Note: Descriptions are shown in the official language in which they were submitted.
7~
METHOD OF FINISHING PAPER
UTILIZING SUBSTRATA THERMAL MOLDING
TECHNICAL FIELD
This invention relates generally to the manufacture of paper and in
particular to a novel method of finishing printing paper in a manner
which improves its properties.
BAC KG ROU N D A RT
High quality printing paper must have a number of physical
properties. Two of the most important are a flat and smooth surface
to facilitate printing in a press and gloss to produce a more attractive
surface, particularly after printing. These properties can be obtained
by a variety of techniques, such as coating the paper with pigments
and binder and finishing it in one or more pressing operations.
One of the most common finishing operations employed in the
manufacture of printing paper is supercalendering, in which paper is
passed through a series of nips formed by steel rolls pressed against
cotton filled rolls at very high pressures, typically at nip loads
between 175 KN/M and 437.5 KN/M (1000 and 2500 pounds per lineal
inch). This typically results in nip pressures of 13,780 KN/M2 to
27,560 KN/M2 (2000 to 4000 p.s.i.).
Traditional supercalender stacks are not externally heated, but
heat is generated when the cotton filled rolls subjected to the
extremely high pressures in the nip flex intermittentiy with each
revolution. The nip temperatures in such supercalenders typically
reach levels of about 71C (160F). Another important element in
producing good results is having a high moisture content in the paper
as it passes through the supercalender. Typically, the moisture
content will be 7% to 9O, or higher, of the bone dry fiber weight.
Flatness, smoothness and high gloss are obtained in supercalenders
because of extreme compression and densification of the sheet. The
densification undesirably results in reduced opacity and a blackening
effect in overly moist portions.
Supercalenders commonly consist of a large number of rolls (9 to
14), alternating steel and resilient, in order to obtain the desired
smoothness and gloss. In order to obtain smoothness on both sides it
is necessary to run an even number of rolls and with two resilient
rolls (so called "cushion rolls") running together midway in the stack
to perform the necessary reversing of the side toward the steel rolls.
5 This action is only partly successful at providing two smooth sides
since the first side finished towards the steel is later deformed by the
exposure to the resilient rolls.
Because of this shortcoming and the inherent mechanical problem
associated with the "cushion roll" nip, many supercalenders operate
10 today with an uneven number of nips and no "cushion roll" nip, which
results in only one side being finished against the steel, and while
gloss values may be manipulated to be close on the two sides,
inevitably one side is noticeably rougher than the other.
Another form of finishing is machine calendering wherein the
15 paper web is passed between two normally unheated steel rolls pressed
together at high pressures. This process produces smoothness, but
little gloss because of the absence of shear in the nip.
Another common finishing operation is gloss calendering, which
uses heated finishing rolls to produce high gloss finishes on coated
20 paper or board without the high pressure of supercalendering. The
nip pressures for commercial machines are typically between about 87.5
to 175 KN/M (500 to 1000 pounds per lineal inch) of nip loading. This
typically results in nip pressures of 6,890 KN/M2 to 13,780 KN/M2
(1000 to 2000 p.s.i.). The lower pressure causes less densification of
25 the paper, and therefore, better opacity, while the higher temperature
softens the coating and permits better gloss enchancement. However,
the finishing effect is limited to the coating and the uppermost surface
of the web. Thus, the surface of the sheet is not as smooth and flat
as that produced in supercalendering and has generally been applied
30 to coated board rather than high quality papers. As a result, gloss
calendered sheets do not print as satisfactorily in a printing press as
do supercalendered sheets.
In recent years, many modifications have been made to gloss
calendering, machine calendering and supercalendering operations.
-
~ w~
--3--
Some supercalenders have been heated, primarily to improve the
uniformity and control of the temperature. Typically, heated
supercalenders reach nip temperatures of about 82C (180F).
Temperatures of some machine calenders or supercalenders have been
5 further increased in an attempt to allow a decrease in pressure to
produce the same results. In spite of this modification of
supercalendering in the direction of gloss calendering, the fundamental
effects of the two processes have remained distinct. Supercalendering
uniformly compacts the entire sheet to a high degree, thus flattening
10 the surface fibers and all others, as well as producing gloss on the
surface. In contrast, gloss calendering molds, flattens, and glosses
the surface of the coating and, in the case of uncoated paper the top
surface of the fibrous substrate, but compacts the remainder of the
sheet much less than supercalendering.
Examples of gloss calendering are disclosed in (J.S. patents Nos.
3,124,504; 3,124,480; 3,124,481; 3,190,212; and 3,254,593. These
patents collectively describe apparatus capable of nip temperatures
from below the boiling point of water to as high as 232C (450F) and
nip pressures from 1,722 to 17,220 KN/M2 (250 to 2500 p.s.i.). No.
20 3,124,480 describes finishing steps designed to heat the coating on
paper to a temperature which temporarily plasticizes at least the
surface of the coating in contact with the hot drum. A form of
supercalendering in which the rolls are heated to relatively high
temperatures is disclosed in U.S. patent Nos. 3,442,685 and 3,451,331.
25 These patents disclose a method and apparatus capable of producing
high gloss on coated paper by heating at least one roll of a
supercalender stack to a temperature between 82C and 163C (180F
and 325F) to plasticize the coating.
The one parameter which has been found to be the most critical
30 in gloss calendering and supercalendering has been the moisture
content of the paper. High moisture improves the smoothing and
glossing effects of both the coating and the paper substrate. Many
developments in supercalendering and gloss calendering involve
7~
techniques for increasing the moisture in the web or at least in some
portions of it before finishing.
