EFFECT OF pH AND BUFFERING CAPACITY OF

WOOD ON THE GELATION TIME OF

UREA-FORMALDEHYDE RESIN1
William E. Johns
Assistant Professor

and

Khan A. Niazi
Graduate Research Assistant
Wood Technology Section, Materials Science and Engineering Department,
Washington State University, Pullman, WA 99164
(Received 8 February 1980)

ABSTRACT

The pH and buffering capacity for water extract of the heartwood and sapwood of ten hardwoods
and nine softwoods were determined. pH values ranged from 4.00 to 5.86 for hardwood and 4.02 to
5.82 for softwood. The gelation time for a urea-formaldehyde resin as affected by wood flour for each
wood was determined. Analysis shows that a strong correlation between gel time of wood flour and
either pH or acid buffering capacity of water extract exists.
Keywords: pH, buffering capacity, hardwoods, softwoods, urea-formaldehyde resin, gelation time.

INTRODUCTION A N D REVIEW OF LITER.ATURE

Knowledge of the pH and buffering capacity of wood is fundamental to the

efficient use of wood in many gluing and coating processes where the glue or
coating is pH-sensitive. For example, a urea-formaldehyde (UF) resin is acid-
catalyzed and cannot be used to an optimum state-of-cure in a low acid environ-
ment.
Various methods for determining pH of wood have been proposed in an attempt
to correspond to the conditions to which the wood will be exposed in use. For
air-dry wood (5 to 20% MC), Campbell and Bryant (1941) proposed that pH of
wood ". . . is numerically equal to that of a solution containing free hydrogen
ions, which when added to it under the conditions stated, undergoes neither a
net loss nor gain of hydrogen ions." A series of unbuffered solutions of acids and
bases of different pH are prepared and samples of wood are immersed. The
solution whose pH does not change by immersion of wood as measured by a
glass electrode is taken to represent the true pH of the wood. The drawback of
this method is that a large number of solutions have to be prepared approximating
pH value of wood.

I We wish to thank Joe Dede at Georgia-Pacific Corporation, Albany, Oregon, for providing the

urea-formaldehyde resin and staffs of the University of Minnesota College of Forestry, Virginia
Polytechnic Institute, and Weyerhaeuser Company for assistance in obtaining the wood specimens
used for this experiment.
Wood and Fiber. 12(4). 1980. pp. 255-263
@ 1980 by the Society of Wood Science and Technology
256 WOOD AND FIBER, DECEMBER 1980, V. 12(4)

Ingruber (1958) used the method of Campbell and Bryant but with only a few
different unbuffered solutions of varied pH. By plotting the values on a graph of
pH change vs. original pH, he found the point at which no change in pH occurred.
Stamm (1961) compared the above two methods with the method of measuring
pH on the surface of moistened wood itself using flat-headed glass electrodes.
However, he reported erratic pH readings with the contact method unless great
care was taken. Stamm concluded that the three methods gave comparable results
and that simple extraction using comminuted wood and water in the ratio of 1:6
gave satisfactory pH values if the extract solution is sufficiently buffered, as with
most woods.
Gray (1958) used a 1:3 ratio of wood sawdust to water and measured pH by a
spear-type glass electrode and wick-type reference electrode. This ratio of wood-
to-water was used to represent the most drastic conditions to which metal fittings
are likely to be exposed. Gray attempted to correlate pH of wood to the corrosion
of iron nails and concluded that the higher acidity of some woods accelerates the
corrosion of iron under moist conditions.
In general, wood species range in pH from 3.0 to 5.5 (Stamm 1964). The knowl-
edge of pH and buffering capacity of wood is an important consideration to a
better understanding of the wood gluing processes (Maloney 1977; Foster 1967).
Extreme values of wood pH have been reported to be troublesome for achieving
good adhesive bonds. For example, the rate of cure of U F resin is retarded by
Indian alkaline timber Quirpo and the bond is adversely affected (Narayanamurti
1957). Gluing of English oak gives low strength with resorcinol or resorcinoll
phenol glue but excellent strength with U F resin. However, when the surface of
wood is washed with dilute solutions of sodium acetate, excellent adhesion with
both resorcinol and resorcinol/phenol adhesives is obtained. The acidity of oak
(pH 3.7) is believed to seriously prolong the gelation and curing of relatively
unbuffered resorcinol glue at room temperature (Rayner 1965). Thus, the wood
pH may change the pH of adhesive at the interface and modify the curing of glue.
Although U F is the most commonly used wood glue for the particleboard in-
dustry, no published work is available in correlating pH and buffering capacity
of aqueous wood extract to the gelation time of this acid curing adhesive. It is
the purpose of the present study to find out if such a correlation can be estab-
lished.

