"The Darkening and Bleaching of Paper by Various Wavelengths in the Visible and Ultraviolet 11*

R. L. Feller, S. B. Lee and J. Bogaard Research Center on the Materials of the Artist and Conservator

Mellon Institute Pittsburgh, PA 15213

Introduction

These remarks will present some of the information that we are gaining from our own investigations at Mellon Institute, but mainly will review pertinent data that have been reported in the literature regarding the darkening and bleaching of various types of paper by ul­traviolet and visible radiation. The subject is a challenging one be­cause both darkening and bleaching may take place at the same time. Perhaps we can start this discussion no better than by showing some data that were derived from a previous investigation sponsored by the National Museum Act.Cl) The results may not represent more than the behavior of these particular paper samples, but they will certainly stimulate discussion. In Figure l, we have plotted the tendency to darken or bleach as a function of the initial Tappi brightness of various pulps (the reflectance at 457 nm). The very dark, largely lignin­containing, papers tended to bleach under 11daylight 11 fluorescent lamps, whereas the handsheets of intermediate brightness tended to darken, particularly when they contained extensive oxidized material as measured by hot-1%-alkali-soluble (HAS) matter.(l) In the review that follows, we 1 11 try to organize some of the facts concerning just when one may ex­pect darkening; when bleaching.

As we said a number of years ago, one of the first things that one should ask in any bleaching problem is: What is it -- what is the chemical substance -- that you are trying to bleach?{2) Is it l ignin? Lignin discoloration products? Rosin? Other? Is. it the degra­dation products of cellulose and hemicellulose? Are the colored products the result of heat- or light-induced changes? Until we know a lot more about this subject, we suggest that one always specify whether one's conservation problem, or the research report being studied, involves papers high in lignin or practically free of lignin, and whether the discoloration had been caused primarily by ultraviolet radiation or by heat.

One should also carefully consider the type of radiation being used to carry out the bleaching. Is it a narrow range of wavelengths? A broad set of wavelengths perhaps including ultraviolet as well as visible? Middle or near ultraviolet? These are questions that one should particularly ask when reading the literature. Some effort should be made also to determine whether the temperature was or will be much above 3O°C and whether the humidity is high or low.

*Paper given before the Book and Paper Group at the Annual Meeting of AIC, Milwaukee, May 1982.

65

-2-

Importance of Wavelength in Papers High in Lignin

Let us begin by showing the results of some rather simple bleaching tests of our own involving various wavelengths of radiation and various types of paper . Figures 2 and 3 confirm the statement that we emphasized earlier{l): under mild conditions lignin tends to bleach. Note in Figures 2 and 3 that considerable bleaching takes place under visible radiation alone (under the UF-4 filter). Even under a filter that transmits only visible radiation above 460 nm, bleaching takes place.( This is just as predicted by(Nol~n, Van den Akker and Wink in 1945 3), later confirmed by Leary. 4,5) It is the ultraviolet that may cause darkening, as we see in Figure 4, where the unfiltered "daylight" fluorescent lamps cause some darkening in ground­wood pulp. We believe that this may be due to the fact that this par­ticular groundwood pulp sample had a high concentration of alkali­sensitive groups, as evidenced by the high hot-1%-alkali (HAS) matter,(l) rather than having a high percentage of lignin. The fact that there is

little change observed when the sheets made from this pulp was irradiated by visible radiation alone may be due to the fact that both the darken­ing and bleaching effects were balancing out. The fact that both pro­cesses tend to go on at the same time and that short-wavelength ultra­violet light tends to cause darkening of pulps that contain lignin is demonstrated in Figure 5; here, the same unbleached pulp shown in Figure 2 tended to darken at first; later,bleaching took over. This curve seems rather good evidence that the two processes of darkeninq and bleaching are going on at the same time. This same pulp, when bleached with chlorine dioxide or calcium hypochlorite,tended also to darken first, then bleach,when exposed to 11daylight 11 fluorescent lamps.