Unfortunately, moisture is an undesirable control parameter.
Small variations in moisture cause large variations in the finished
properties of the paper. Also, it is undesirable to have more than
about 3.5% to about 4.5o moisture in the finished sheet to avoid
uneven reel building and sheet curl from later drying. This amount of
moisture is a stable amount, and the sheet will not dry significantly
below this level under ambient conditions. To have a finished product
with the desired low moisture content and still have the desired high
moisture content (e.g. 76 to 96) to facilitate calendering, many heated
calendering operations have increased the drum temperature to dry the
moister webs.
Nonuniformity of moisture in the sheet can be even a bigger
problem than too much moisture. By nonuniformity, it is meant that
the moisture content at one place on the sheet is higher or lower than
at other locations across the width of the sheet. The nonuniformity
can also exist in the machine direction and the thickness of the sheet.
Nonuniformity is most severe when calendering takes place immediately
after coating, which is to say when the calender is in line with the
coater. If coating is done in a separate operation from calendering,
the moisture content of the coated paper has time to equalize
throughout the web before calendering.
The above cited patent No. 3,124,504 is primarily concerned with
very moist webs (up to 356 or 50~o moisture) and includes the concept
of drying the web while finishing it. Very high temperatures are
employed for drying, but temperatures above the boiling point of water
are said to be needed only if the web is wetter than 5% to 8% of the
bone dry weight. The web moisture content is also noted as being an
important element in the process disclosed in above cited patent Nos.
3,442,685 and 3,451,331. The patents teach that it is best for the
paper to have about 7% moisture content, and moisture can be added
before the supercalender to improve the finishing effects. The
addition of moisture before finishing is also described in above cited
patent No. 3,124,481 to manufacture glazed uncoated paper. U.S.
Patent No. 2,214,641 also moistens the surface of the web before
finishing. In U.S. Patent No. 4,012,543, gloss calendering is
undertaken immediately after coating before too much of the moisture is
lost from the coating. In this disclosure, finishing is carried out at a
web moisture content of 9-O to 10% of the bone dry weight. In
contrast, U.S. Patent No. 3,268,354, takes special steps to dry the
surface of the coating, but to maintain a wet interface between the
coating and the fibrous web before gloss calendering. The web in this
disclosure has a moisture content of at least 156 at the interface.
DISCLOSURE OF THE INVENTION
The present invention is a new process which permits the
rnanufacture of paper with supercalender smoothness and gloss without
the above noted disadvantages of supercalendering.
The invention is a process for producing gloss and smoothness on
the surface of a paper web, comprising the steps of:
A. providing a finishing apparatus comprising a smooth metal
finishing drum and a resilient backing roll pressed against the
drum at a force of up to 700 KN/M (4000 pounds per lineal inch)
to form a nip with pressure against the paper web of at least
13,780 KN/M2 (2000 p.s.i.);
B. advancing a web of papermaking fibers having a moisture
content in the fibers of from 3o6 to 7% of the bone dry weight of
the fibers through the nip at a speed which results in the web
dwelling in the nip from 0.3 milliseconds to 12 milliseconds; and
C. simultaneously with step B, heating the drum to a surface
temperature having a value no less than 20C below the value
determined by the following formula:
Ts = [Ti x .357t ~ 479 - 234.2e~ 131m]/[.357t ~ 479 -1]
where:
Ts = surface temperature of the heated drum, in C;
Ti = the initial temperature of the web just prior to
entering the nip, in C;
t = dwell time of the web in the nip, in milliseconds;
e = the base of the natural logarithm; and
m = moisture content of the fibers in the web in weight
percent of the bone dry fiber weight.
Much of the prior art discloses broad operating conditions in
which some of the conditions of the present invention fall, but fail to
teach the special requirements for low moisture paper and are far too
broad in their disclosures for one to appreciate the present critical
operating range. They all either calender at a temperature and/or
pressure below the present invention, calender the web too wet, or
teach a very broad temperature range which might accidentally include
the present range.
The invention is believed to owe its success to one phenomenon
believed to be unappreciated before this invention and to another
phenomenon just beginning to be appreciated. With respect to the
first, it has been discovered that an unexpected increment of gloss
and smoothness can be obtained in a critical temperature rangs and
that increment is much greater at nip pressures above 13,780 KN/M2
(2000 psi). With respect to the second, cellulosic fibers, such as
papermaking fibers, appear to exhibit thermoplastic properties and in
particular appear to have a glass transition temperature ("Tg") above
which the fibers become much more flexible and moldable when
subjected to pressing forces. The Tg of cellulose in paper is greatly
dependent upon the moisture content of the paper and is very low for
papers as moist as those traditionally supercalendered. However, this
very property which facilitates supercalendering also results in the
undesirable ultra sensitivity to moisture variations and the
undersirable ultra densification through the entire thickness of the
web .
Although some of the prior art relating to gloss calendering
recognized the effects of temperature on moldability of the coating and
the surface fibers of uncoated paper, none recognized the existence of
a critical strata beneath the surface of the fibers which must be
molded flat to obtain the flatness and smoothness of supercalendering.
~2~ 7
The invention, which can be described as substrata thermal
molding, is based upon molding the critical substrata of the web into a
flat strata permitting the surface of the fibrous web and any coating
to be flattened, smoothed and glossed to the degree obtainable by
5 supercalendering. This strata is the foundation for the surface, and
molding below this level is not critical to obtaining supercalender
flatness. Thus the molding of the entire thickness of the sheet as in
supercalendering is unnecessary, provides little advantage, and results
in the previously noted disadvantages.