PROCEDURE
To achieve the objective, the following determinations were made:
1. pH of aqueous wood extract.
2. Buffering capacity of aqueous wood extract for a pH range of 3-7.
3. Gelation time of unbuffered urea-formaldehyde resins as affected by the sev-
eral wood species.
Air-dried heart- or sapwood of a variety of species was isolated (Table I),
Wiley-milled through a coarse screen (%-inch mesh), and stored in polyethylene
bags until used. No classification of the wood meal was made. The aqueous wood
extract was prepared by refluxing 25 g of dry wood material in 250 g of distilled
water for 20 min. After refluxing, the mixtures were filtered through Whatman
 Johns and Niazi-pH AND GELATION TIME OF RESIN 257

TABLE1. p H , acid buffering ccrpcrcity, and gelation time of hardwood and softwood species.

Acid3-
SWIHW1 buffering Gel time3
Spectes ,MIX pHZ capacity (minutes)

Hardwoods
I . White oak (Quercus alba L.) SW 4.56 0.182 12.13
HW 4.00 0.435 5.92
2 . Aspen (Populus spp.) MIX* 4.30 0.204 11.70
Wet Wood 4.35 0.154 11.92
3. Red alder (Alnus rubra Bong.) HW 4.66 0.081 11.38
4. Yellow poplar (Liriodendron tulipifera L . ) SW 4.75 0.103 15.50
HW 4.79 0.053 12.50
5. Hickory (Carya spp.) SW 4.97 0.105 15.00
HW 5.63 0.049 25.00
6. Sweet gum (Liyuidcrmhar .styrucij?ua L . ) MIX 5.01 0.082 18.25
7. Blackgum ( N y s s a sylvatica Marsh.) SW 5.13 0.079 21.50
HW 5.86 0.019 34.75
8. Red maple ( A c e r rubrum L.) SW 5.40 0.037 20.50
HW 5.45 0.024 18.50
9. Red oak (Quercus rubra L.) SW 5.04 0.100 19.18
HW 5.66 0.009 19.00
10. White birch (Betula papyrifera Marsh.) SW 5.54 0.014 21.75
HW 5.51 0.079 22.50
Softwoods
11. Douglas-fir (Pseudotsugu menziesii Mirb.) Franco SW 4.76 0.065 14.00
HW 4.02 0.277 7.75
12. White pine (Pinus strobus L.) SW 4.15 0.227 9.25
HW 4.90 0.039 12.50
13. Red pine (Pinus resinosa Ait.) SW 4.50 0.125 10.25
HW 4.46 0.155 9.50
14. Southern pine (Pinus spp.) MIX 4.66 0.065 12.08
15. Jack pine (Pinus hnnksiana Lamb) SW 4.70 0.074 14.26
HW 4.94 0.050 12.50
16. White spruce (Picea gluuca Moench) Voss MIX 4.96 0.022 15.35
17. White fir (Abies concolor Gord. and Glend.) SW 5.25 0.022 16.38
18. Tamarack (Larix laricina (DuRoi) K. Koch) SW 5.17 0.027 18.33
HW 5.18 0.044 18.75
19. Northern white cedar (Thuja occidentu1i.s L . ) SW 5.82 0.009 22.75
HW 4.97 0.051 12.13
' SW = Sapwood; HW = Heartwood; MIX = Mixture of sapwood and heartwood (used when precise separation of heartwood and
sapwood was not possible).
' Average of four values, based on two extract replications.
Average of two values, based on two extract replications.
* Isolated as an obviously discolored zone of wood near pith.