These simple experiments indicate how important it is to con­sider the influence of wavelengths of radiation involved. Perhaps no better presentation of the basic facts can be cited than the work of Leary(4), shown in Figure 6 and in Tables 1 and 2. Leary used a mono­chromator to irradiate newsprint with narrow-band-widths of radiation (10 to 20 nm). As Figure 6, a qualitative representation of the infor­mation in Tables 1 and 2, indicates, darkening of newsprint, a pulp that undoubtedly contains considerable lignin, takes place principally under radiation between 360 and 390 nm in the near ultraviolet (defined as 320-400 nm) and bleaching occurs at wavelengths perhaps between 420 and 510 nm in the visible. We see in Figure 6 how confusing the results might be if one wished to irradiate such a paper with a source that contained both near ultraviolet and visible,such as the radiation from a carbon-arc fadeometer. One would have every reason to expect that both darkening and bleaching were going on at the same time under such a source or under a filter that transmitted radiation above 385 nm, as Nolan, et. al. used.(3) With such a filter, bleaching first occurred, then darkening in their experiments, leading them to conclude that there were at least two reactions going on.

The darkening of pulps that contain lignin was very well investi­gated in 1945 by Nolan and his co-workers .. (3) We see their results in

-3-

Figure 7 where the absorption curve of various lignin-containing liquors is compared to the sensitivity of lignin to discoloration (called 11fading 11

by these authors, but really referring to darkening). Figure 7 clearly indicates what we mean when we say that lignin tends to darken as in­creasingly lower wavelengths of ultraviolet are employed.

Figures 6 and 7 scarcely begin to tell tbe story; we see in Figure 8, taken from the work of Lin and Kringstadl6), that irradiation by the near ultraviolet caused the extinction coefficient to decrease in the ultraviolet region; in a sense, the lignin is fading in the ul­traviolet. If one looks carefully at the curves in the region of 400 nm, however, one notices that the tail of the absorption curve after exposure is actually slightly higher as it enters the visible region. Apparently we have fading in the ultraviolet and increased absorption (darkening) in the visible.

Indeed, the change in the spectral absorption and reflectance curves is complex. If one simply measures the spectral reflectance curves in the visible region as, for example, in the case of wood pulP,s being extracted,in the work of Polcin and co-workers seen in Figure 9(7), we find that subtle changes in the reflectance curves can hardly be interpreted. If the two sets of curves are subtracted, however, as is done in their Fig. 7, we display the spectral absorption curve of the extracted material. Subtraction of curves has frequently proven useful in discoloration work, for by means of subtraction the nature of the colored substances that were developed or removed can be detected.

In Figures 10 and 11, taken from the work of Claesson, et. al .( 8), we see in detail the results of discoloration caused by 365 nm uITra-=­violet and bleaching brought about by the visible wavelength of 436 nm. These data represent the differences in percent reflectance after ir­radiation. In these curves, we see bleaching in the ultraviolet and darkening in the visible, with a peak of absorption around 430 nm. The shape of these absorption curves, in a sense, describes the colored matter that was formed upon irradiation by the ultraviolet. We also see the changes brought about during bleaching by 436 nm light.

Figures 10 and 11 (the authors' Figures 14 and 15) clearly demonstrate that one cannot learn very much about the discoloration and bleaching of pulps by consideration of the visible region alone (such as shown in Figure 9). It is in the ultraviolet that the absorptio9 peaks, which depend upon the particular substances involved, occur.t9)

Changes in Pure Cellulose

Thus far we have been considering the changes in the spectra of pulps high in lignin.(9) The work of Silvy and LeNest(l0), on the other hand, specifically concerned the darkening of pure cellulose. Referring to Figure 12, we see, in Silvy and LeNest's Figure 2, the

67

-4-

effects of increased heating of cellulose and note the development of absorption peaks at 260, 310 and 420 nm. (PNCR% is essentially the relative post-color number, tK/S/ K/S.) Curves l, 2 and 3 in this figure are the results of heating the cellulose at 120°C in the dark for 13 hours, 148 hours and 3 months. Curves 4, 5 and 6 show the ef­fects of exposure to simulated 11daylight 11 in a Xenotest apparatus for 40 minutes, 6 hours and 70 hours, respectively. We know from our own work that thermally discolored cellulose tends to bleach, as stated earlier.(l) The authors point out that the peak absorption at 420 nm disappears most readily, the peak at 310 nm less readily and that at 260 nm is scarcely reduced at all by the light. This is a particularly illuminating illustration of the complexity of the darkening and bleach­ing phenomenon and shows clearly the importance of considering the ab­sorption spectra in the ultraviolet in order to learn what is going on.