The present invention does not require a web as moist as those
generally subjected to supercalendering and gloss calendering. The
present invention performs satisfactorily on a web having a moisture
content less than 7% of the bone dry weight of the fibers and even
less than 6o or 5%. Surprisingly, the invention works satisfactorily at
15 even lower moisture contents, even as low as 3%. Consequently,
finished products can be easily produced at desirable moisture levels
without having to dry them in the finishing process. In addition, the
ability to finish the web at lower moisture contents permits drying
down the web immediately before finishing to a low level where
20 moisture content is substantially uniform throughout the web,
preferably with no variation greater than 0.5% from the average.
Thus, the invention is particularly valuable where coating and
finishing are done continuousiy in line with each other. It is even
more valuable when coating and finishing are done continuously in line
25 with the papermaking machine.
The principal shortcoming of the prior art hot calendering of
coated paper was that it only molded the coating with little effect on
the fibrous substrate. Consequently, while high gloss could be
obtained, the very flat smooth surface of supercalendering was not
30 obtainable. With uncoated paper, the prior art molded only the
surface fibers to coalesce or seal the surface of the sheet. The effect
needed to reach the critical substrata, which is believed necessary to
flatten the web, was not appreciated. Adding confusion to these
teachings was a failure to understand the role of moisture and
--8--
temperature in molding the shee-t. For example, much of the prior art
teaches that temperatures below the scope of the present invention will
suffice at low moisture, but higher temperatures are needed at higher
moistures to dry the sheet.
In a preferred embodiment of the present invention, the finishing
apparatus includes a second resilient backing roll pressed against the
drum preferably within the same pressure range as the first to form a
second nip. The web is advanced through the second nip after the
first nip within a short period of time, less than 4 seconds, to provide
a great advantage, uniquely valuable to this invention and explained
as follows. The key to the invention is to heat a critical substrata of
the web to its Tg. Obviously, this requires a drum surface
temperature hotter than the Tg. At the same time, the Tg increases
with reduction in moisture. Thus, conflicting goals exist in selecting
the drum temperature. If the temperature is too low, the heating time
required, which is limited to dwell time in the nip, will be too long
and cause too much loss in web moisture, as well as a tendency to
raise the temperature of the entire web to the same temperature. If
the temperature is too high, the web must be specl through the nip too
fast to provide the dwell time needed as well as perhaps being beyond
commercially feasible machine speeds.
As set forth in the above description of the invention, there is a
drum temperature range wherein the invention works satisfactorily.
However, the use of two nips on one drum will permit the drum
temperature to be lower and the invention to work more satisfactorily.
The web is heated quickly in the first nip to a relatively high
temperature on its surface which is in contact with the drum, but the
temperature on the opposite side will increase little, if any.
Immediately upon leaving the first nip, the temperature of the web
through its thickness tends to equalize, while of course losing some
heat to the air from both surfaces. As a result, the entire web, and
most importantly the critical substrata, has a temperature raised above
its previous temperature, but below its Tg, when it enters the second
nip. In the second nip the same type of temperature gradient that
.
-9 -
existed in the first nip is established, but with the interior
temperature of the web higher than before. Thus, the critical portion
of the web can be brough-t to the critical temperature using a lower
drum temperature or faster process speed than needed with only a
single nip. Of course, the additional pressing time provided by two
nips will result in surface improvements also.
In the preferred form of the invention, the web will be passed
through the nip or nips without contacting the heated drum except in
the nips for the reasons stated above. However, there may be cases
where it is desirable and not too disadvantageous to have some
additional drum contact. In those cases, it will be preferable to limit
the contact to less than 20% of the drum circumference.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 illustrates schematically an apparatus suitable for
practicing the present invention;
Fig. 2 is a graph illustrating the gloss and smoothness values for
the uncoated paper finished at various temperatures in Example I;
Fig. 3 is a graph illustrating the gloss and smoothness values for
the coated paper finished at various temperatures in Example 2;
Fig. 4 is a graph illustrating the gloss and smoothness values for
the coated paper finished at various temperatures in Example 3;
Fig. 5 is a graph showing the dynamic Tg of cellulose fibers for
va riou s moi stu re contents;
Fig. 6 illustrates schematically the temperature gradient into the
thickness of the paper in a nip of the apparatus illustrated in Fig. 1;
Fig. 7 is a graph showing the temperature gradient into the
thickness of the web for various dwell times of the web in the nip;
Fig. 8 is a graph showing the drum surface temperature required
for the invention for various moisture contents and various dwell
times; and
Fig. 9 is a graph illustrating the gloss values for the coated
paper finished at various temperatures and pressures in Example ~1.
BEST MODE FOR CARRYING OUT THE INVENTION
The following definitions are provided to better understand these
terms in this specification and claims.
-10-
Parker Print-Surf - a quantitative measurement commonly used in
the papermaking field for the printing roughness and porosity of paper
made by sensing the leakage of air at low pressure between the
surface of the sample and the measuring sensing head. The lower the
5 value, the smoother the paper. Parker Print-Surf can be measured
with several different pressures of the dam against the paper being
measured. In the present specification and claims, all were measured
with a pressure of 10 Kg/cm2. Supercalendered coated woodfree paper
will typically have a Parker Print-Surf of less than 1.4 and less than
10 1.0 for very high quality. Gloss calendered coated woodfree paper
will typically have a Parker Print-Surf of between 1.2 and 2Ø
756 Hunter Gloss - a well-recognized quantitative measurement of
the amount of light specularly reflected at an angle 75 from a line at
a right angle to the plane of the paper. Glossy grades of coated
papers typically have a gloss of from 50 to 90. Above 70 is considered
as very high gloss.