# 1 filter paper with an aspirator vacuum. The mixtures were cooled to room
temperature before titrating.
All pH and buffering capacities measurements were made with a Fischer Ac-
cumat pH meter, Model 310 after the technique of Lambuth (1967). Prior to each
 WOOD AND FIBER, DECEMBER 1980, V. 12(4)

SOUTHERN PINElHEART BSAP WOOD1

A R E 0 PINE
0 WHITE PINE
0 J A C K PIME

- SAP w o o 0

1 0 9 8 1 6 5 4 3 2 1 0 1 2 3 4 5 6 T 8 3 l O
OoZ5'
MILLILITRES OF Oo2' H2s04

FIG.1. The pH, acid, and base buffering potential for selected pines.

titration, the pH meter was calibrated with standardized buffer solution to a pH

of either 4 or 7, depending on the buffering capacity measurement to be made.
After calibration, exactly 50 ml of wood extract solution was pipetted into a 150-
ml beaker, the pH of the extract solution noted, and then titrated to a pH of
either 3 or 7 with nominal 0.025 N NaOH or H,SO, solutions. The pH of the
constantly stirred solutions was noted after each incremental ml of acid or base.
The acid buffering capacity or acid equivalent is defined by Lambuth (1967) as
the number of ml of NaOH solution required to raise the starting pH of the wood
extract to a pH of 7.0 times the normality of the base solution. Similarly, alkaline
buffering capacity determinations utilized the H,SO, solution with the comparable
calculation. Total buffering capacity is merely the sum of these two values.
An unbuffered urea-formaldehyde resin was obtained from Georgia-Pacific Cor-
poration, Albany, Oregon. Experimentally it was determined that a gel time of
25.7 min resulted when 5 drops of 0.5% NH,Cl solution were added to 4 ml of
the resin in a 13- x 100-mm test tube, and the test tube immersed in boiling water
and stirred with a stainless steel wire. Gelation time is defined as that point when
the adhesive or adhesive-wood mixture suddenly loses its fluidity; it becomes a
solid. Gelation time as affected by various wood species was found by mixing
0.7 grams of wood meal (O.D. wood basis), 5 drops of 0.5% NH,Cl, and 4 ml of
the resin in the test tube and heating in boiling water with constant stirring.
STATISTICAL CONSIDERATIONS

All initial pH readings are the average of four measurements, while buffering
capacity and gelation times are the means of two determinations. A total of thirty-
three samples of sapwood and heartwood (ten hardwood and nine softwood) were
obtained from the species noted in Table 1. In cases where heartwood and sap-
wood were not readily separated, representation samples from the total wood
cross-section were used.
Linear regression and correlation analysis were completed on the following
combinations for hardwoods, softwoods and combined hardwoods/softwoods.
 Johns and Niazi-pH AND GELATION TIME OF RESIN

D
--EL.
-- D.
\
n
----
0
m
0
WUGLAS FIR
TAMARACK
WHITE FIR

HEART WOO0

FIG.2. The pH, acid, and base buffering potential for Douglas-fir, western larch (tamarack), and
white fir.

gelation time vs. pH

gelation time vs. acid equivalent
gelation time vs. base equivalent
gelation time vs. total buffering capacity
In addition, nonlinear analysis was completed for gelation time vs. acid equivalent
for hardwoods, softwoods, and combined data.
Significance of correlation coefficients, regression analysis, and differences be-
tween mean gelation times was tested (Steele and 'Torrie 1960).
RESULTS A N D DISCUSSION

The data for pH and buffering capacity are graphed in Figs. 1 through 6. In
this study, initial pH values ranged from 4.00 to 5.86 for hardwoods and from

pH

?\
\
I
I -
0
WHITE SPRUCE (HEART I) SAP WOOD)
N WHITE CEDPR
SAP W O O

1 0 9 8 7 6 5 4 3 2 1 0 1 2 3 4 9 6 7 8 9 D

0025N Naon MlLLlLlTRES OF O OZ5 " %?so*

FIG. 3. The pH, acid, and base buffering potential for white spruce and northern white cedar.
 WOOD AND FIBER, DECEMBER 1980, V. 12(4)

0 02% NOOY
MILLIL1TRES OF 02!'N '2 "4

FIG.4. The pH, acid, and base buffering potential for maple, white birch, and red alder.