As mentioned, the work of Silvy and LeNest concerns the ef­fects on thermally-darkened cellulose, not involving lignin. Their Figure 4, curve number 2 (in our Figure 12), shows the result of heating cellulose that had been oxidized by sodium periodate. Such material is known to have a high concentration of aldehyde, and perhaps some ketonic and acidic, groups; the authors wished to show that the absorp­tion peaks, when this chemically oxidized material is discolored by heat (curve 2 in their Figure 4), are much the same as those found in aged papers. Curve number l in their Figure 4 is based on the differ­ence in reflectance just after the cellulose had been oxidized by the periodate; the peak at 260 nm, they point out, is likely to be due to carbonyl groups for it is present in this chemically-oxidized cellulose even before thermal aging.

Rate of Change; Post-Irradiation Effect

Having stressed that the most significant changes in the ab­sorption spectra of pulps must be observed in the ultraviolet region, what else can we say on the basis of what has been published in the literature concerning the bleaching of darkened pulps by vi~ible radia­tion? We can point out, from the work of Claesson, et. al .l8), that the darkening caused by near ultraviolet takes placeveryrapidly at first and then tends to slow down, perhaps reaching a constant value (Figure 13; the authors' Figure 4). The same is true for bleaching; very rapid at first and then becoming negligible as the exposure is increased (Figure 14).

We also note that there is likely to be post-irradiation dar­kening. This is illustrated in Figures 10 through 14. It is a pheno­menon that was also pointed out by Silvy and LeNest.(10) Their Fig. 3 (our Figure 12), curve 1, shows the absorption curve after heating pure cellulose; curve 2 is the result after bleaching in 11daylight 11 for 70 hours; and curve 3 is the intense absorption after reheating that paper in the absence of light for 25 days. Claesson's results reported in

68

-5-

Figures 10, 11, 13 and 14 are particularly convincing. At the present stage of our knowledge regarding the bleaching of papers by visible radiation, we believe that the possibility of a post-irradiation effect that is, thermally-induced darkening that can occur after the papers have been exposed to light -- is the aspect of this procedure that must be investigated most carefully in the immediate future.

Summary

What can be said at the end of this brief review of the sub­ject? In general, when moderate temperatures are involved, lignin tends to be darkened by ultraviolet radiation of decreasing wavelength, as shown by the work of Nolan, Van den Akker and Wink in 1945; discolored papers tend to be bleached by visible radiation, perhaps very much as Leary demonstrated for newsprint in 1967 and 1968. If radiation that contains both ultraviolet and visible is involved, both bleaching and darkening processes probably go on at the same time. If either effect predominates, the change appears to be rapid at first, and frequently tends towards a limit in the net change at any given temperature and humidity. The most significant changes in absorption tend to take place in the ultraviolet region; investigation of bleaching or darkening effects by measurement of changes in color and Tappi brightness alone are not likely to lead to major enlightenment. Until much more is learned about the processes, it is recommended that the wavelengths of radiation being employed be specified, the temperature and humidity of the sam-ples noted and consideration given as to whether one is attempting to bleach primarily thermally or photochemically discolored lignin, degra­dation products of cellulose and hemicellulose, or both. An important aspect of photochemical bleaching of paper for future study is a post­irradiation effect, darkening of the sheets that may largely be ther-mally induced.

The authors gratefully acknowledge that this work was sup­ported in part by a grant from the National Museum Act in FY 1981.

69

-6-

References

1. Lee, S. B., Bogaard, J. and Feller, R. L., 11Concerning the Exposure of Paper to Light: Discoloration of Handsheets of Known Initial Lignin and Hot-Alkali-Soluble Content11 , Paper 81/14/4, 6th Triennial Meeting, ICOM Connnittee for Conservation, Ottawa (September 1981).

2. Feller, R. L., 11Notes on the Chemistry of Bleaching 11 , Bulletin of the American Group - IIC, l!_, No. 2, 39-56 (1971).

3. Nolan, P., Van den Akker, J. A. and Wink, W. A., 11The Fading of Ground­wood by Light. II. The Physical Mechanism of Fading11 , Paper Trade Journal, 121, No. 11, Tappi Section 101-105 (1945).

4. Leary, G. l., 11The Yellowing of Wood by Light 11 , Tappi, 50, 17-19 (1967).

5. Leary, G. L., 11The Yellowing of Wood by Light: Part II", Tappi, B_, 257-260 (1968).

6. Lin, S. Y. and Kringstad, K. P., 11Photosensitive Groups in Lignin and Lignin Model Compounds11 , Tappi, g, 658-663 (1970).

7. Polcin, J., Wayman, M., Anderson, C. B. and Rapson, W. H., 11Sapwood and Heartwood Groundwood of Western Hemlock and Jack Pine. Part l. Influence of Extractives on Optical Properties 11 , Pulp and Paper Magazine of Canada, 69, No. 21, T410-T417 (1969).