The present invention can be carried out on an apparatus like
that illustrated in Fig. 1. A paper web 1 is advanced through the
first nip formed by smooth surface finishing drum 2 and resilient
20 backing roll 3, around guide rolls 4, and through a second nip formed
by drum 2 and a second resilient backing roll 5 pressed against drum
2. Thereafter, if desired for finishing the other side of the web, the
web 1 is advanced to a second smooth surface finishing drum with a
pair of nips formed by resilient backing rolls similar to the first unit
25 (not illustrated for simplicity). The finished web is then wound onto
reel 6. Variations in the process can be carried out by omitting or
bypassing the second nip on each drum and/or finishing on one side
only, in which case the second drum is bypassed or omitted.
The web 1 supplied to the finishing apparatus can come directly
30 from a papermaking machine 7 and/or coater 8, if the paper is to be
coated. In the alternative, the web 1 can be supplied from a roll of
previously manufactured paper which may or may not have already
been coated. The papermaking machine and coater are illustrated only
77
as blocks since they can be provided by any conventionai apparatus
well known in the art.
The finishing apparatus employed in the invention can be
provided by any of the many disclosed in the previously described
prior art relating to gloss calendering if they are designed or can be
adapted to operate at the temperature, pressure and speed conditions
of the invention. Accordingly, little description of the apparatus will
be given herein except to emphasize the importance of choosing a
finishing drum which can be heated to the temperatures required by
the invention and has a smooth metal surface and choosing a resilient
backing roll which is yieldable but will have sufficient hardness at
operating temperatures to provide a nip force between 35 and 700
KN/M (200 and 4000 pounds per lineal inch) of nip, which could
require pressures as high as 60,000 KN/M2 (8,700 p.s.i.) at the extreme
end of the range. The actual pressure to which the paper web is
subjected in the nip will depend upon the force applied and the width
of the nip. Resilient backing rolls flatten somewhat at the nip and will
preferably have a nip width of from 1.27 to 2.54 cm (0.5 inch to 1.00
inch) for the present invention. Nip widths shorter than 1.27 cm and
longer than 2.54 cm could be usable with the invention. However,
widths shorter than about .635 cm will likely require undesirably slow
machine speeds and nip widths wider than 2.54 cm will likely require
backing rolls of undesirably large diameter and/or softness. It is
preferable for the backing roll surface to have a P. ~J. hardness of
about 4 or harder at operating temperatures to develop the desired nip
width and pressure. To maintain this hardness may require internal
cooling of the roll, since the typical resilient roll materials become soft
very quickly at elevated termperatures. An example of a roll which
can perform satisfactorily in the invention is disclosed in U.S. Patent
No . 3,617,445.
The following examples illustrate the invention.
Example 1
An uncoated and uncalendered bodystock of a mixture of Northern
B~ hardwood and softwood fibers produced in a Kraft pulping process was
-12 -
unwound from a roll and passed through an apparatus simiiar to that
illustrated in Fig. l. The web had been mineral filled and sized to
have 10% ash content by weight, and -the web weighed 93.3 g/m2 (63
pounds per ream of 3300 ft2). The finishing apparatus was operated
with only one nip at a force of 175 KN/M (1000 pounds per lineal inch)
and a nip width of .47 cm (.185 in). The temperature of the web was
about 26.7C (80F) just before entering the nip. The moisture
content of the web was measured to be 4.8-o of the bone dry weight of
the fibers.
The web was passed through the finishing apparatus at 1.02 m/s
(200 feet/min), resulting in a dwell time in the nip of 4.5 milliseconds.
The temperature of the drum was adjusted throughout the test from a
surface temperature of 82.2C (180F) to 171.1C (340F), and samples
of the finished product were taken at various intervals. The samples
15 were tested for 75 Hunter gloss values and Parker Print-Surf values,
which were plotted against drum surface temperature in Fig. 2.
Example 2
A bodystock like that of Example 1 was coated on one side with
a conventional pigment binder coating having a weight of 14.8 g/m2
20 (lO pounds per ream of 3300 ft2), dried and passed through the same
apparatus and same procedure as Example l, except the finishing drum
surface temperature was adjusted from 25.6C (78F) to 190.6C
(375F). The coater was in line with the finishing apparatus. The
moisture content of the coated web was about 3.9% of the bone dry
25 weight of fibers. The temperature of the web was about 48.9C
(120F) just before enterin~3 the nip. Samples were taken for different
temperature intervals and tested for 75 Hunter gloss values and
Parker Print-Surf values, which were plotted against drum surface
temperature in Fig. 3.
Because the data was a little scattered due to the small number of
readings taken on each sample involved, a ratio between gloss and
Parker Print-Surf was determined (which was constant) and an
-13-
on-machine produced gloss curve (which measured a large number of
samples) was used to produce the gloss curve and to determine the
proper curve within the Parker Print-Surf points.