4.02 to 5.82 for softwoods (Table 1). The difference between sapwood and heart-
wood pH values was not consistent, as was also reported by Gray (1958). In
hardwoods, two species (white oak and white birch) gave higher values of pH for
sapwood, while five species (yellow poplar, hickory, black gum, southern pine,
and jack pine) gave higher pH values for heartwood. It was not possible to com-
pare these pH values with those in the literature because of lack of identical
combinations except for one comparison in the case of Douglas-fir. Our method
gave slightly higher pH values for Douglas-fir than those reported by Crawford
(1968) (4.76 vs. 4.61 for sapwood and 4.02 vs. 3.94 for heartwood). It was not
practical within the scope of this experiment to seek possible chemical reasons
for the differences between heartwood and sapwood of the same species.
Figures 1 through 6 show that acid equivalent, base equivalent, and total buff-
ering capacity of aqueous extracts of various wood species and those of sapwood
and heartwood within the species vary considerably. No similar information is
available to compare these with the results of others.

O $WELT OUY
m BLACK OYM
0 VELWW POPLAR

0 02BN Noon MILLILITREs 0026N nzsC,,

FIG.5. The pH, acid, and base buffering potential for. selected hardwoods.
 Johns and Niazi-pH AND GELATION TIME OF RESIN

FIG.6. The pH, acid, and base buffering potential for selected hardwoods.

Figures 7 and 8 give the gelation time of UF resin as a function of pH and acid
equivalent, respectively, for hardwoods, softwoods, and combined data. A highly
significant linear correlation was found for each of these cases (Table 2).
Statistically, there is no difference between the slopes of the lines of gelation
time vs. pH for hardwoods and softwoods at the 5% level (Fig. 7). In other words,
increase in pH and acid equivalent of hardwoods and softwoods have the same
corresponding effect on gelation time.
Curiously, the data in Figs. 7 and 8 show that the hardwoods gelled more
slowly than the softwoods at the same pH and acid-equivalent values. The exact

35.01 C- -- -0 SOFTWOODS y = -26.637 + 8.3561 , r = 0.929

I -----a HARDWOODS

COMBINED
y = -37.721

y = -35.513
+10.9921, r ~ 0 . 8 9 3

+ 10.394r, r ~ 0 . 8 9 4
30.01 -

FIG.7. The relationship between the pH and wood extracts and the gel time of a urea-formal-
dehyde resin as affected by wood meal of the same species.
262 WOOD AND FIBER, DECEMBER 1980, V. 12(4)

II --+ HARDWOODS

COMBINED
lny = - 3 . 2 0 6 ~ + 3 . 1 3 0 , r = -0.819

lny= -3.0671 + 2.985, I= -0.756

ACID EQUIVALENT

FIG.8. The relationship between the acid equivalent of wood extracts and the gel time of a urea-
formaldehyde resin as affected by wood meal of the same species.

reason for this is not known but is probably related to the experimental technique
used for wood extraction and hence, pH and acid-equivalent values as compared
to the technique used for the gelation time.
The pH and the acid-equivalent values of wood extract were determined in a
closed system with the probability of no volatile acid(s) loss, while gelation time
was measured in an open system, and any volatile acids were capable of escape.
Modification of the gelation technique to a closed system is not direct in that
constant stirring is required. Thus, wood-extract pH and the buffering capacity
values were not necessarily representative of the values for the gelation time
process.

TABLE2. Summary of statistical rests for correlation of gelatron time vs. p H , acid equivalent, bose
equivalent, and total buffering capacity of various wood species.