8. Claesson, S., Olson, E. and Wennerblom, A., 11The Yellowing and Bleach­ing by Light of Lignin-rich Papers and the Re-yellowing in Darkness11 ,

Svensk Papperstidning, IL, No. 8, 335-340 (1968).

9. Pew, J.C. and Connors, W. J., 11Color of Coniferous Lignin 11 , Tappi, 54, No. 2, 245-251 (1971).

10. Silvy, J. and LeNest, J. F., "Mise en evidence par diffusion de la lumiere d 1 un effet photochrome de la cellulose 11 , Comptes Rendus Des Seances De L'Academie Des Sciences, Paris, Series C, Sciences Chimiques, 271, 221-224 (1970); see also 11A Photochromic Effect Shown During the Ageing of a Cellulose Material by Heat and Light 11 , Transactions of the Symposium held at Cambridge, September 1973, ed. F. Bolam. Technical Division, The British Paper and Board Industry Federation, Plough Place, Fetter Lane, London EC4A lAL (1976).

70

-7-

Table I. Effect of Wavelength on Yellowing of Newspaper (from Leary, 1967)

Wavelength of Irradiation, nm

330

342

355

360

365

370

375

380

385 390

Observed Effect

Nil Nil Nil

Very slight yellowing Slight yellowing Yellowing Yellowing Yellowing Yellowing Very slight yellowing

Table II. Effect of Wavelength on Bleaching of Photochemically Yellowed Newspaper (from Leary, 1967)

Wavelength of Irradiation, nm

400

420

448

480

508

Observed Effect

Slight greying Bleaching Bleaching Bleaching Slight bleaching

71

-8-

0.4-----,r------..---------

(.!) z ::c u

0.2

0.0

-0.2

~ -0.4 _I

ca

t -0.6

-0. 820

I I I

• UBP , I

' I I I

' I

' 1 AT

Figure 1

WR BP

--o- Exposed 75 Hrs. - -•--Exposed 650 Hrs.

Data from Lee, Bogaa rd and Feller, I COM Paper 81/14/4,

1981

40 60 80

INITIAL REFLECTANCE (R457)

BEFORE EXPOSURE TO 11DAYLIGHT'1

FLUORESCENT LAMPS

100

72

V'> -~ -<]

-9-

UNBLEACHED PULP 4. 4% Lign in, 3. 7% HAS Matter

0.2 --------------

0.0

-0.2

-0.4

-0.6

Figure 2

initial R457 = 31. 0%

o no filter

0

a UF-4 filter > 400 nm A 3-71 filter >460 nm

100 200

HOURS OF EXPOSURE TO DAYLIGHT FLUORESCENT LAMPS

300

73

V') -~ <]

-10-

Fl LTER PAPER WITH LIGNIN ADDED-ON

3% Li nin 5. 2% HAS Matter 0.2-----------.--------,.----

0.0

-0.2

-0.4

-0.6

Figure 3

initial R457 = 33. 8%

Bleaching

o no filter • UF-4 filter> 400 nm A 3-71 filter > 460 nm

100 200

HOURS OF EXPOSURE TO 11DAYLIGHT11

FLUORESCENT LAMPS

74

V) -~ <l

-11-

GROUNDWOOD PULP 29% Lignin, 15% HAS Matter

0.0 t,-c::::.--A----6.---Jr---~A­Little Change

-0.2

-0.4

-0.6

0

Figure 4

initial R457 = 57.3%

o no filter a UF-4 filter> 400 nm A 3-71 filter > 460 nm

100 200

HOURS OF EXPOSURE TO "DAYLIGHT" FLUORESCENT LAMPS

300

75

0.5

0.4

t E

XP

OS

UR

E

UN

DE

R FLU

OR

ES

CE

NT

S

UN

LAM

P

(.!)

(U

LTR

AV

IOLE

T

290-

350 n

m)

0.3

z z LL

.l

0.2

~

a:::

<

( 0

0.1

Vl

Ble

ache

d Pul

p 0.

24%

Lign

in 5

.0 H

AS

Mat

ter

--0.

0 ~

<J

I _,

-0.1

N

(.!)

I

z U

nble

ache

d Pul

ps 4

.4%

Lig

nin

3. 7

% H

AS

Mat

ter

-0.2

:::

c u ~

_J

-0.3

ca

-0.4

t

Fig

ure

5

-0.5

10

0 20

0 30

0 40

0 50

0 60

0

-....