Example 3
A bodystock like that of Examples 1 and 2 was coated on both
sides with coatings of the same type and amount as in Example 2 and
passed through a finishing apparatus in line with the coater and
similar to that employed for Examples 1 and 2, but with two finishing
drums. Each of the drums had two resilient backing rolls forming a
pair of nips. One side of the paper was finished against one drum
and the other side against the other drum. The nip pressure for the
first drum was varied during the test from 263 KN/M (1500 pounds per
lineal inch) to 333 KN/M (1900 pounds per lineal inch). The nip
pressure on the second drum was held at 333 KN/M (1900 pounds per
lineal inch) and its drum surface temperature at 162.8C (325F)
throughout the test. One of the resilient backing rolls on the first
drum was removed during part of the test. The moisture content of
the web was about 4.7-O just prior to the first drum and about 0.5%
less at the second drum. (The decrease was due to evaporation of
moisture from the heated web surface between drums.) The web was
passed through the nips at 8.89 m/s (1750 feet per minute). The nip
widths were about 2.21 cm (.87 in), resulting in a nip dwell time of
about 1.5 milliseconds. The temperature of the web was about 71.1C
(160F) just before entering the first nip. Samples of the product
produced were taken at the following conditions for the first side and
first finishing drum.
Sample No.No. of NipsPressure Temperature
2 333 121C
2 2 333 147.8C
3 2 263 1~7.8C
4 1 263 147.8C
1 263 135C
6 1 263 121C
7 2 263 121C
77
The samples were tested on the first side for 75 Hunter gloss
and Parker Print-Surf values, which were plotted against temperature
in Fig. 4. The gloss values for the second side on the same samples
were very constant (71.7, 72.5, 71.9, 71.5, 71.6, 71.7, 71.8), as
were the values for Parker Print-Surf (.95, .95, .97, .995, .96, .95,
.88). This shows the ability to control the surface properties of one
side independently from those of the other, in contrast to
supercalendering. This is believed to be possible because temperature
and not pressure is the predominant factor, and the high surface
temperature of the drum does not transfer through the web to the
other surface of the web.
Example 4
An uncoated and uncalendered bodystock of a mixture of Southern
hardwood and softwood fibers produced in a kraft pulping process was
prepared for this example. The web was mineral filled and sized to
have an ash content of about lOs~o by weight. The web weighed about
79.9 g/m2 (54 pounds per ream of 3300 ft.2). The web was coated on
one side with a conventional pigment binder coating having a weight of
12 g/m2 (8.1 pounds per ream of 3300 ft.2), dried and passed through
an apparatus similar to that illustrated in Figure 1. The apparatus
was operated with both finishing nips. Operating runs were made at
four different nip loads: 78.8 KN/M (450 pounds per linear inch),
122.5 KN/M (700 pli), 157.5 KN/M (900 pli) and 262.5 KN/M (1500
pli). These loads produced nip widths of .816 cm (.321 in), .912 cm
(.359 in), .99 cm (.391 in) and 1.27 cm (.50 in) respectively, and
average nip pressures of 9,500 KN/M2 (1400 psi), 13,500 KN/M2 (1950
psi), 16,000 KN/M2 (2300 psi), and 20,500 KN/M2 (3000 psi),
respectively .
The coater was in line with the finishing apparatus. After
coating, the web was dried and it entered the finishing apparatus at a
moisture content of about 4.0O of the bone dry weight of the fibers
and with a web temperature of about 60C (140F) just before entering
the first nip. The web was passed through the finishing apparatus at
a speed of 2.73 m/s (500 feet/min.), resulting in nip dwell times of
3.21, 3.59, 3.91, and 5.0 milliseconds for the aforementioned
pressures. At each of these pressures, the temperature of the drum
surface was allowed to drop from a starting surface temperature of
177C (350F) to a temperature of 110C (230F) at the finish while
taking on-machine measurements for 75 Hunter gloss values. The 75
Hunter gloss values for each nip load were plotted against drum
surface temperature in Fig. 9. It should be noted that off-machine
measurements of actual samples indicated that gloss values were several
points higher than the on-machine measurements, but the latter
measurements were plotted because of the advantage of much greater
number of samples.
Referring now to Fig. 2, the curve is shown in two portions, the
left covering temperature ranges up to about 110C (230F) and the
right from about 104.4C (220F) up. On the left, one can see that
gloss and Parker Print-Surf increase at a steady rate with increasing
temperature up to about 104.4C (220F). This is believed to be the
effects from molding and coalescing the surface of the web and is what
one would expect from the prior art.
On the right side of Fig 2 is illustrated the unexpected results of
the invention. That is, at a specific temperature, about 110C (230F)
in this case, there is a sudden rapid improvement in Parker Print-Surf
for increasing temperatures. There is also a similar increase in gloss,
and this is believed to be due to the interrelationship of flatness to
gloss. This additional increment of gloss and flatness was unexpected,
but once discovered is believed to be due to the portion of the web
beneath the surface, or the subsurface strata, being heated to its
glass transition temperature and suddenly softening and becoming
moldable to allow the surface to be flattened to a greater degree than
before. The advantages provided by the thermal moldability of the
subsurface strata continue only up to about 148.8C (300F), after
which there is no improvement in gloss or flatness for the next 16.7C
(30 F )
Fig. 3 displays a similar phenomenon to Fig. 2. On the left side
one can see the Parker Print-Sur-f and gloss increase at a steady rate
--16--
with increasing temperature up to about 93.3C (200F), after which
there appears to be no further increase with increasing temperature.
