600

<750

<900

<1000

<1500

<2000

—Feedwater

-Concentration, parts/billion-

Iron

100

Cu Hardness

total as CaCOj

50 300

25 300

20 200

15 100

15 50

10 0 as OH-

IO 0 as OH"

Boiler

ppm

SiOi

150

high concentration must be used to regenerate the

column by backwashing. For water softening the

column can be regenerated with about 10% sodi-

um chloride. For demineralization, an acid such

as hydrochloric acid must be used. Since not all

of the hydrochloric acid will be absorbed by the

column, an acid waste stream is created.

33.5 BOILER FEEDWATER TREATMENT

The treatment of water for boilers began with

the invention of the first practical steam engine by

James Watt in 1763 (patented in 1769). Although

Watt alluded to the possibility of high pressure

engines in his patent, he usually used steam pres-

sures on the order of 50 kPa (7 psig). Engineers

soon found that adding a few potatoes to the boiler

reduced the amount of scale formed in the boiler.

With the development of higher temperature

and pressure

boilers,

water treatment became more

sophisticated. From 1857 to 1900 numerous

patents described the use of tannins and other

materials for treating boiler water. The use of

disodium phosphate and trisodium phosphate came

during this time period. Still, boiler explosions

killed hundreds of people each year during this

period of time due to corrosion and alkaline

embrittlement of metal. The buildup of scale

inside boiler tubes decreases heat transfer, so that

the metal tubes will be much hotter, leading to

higher rates of corrosion and reduced strength.

Boiler feedwater is purified to remove con-

taminants that would interfere with the safe opera-

tion of boilers. Methods used to purify boiler

feedwater include those already discussed and

additional methods. Additives are used in the

feedwater to counteract those impurities that

remain in the feedwater.

Specifications

Table 33-4 gives some guidelines of the

ASME Research Committee on Water in Thermal

Power Systems in terms of the tolerable amounts

of impurities in boiler waters.

Deaeration

Deaeration of water is accomplished by

decreasing the concentration of the offending gas

species

(O2,

COj, NH3, HjS) over the water some-

times with heating of the water, which decreases

the solubility of most gases (e.g., Fig. 11-1 for

oxygen). This can be accomplished by applying a

vacuum over the water, a method suitable for

water distribution systems. However, the gases

over boiler water are usually purged (flushed

away) with the use of steam. Steam does not

contaminate the water and raises the temperature

of the water (which must be done anyway), there-

by decreasing the solubility of the gases.

Oxygen scavengers

Oxygen may be introduced into the system

inadvertently even with proper preparation of

boiler feedwater. Trace amounts of oxygen can be

removed by oxygen scavengers. Sodium sulfite is

the most commonly used additive. It reacts with

oxygen to form sodium sulfate: INajSOj + O2 ->

2NaS04. It is added continuously at about 10 lb

per lb of oxygen in the feedwater. The reaction is

quick at temperatures above 100°C (212°F). The

reaction is most rapid at a pH of 9—10. Catalysts

such as the divalent cations of the transition ele-

ments are often used.

Hydrazine (N2H4, which is converted to

nitrogen and water) is also used as an oxygen

scavenger. Because it reacts slowly, it is useful

only in high—temperature applications (900 psi

pressure corresponding to 278 °C or 532 °F) where

sodium sulfite may decompose and sodium sulfate

is undesirable. Hydrazine also reacts with iron to

give a protective coating of magnetite (Fe304).

Other boiler feedwater additives

Maintaining the pH of boiler water in the

range of 9 to 11 decreases the rate of corrosion;