J H

OU

RS

OF

EX

PO

SU

RE

O

'l

+

•

t Y

ello

win

g 0 z .

--u

+

a.:

Ble

achi

ng

<l

"CJ

a, - -·- E

w

a::

3:

a.

a, > -0 a,

a::

a,

(J

C

0 -+- E

Cl)

C

0 i...

t-

-....

J 30

0 35

0 -.

...J

400

DA

TA

OF

LE

AR

Y,

TA

PP

I, 19

67

EM

ISS

ION

OF

C

AR

BO

N A

RC

TR

AN

SM

ITT

AN

CE

O

F

" 38

5 11

F

I LT

ER

450

Wav

elen

gth,

nm

::!!

l.C

I C

_.

""

"I

w

ct>

I

°'

500

0

......

. ":

!"~

•-;~tt

t•lt

, et

--...

...

... ~

-·-

.aa.

.1...

..-...

.....

,.. ..

u_

300

400

WA

VE

LE

NG

TH

, m

mu

Fta

. l

Co

mp

ari1

011 o

f Sp

ectr

al

Sens

itiv

ity

of

''Fad

lnc"

of

Gro

undw

ood

and

Spec

tral

A

bsor

ptio

n C

urve

s o

f L

icni

n C

ompo

1111

d1

Sept

embe

r 13, 1

945

Figu

re 7

. Se

nsiti

vity

of

Lig

nin

to

Dar

keni

ng

(cal

led

"fad

ing"

her

e).

Dat

a of

Nol

an,

Van

den

Akk

er a

nd W

ink,

194

5

-Ill ~l--

--+

----

+--

----

+--

---4

-1

4,i--

--¥-

---+

----

+--

-'-'

0 3r

---f

-l~--

1.-4

----

-1--

---I

-I

- ,c

Ot..

.i..-

-~.i-

.a.

.....

...i..

.1...

1.-'-

i...J

.....:

...1.

.i..;=

-i

250

300

350

400

>.,

nm

Fig.

3.

lJltr

avio

let

abso

rptio

n sp

ectr

um

of s

pruc

e m

illed

-woo

d lig

nin

in

meJ

hyt

cello

solv

e bef

ore(

-)

and

afte

r(--

-)

ex­

posu

re t

o ne

ar u

ltrav

iole

t lig

ht f

or 3

hr.

Figu

re 8

. C

hang

es in

Abs

orpt

ion

Coe

ffic

ient

, E

, of

Lig

nin

in

Ultr

avio

let

Reg

ion.

D

ata

of

Lin

and

Kri

ngst

ad,

1970

I ....

+" I

8 c:::

0·8

0·7

0·6

0·5

0·4

0·3

0·8

0·7

0·6

0·5

0·4

0·3

:::?

=:: ..

~--:

:-.-

···

~~···

·· #,

: .. ·

· .,

,6;,,

-·

1/,::

..--·

····

~···

······

······

·

;7· . ..

:.···

·· ~

f····

·~

,¢"

::;;-

::·

-~:,_

-;::~

-,::

::;,-

;.,,,-

-:.•

·· .

~"'·

···

~ ..

.... -

~-__

..~:::

-··

.,,,..

,,,,..

;,

,_":

-···

. ..-:

.,--:

...

· ,, ./

.,,,

.··

%~"

'.~ .. -·

¼·,,

, .•

· ; .

.;.:·,

,," .. "

...

_ 3a

;/ .. -

-·· ~

r,/··

· .

I.­ , ...

,_.

,_.-

, ...

1:

(

400

500

600

~,m

F

IG. 6

. R

efle

ctan

ce s

pec

tra

of

wes

tern

hem

lock

sap

­w

oo

d p

ulp

s b

efo

re (

a)

and

aft

er (

b)

ther

mal

ag

ein

g:

1)

ori

gin

al

pu

lp.

2)

EO

TA

tr

eate

d,

3)

eth

ano

l•

ben

zen

e ex

trac

ted

, 4)

ac

eto

ne-

wat

er e

xtra

cted

.

0

-O·I

-0-2

I

-0·3

400

... -··· ,,,

500

----

--·-

-- ... -··

· .. •·

... •····

······

····

..

,,,.,,

,.,,, .... --

----

----

600

~, n

m

FIG

. 7.