This flattening of the curve is believed to be due to the behavior of
coating being thermally molded and is believed to be what one would
5 expect from the prior art. This may also explain why gloss
calendering, which is more temperature controlled than
supercalendering, was thought to have limited ability to improve
Parker Print-Surf values. On the right of Fig. 3 is illustrated the
results of the invention. At about 126.7C (260F) there is a rapid
improvement in gloss and flatness for the next 36.8C (65F). This
result is totally unexpected.
A study was undertaken to attempt to better explain the results
of the invention and to determine if the temperature at which this
phenomenon occurs can be predicted for various conditions. The
15 study starts with the belief that a substrata of the fibers in a fibrous
web can be heated to the Tg of the fibers to flatten the surface of the
web. The invention proves that this can be dGne at commercially
feasible speeds and at a moisture content which is more desirable than
those previously found necessary. To determine this temperature a
20 number of factors are involved. First, the Tg must be adjusted for
the dynamic conditions involved in high speed finishing ti.e., between
2.54 and 25.4 M/S or 500 and 5000 feet per minute). This means in
effect that the flexibility or moldability of the fibers is not only
dependent upon their temperature, but upon the rate at which they
25 are compressed. They in effect have an apparent glass transition
temperature which is based upon dynamic conditions and will be higher
than the static Tg. (Unless otherwise stated, reference to "Tg"
hereafter will refer to the apparent glass transition temperature at
dynamic conditions.) In addition, the dynamic heat transfer conditions
30 must be met to raise the temperature of the critical substrata of the
web to its Tg while in the nip.
Moisture plays a major role in determining the Tg of the fibers,
and the present invention surprisingly is capable of producing
supercalender quality at much lower moisture levels than those
employed in supercalendering. The same phenomenon which facilitates
flattening of the critical substrata in the present invention causes the
entire thickness of the web in supercalendering to be molded at a
temperature above its Tg. The reason is that the high moisture
content of paper employed in supercalendering, can res-llt in a Tg low
enough to be reached throughout the web by the temperature
conditions of supercalendering, even when unheated.
Some moisture will be lost between nips in a multinip apparatus,
due to evaporation of the moisture while traveling between nips. At
10 the low moisture levels of this invention, that amount is about 0.25% to
0.5% per nip (e.g. from 5% to 4.75% or 4.506). However, that amount
will cause a need for a significant increase in temperature in
subsequent nips. Preferably, the first drum temperature in a two
drum apparatus will be set for the moisture content at the second nip.
15 If there are two drums, the second drum temperature will preferably
be higher than the first to accommodate the lower moisture content of
the web resulting from neating at the first drum. Since satisfaction of
any needed drum surface temperature for any one nip will provide
some of the advantages of the invention, this invention includes a
process wherein one or more of the nip conditions do not satisfy the
tempe ratu re req u i rements .
Fig. 5 illustrates Tg values for cellulose fibers at various
moisture levels. The curve was derived from the experimental work of
N.L. Salmen ~ E.L. Beck (The Influence of Water on the Glass
Transition Temperature of Cellulose, TAPPI Journal, Dec. 1977. Vol.
60, No. 12) and (Glass Transitions of Wood Components Hold
Implications for Molding and Pulping Processes, TAPPI Journal, July
19$2, Vol. 65, No. 7, pp. 107-110). The curve was adjusted for the
dynamic conditions in a finishing nip. That is, the Tg values have
30 been increased over those derived by Salmen ~ Beck by about 12C.,
since the yieldability of any polymer-like material will become less for
any given temperature if the force is applied over a shorter time span.
The result is that the Tg of the material appears to be higher at
dynamic conditions than for static conditions. To make this
7~
-18-
adjustment, the Williams-Landel-Ferry equation was employed. The
very large increase in Tg for small reductions in moisture content in
the range of the invention, 3O to 7%, should be noted.
When practicing the preferred forms of this invention, the web
dwells in the nip very briefly, due to short nip widths and fast
operating speeds. For example consider nip widths of .635 to 2.5~ cm
(~" to 1") and machine speeds of 2.54 to 25.4 M/S (500 to 5000 feet
per minute). The web dwell time in the nip will be from 0.3 to 12
milliseconds. At these short dwell times, the heat from the drum does
not penetrate very far into the web.
Fig. 6 illustrates the temperature gradient into a web at 1.5
milliseconds of dwell time (corresponding to a nip width of 1.32 cm and
a machine speed of 8.9 M/S~. For this illustration, the drum surface
temperature is 138C, the web temperature prior to entering the nip is
71C, and the backing roll surface temperature is 71C. The
temperature gradient in the web was determined by the formula:
T(x,t) - To = erf X
Ti - To 2 ~
20 where: T(x,t) = temperature in C at distance X into the web
and at time t;
To = surface temperature in C of the drum;
Ti = initial temperature in C of the web entering the nip;
X = distance in feet into the web;
a ~ .005 ft2/hr;
t = time in the nip in hrs.
Fig. 7 illustrates the temperature gradient into the thickness of
the web for various nip dwell times. In this illustration the drum
surface temperature is 137.8C (280F) and the paper temperature just
prior to reaching the nip is 71C. The approximate location of the
critical substrata is believed to be about .0076 mm (0.3 mils ) deep
and is illustrated by the cross-hatched portion. It can be seen that
the temperature of the critical substrata will depend upon dwell time
and surface temperature. Whether or not the critical substrata
temperature is as high as its Tg will depend in part upon its moisture
~L~ 7
_19_
content. Thus, for the conditions illustrated in Fig. 7, the critical
temperature will be reached for moisture contents from 5% to 7.5%,
depending upon the dwell time chosen.