Gel time Gel time vs. Gel time vs. Gel time vs. total
Species vs. p H acid equivalent base equivalent buffering capacity

Combined hardwoods ** ** * ns
and softwoods r = 0.894 r = 0.756 r := 0.427 r = -0.191

Hardwoods ** ** ns ns
r = 0.893 r = 0.819 r := 0.421 r = -0.248
Softwoods x* ** ns **
r = 0.929 r = 0.914 r = 0.193 r = -0.738
Remarks: ns = statistically not significant at 5% level
= statistically significant at 5% level but not significant at I% level
I* = statistically significant at 1% level.
 Johns and Niazi-pH AND GELATION TIME OF RESIN 263

A higher rate of acid liberation from hardwoods (oak and birch) as compared
to softwoods (Douglas-fir and Parana pine) has been reported by Packman (1960);
higher rate of acid liberation and thus loss in the open system is consistent with
our results-that is, hardwood species when compared to softwood species have
a longer gelation time because of the loss of acid associated with the procedure
used for measuring the gelation time.
Table 2 shows that base equivalent and total buffering capacity are not signif-
icantly related to the gelation time. The one exception seems to be the relationship
of total buffering capacity vs. gel time for softwoods. This may very well be
related to the very strong correlation of pH and acid equivalents for hardwoods.
The best fit equations are provided in Figs. 7 and 8 for future reference.
CONCLUSIONS

1. A pH range of 4.00 to 5.86 for hardwoods and 4.02 to 5.82 for softwoods was
found.
2. Gelation time of a urea-formaldehyde resin was directly correlated to the pH
and inversely correlated with acid-buffering capacity for both hardwood and
softwood aqueous extracts.
3 . An increase in either pH or acid equivalent has the same effect on the gelation
times for U F resins when in contact with either hardwood or softwood
aqueous extracts.
REFERENCES

CAMPBELL, W. G., A N D S. A. BRYANT.1941. Determination of pH in wood. Nature 147:357.

CRAWFORD, R. J. 1968. Production experiences in large plant using several species for furnish. A
description of the laboratory method used in measuring pH and buffer of wood. Proceedings of
the Second Annual International Particleboard Symposium, pp. 129-138.
FOSTER,W I L L I A M G. 1967. Species variation. Proceedings of the First Annual Particleboard Sym-
posium, pp. 13-22.
GRAY,V. R. 1958. The acidity of wood. J. Inst. Wood Sci. 1:58-64.
INGRUBER, 0. V. 1958. The behavior of wood and wood constituents as acid-buffering systems. Pulp
Paper Mag. Can. Nov 59: 135-141.
LAMBUTH, A. L. 1967. Procedure for determining the pH and buffering capacity of wood. Lab Test
Method No. 142, Monsanto Company Seattle Research.
MALONEY, T. M. 1977. Modern particleboard and dry-process fiberboard manufacturing. Miller
Freeman Publications, San Francisco.
MOOR,W. E., A N D D. B. JOHNSON.1967. Procedures for the chemical analysis of wood and wood
products, as used at the U.S. Forest Products Laboratory. Forest Products Laboratory, Forest
Service, U.S. Department of Agriculture, Madison, Wisconsin.
NARAYANAMURTI, D. 1957. Die Bedeutung der Holzextraktstoffe. Holz Roh-Werkst. 15:370.
PACKMAN D., F . 1960. The acidity of wood. Holzforschung 14(6):178-183.
RAYNER C., A. A. 1965. Synthetic organic adhesives, pp. 186-352. In: Houwink, R. and G. Salomon
eds. Adhesion and adhesives, Voi. 1. Elsevier Publishing Company, New York.
STAMM, A. J. 1961. A comparison of three methods for determining the pH of wood and paper. For.
Prod. J. 11(7):310-312.
STAMM, A. J. 1964. Measurement of pH. Para. 10.1. Selective adsorption from solutions. Chapter
10. Wood and cellulose science. The Ronald Press Company, NY.
STEELE,R . G. D., A N D J. . H. T O R R I E 1960.
. Principles and procedures of statistics. McGraw Hill
Book Company, NY.