[DO

.]

curv

es

of w

este

rn

hem

lock

sa

pwoo

d pu

lps

befo

re

and

afte

r th

erm

al

agei

ng:

1)

orig

inal

pu

lps,

2)

E

OT

A

trea

ted,

3)

et

hano

l-be

nzen

e ex

­tr

acte

d,

4)

acet

one-

wat

er

extr

acte

d (1

:1).

Figu

re 9

. A

utho

r's

Figu

re 7

Illu

stra

ting

Cha

nges

in R

efle

ctan

ce

of P

ulps

in

the

Vis

i­bl

e R

egio

n, S

how

ing

Use

fuln

ess

of S

ubtr

act­

ing

the

Cur

ves

of

Ref

lect

ance

Se

en i

n A

utho

r's

Figu

re 6

.

Dat

a of

Pol

cin,

et

. ~-

, 19

69

[DC

A] =

K

/So

-K

/S1

K/S 0

I __,

0, I

,. IIO

IO

-16-

Figure 14

'°o._ ___ .._n ___ _.so.__ ___ n.__ __ ~100,-----Qt

436""'

Figure 13 .,_

100

90

10

Fig. 13. Relative reflectivity at 430 nm of newsprint, irra­diated at 365 nm for 30 minutes, as a function of the dura-tion of the bleaching at 43,; nm. I. Measured immediati:!y 70 after bleaching. 2. Measured 2. weeks later.

Figure 10

120

2

~~:;c~::.:-:::~.;cf~t: 3

9

90

i,00 500 600 nm

Fig. 14. Relative reflectivity as a function of wave-length of !lewsprint showing re-yello\\·ing after irradi:1tion and bleach­mg. 1. Irradiated at 365 nm for 30 minutes. :. As I but bleached at 436 nm for 60 minutes, me.isured immedi.itely after bleaching. 3. As 2 but measured .i weeks later.

O !la JOO 150 min ol 36SM1

Fig. 4. Relative reflectivity al 470 nm o: high yii!d bisul­phire liner· as a function of the duration of the irr:1diation at 365 nm. I. Irradiated only at 365 nm. 2. As t but ble:1ched at 436 nm for 60 minutes. measured immediately after bleaching. ;. As 2 but measured 24 hours later.

.,_ Figure 11

150

70300 400 . 500 . 600 nm

Fig. I 5. Relative reflectivity as a function of wave-length of irradiated wood from spruce. Reference: Unirradiated sample. 1. Irr.adi:md at 365 nm for 8' hours. 2. As I but bleached at 436 nm for 2 hours, measured immediately after bleaching. 3. As 2 but measured 3 weeks later.

Data of Claesson, Olson and Wennerblom, 1968

80

co

.....

Figu

re

12

Dat

a of

Sil

vy

and

LeNe

st,

1970

, 19

76

w

0·10

..... w

800

B 60

0 ..

Cl)

Q.

B 4

00

0::

a. 20

0

0 ·0 5

2000

300

0 Fig.

1

Fig.2

0·6 0

0·50

0-40

0·30

0·20

0·10

5000

.8

6000

1

00

0 a

Auth

ors

' Fi

gur

e 1.

Sp

ectr

a of

Pu

re

(1)

and

Ligni

fie

d (2

) Ce

llul

ose

.

Auth

ors

' Fi

gur

e 2.

Ab

sorpt

ion

of

Pure

Ce

llu

­lo

se H

eate

d at

12

0°

C (c

urve

s l,

2

and

3)

an

d Su

bseq

uent

ly

Bleac

hed

in

11Da

yli

ght

11

(dot

ted

curv

es

4,

5,

6).

c Cl)

0

6000

t 45

00

0.

z· � 3

000

Q.

1500

2000

300

0 40

00

sooo-

6000

A

Fig. 3

800

c 60

0 IV

0

... �

400

2 u

a::

a.. 2

00•-

0

t

2000

300

0 40

00

sooo

60

00 A

Fig.

4

Auth

ors

I Fi

gur

e 3.

Pu

re

Ce 1

1 ul o

se T

herm

ally

Da

rken

ed

(l)

, Bl

each

ed

in "

Dayl

ight

" at

35

°

C (2

)an

d Su

bseq

uent

ly H

eate

d in

Da

rk

(3).

Auth

ors•

Fi

gur

e 4

. Re

sult

of

Pe

riod

ate

Oxid

a­ti

on o

f Ce

llul

ose

(1

) an

d Su

bseq

uent

He

atin

g of

th

e Ox

idiz

ed

Mate

ria

l (2

).

I .....

'.I

I