It should be noted here that the exaçt location of the critical
substrata is not known. The above noted location of .0076 mm (0.3
mils) into the web is an estimate based upon typical roughness of
paper, it being necessary to heat fibers down into the valleys of the
web. However, it is not critical that this assumption be correct, as
will be explained later.
Fig. 8, further illustrates the effects of dwell time, moisture
content and surface temperature of the drum in raising the critical
substrata to its Tg. The curves illustrated in Fig. 8 assume the same
.0076 mm (0.3 mils) of depth for the critical substrata as in Fig. 7
and a web temperature of 71C just prior to entering the nip. This
temperature is not uncommon where finishing takes place immediately
after coating and drying. It is expected that the webs may be at
other temperatures from ambient to about 93.3C (200F), in which
case the curves would vary somewhat.
The drum surface temperature needed for a web entering the nip
can be determined by the formula:
Ts = [Ti x .357t 479 - Tg]/[.357t 479 -1]
where:
Ts = surface temperature of the heated drum, in C;
Ti = the initial temperature of the paper entering the
nip, in C;
t = dwell time of the web in the nip, in milliseconds;
Tg = the dynamic glass transition temperature of the web
at the moisture conditions existing in the nip, in C.
The Tg can be determined from the curve in Fig. 5. A formula
which very closely approximates that curve is the foilowing:
Tg = 234.2 X e .131m
where: Tg = glass transition temperature under the dynamic and
moisture conditions existing in the nip in C;
-20-
e = the base of the natural iogarithm;
m = moisture content of the fibers in web in % of the
bone dry weight of the fibers.
The following is a guide for determining the drum surface
5 temperature Ts, in C required for the present invention for various
moisture contents, initial web temperatures and dwell times.
Ti = 26.7C (80F)
DwellMoisture Content
Time(ms) 76 6% 5-6 4% 3o
.5 160 187.2 217.2 251 288.9
1.0 132.2 153.3 177.5 204.2 233.9
2.5 114.3 132.2 152.1 174.2 198.9
107.1 123.6 141.8 162.2 1~4.9
102.7 118.1 135.4 154.7 176.2
100.8 116 132.8 151.6 172.6
Ti = 48.9C (120 F)
Dwell- Moisture Content
Time(ms) 7% 6~o 5% _ 3%
.5 137.9 165.1 195.6 229.1 266.7
1.0 119.3 140.8 164.8 191.7 221.1
2.5 107.4 125.3 145.2 167.4 192.1
102.7 119.1 137.4 157.7 180.4
99.7 115.2 132.4 151.7 173.2
98.4 113.6 130.4 149.2 ~70.2
Ti = 71.1C (160F)
DwellMoisture Content
Time(ms) 7~0 6% 5o6 4%
.5 115.5 142.8 172.8 206.~ 243.3
1.0 106.7 127.8 152.2 178.3 208.3
2.5 100.6 118.3 138.2 160.3 185
98.2 114.6 132.8 153.3 176,1
96.7 112.1 129.4 148.9 170
96.1 111.1 128.1 146.7 167.8
~2~
-21--
Ti = 93.3C (200F)
DwellMoisture Content
T i me ( ms ) 7_ 6% 5% 4O
.5 93.9 121.1 151.4 185.1 222.8
1.0 93.8 115.3 139.2 165.9 195.6
2.5 93.7 111.6 131.5 153.7 178.3
93.7 110.1 128.4 148.8 171.4
93.65 109.2 126.4 145.7 167.1
lS 93.64 108.8 125.7 144.4 165.3
Based upon the above formula developed, a needed drum surface
temperature (Ts) can be determined for each of the Examples. For
Example 1, where moisture content was 4.8%, nip dwell time was 4.5
milliseconds, and the initial web temperature was about 26.7C, the Ts
value is about 147.8C (298F). Looking at Fig. 2, this value,
15 illustrated by the line identified as Ts, can be seen to be at the top
of the temperature range where the unexpected rise in gloss and
flatness occur. The advantages of the invention actually begin about
40C (70F) lower.
For Example 2, where moisture content was about 3.9%, nip dwell
20 time was 4.5 milliseconds, and initial web temperature was ahout
48.9C (120F), the Ts value is about 161.7C (323F). Looking at
Fig. 3, this value, illustrated by the line identified as Ts, can be
seen to be at the top of the temperature range where the unexpected
rise in gloss and flatness occur also. The advantages of the invention
25 actually begin about 40C (70F) lower. This is considered good
correlation with the results for Fi~. 2.
Example 3 produced too little data to produce the full curves of
the other examples, but the temperature settings in that test were
chosen in accordance with the above formula with the intent to show
30 the inflection of gloss and flatness near the unexpected rise. Moisture
content of 4.7%, nip dwell times of 1.5 milliseconds, and initial web
temperature of 71.1C (160F) result in a calculated Ts value of about
153.9C (309F). Fig. 4 shows by the line identified as Ts where this
point is located on the gloss and flatness curves. This part of the
curve appears to correspond to the end of the unexpected rise, this
being consistent with the results from Examples 1 and 2 and the
formu la .
There are components involved in the formula which can only be
estimated. The location of the critical substrata is one already
identified. Another is the exact value of the nip dwell time. The
formula assumes that heating of the web occurs through the entire
nip, but the greatest molding pressure only occurs in the center of
the nip. Thus, the temperature reached upon exiting the nip is not
as meaningful as that reached at some point between the center and
the end. Determining what portion of the nip that should be used in
the formula is difficult and nGt necessary. Also, the meaning of
reaching the Tg of the fibers needs further explanation. The
softening of polymeric materials is a second order transition and occurs
over a range of temperature rather than sharply as in a first order
transition, such as in the melting of ice. The breadth of the range is
also a function of the molecular weight distribution with a wider
distribution giving a wider range. This same softening may occur
prior to reaching the temperature where the maximum effects are
noted. None of these components need to be known precisely to
develop a useful formula, because the formula need only be compared
to the test results in the examples and a correction made to determine
the starting and ending point of the unexpected rise in gloss and
flatness. It is not known nor important to know which component or
components have been estimated incorrectly,if any. The empirically
determined adjustment corrects them and provides a formula suitable
for determining the invention for all conditions contemplated by the
invention. The good correlation between the examples is evidence of
this .
Fig. 4 also includes in dotted lines the results of samples 3 and 7
of Example 3. They are located, as expected, slightly higher due to
increased pressure effect of 2 nips, but in a nonimproving relationship
to each other with increase in temperature. This is believed to be for
the reason stated earlier, that two nips in rapid succession are
7~
-23-
equivalent to higher drum temperature. Thus, if the solid curves
were extended into higher temperatures in the manner predicted by
Fig. 2, they would be flat. The single point represents the higher
pressure of sample 2.
Fig. 9 illustrates the unexpected large benefits of the invention
when the process is carried out at nip pressures above 13,780 KN/M2
(2000 psi). The calculated Ts is shown for each nip pressure at
approximately 142C - 147C, with each being different because of the
different dwell times (due to larger nip widths with increasing
pressure). Each curve starts approximately 20C below its Ts, and
shows the much more rapid gloss improvement which occurs at nip
pressures above 13,780 KN/M2 (2000 psi) with increasing temperature
over the critical temperature range (Ts-20-Ts). Specifically, gloss
only improves about 2 points when the nip pressure is under 13,780
KN/M2 (2000 psi), while it improves about 5 points when the nip
pressure is over 13,780 KN/M7 (2000 psi). Moreover, the improvement
at the higher pressures is at the higher gloss range where a point of
ir,lprovement is harder to obtain.
Sufficient data was collected from Example 4 to produce full
curves similar to that illustrated in Fig. 2. Only that portion near the
temperature region of Ts-20 to Ts is shown in Fig. 9 to highlight the
more greater gloss gain in that critical temperature region. On either
side of the region the curves have slopes for all pressure ranges
similar to Fig. 2.
Although the temperature benefits of the invention begin at a
temperature about 40C below the calculated Ts, the drum surface
should be heated to no less than 20C below the Ts to provide the
pressure benefits of the invention. It is even more preferable that
the drum be heated to no less than the calculated Ts to obtain all the
temperature benefits of the invention. There is no well defined
critical upper limit, but for economy and other obvious reasons it is
preferable that the Ts not be exceeded by more than about 25C,
particularly for coated paper.
-24-
lt is also desirable to limit the depth of the web heated to its Tg
to only the critical substrata. The reason is that all portions pressed
which are hotter than the Tg will be excessively densified, in the
manner of supercalendering, with the accompanying undesirable loss in
5 thickness and opacity. To obtain supercalender quality on the
surface, only the critical substrata need be so densified and any
additional flatness obtained by heating further into the web will be
costly. The greater drum temperature, slower process speed, and/or
greater sheet moisture needed to accomplish this reduce process
10 efficiency, may require more expensive equipment and greater energy
costs and can have the disadvantages of supercalendering~
Referring again to Fig. 2, another rapid rise in gloss and
smoothness on uncoated paper begins to occur at drum surface
temperatures beyond about 160C (320F), about 17C (30F) above
15 Ts. This discovery is believed to be an invention in itself. It is
believed to be thermal molding of another, deeper substrata, perhaps
providing a discrete benefit from the first because of the discrete
properties of the fibers in the web. Although
20 operating in the range of this additional benefit has the disadvantages
mentioned above, it may be valuable to do so when exceptionally high
smoothness is desired.
A further surprising and unexpected benefit was obtained from
the invention. If one were to theorize the ideal finishing operation to
25 produce glossy paper with the very smooth flat surfaces of
supercalender quality, it would be necessary to closely evaluate the
control parameters of pressure, temperature, moisture content, and
dwell time in the nip. The one most controllable is pressure, because
it can be changed precisely and instantaneously. The least
30 controllable is moisture content, since it can be changed only slowly
and is often difficult to maintain uniformly. Thus, the ideal process
would be one in which large property changes result from small
pressure changes and small property changes result from large
moistu re changes .
-25-
The present invention provides control parameters which provide
the ideal controls described above and also supercalender quality.
These advantages cannot be obtained with supercalendering because its
range for control parameters cause pressure to be the least effective
5 control and moisture the most.
The temperature effects of the invention are believed applicabie
for almost any pressure applied in the nip. That is, it is expected
that the effects of increasing pressure will follow their known curve,
except of course, the results will be significantly better. However, to
10 obtain the greatest value from the invention, the pressures will
preferably be over 13,780 KN/M2 (2000 pounds per square inch). It
is at these pressures that supercalender and bett0r quality can be
obtained, and it is at these pressures where the temperature effects of
the invention are greatly increased.
It should be noted that nip pressure determination can be
complex. Accurate nip loads (unit force per unit roll length) are easy
to determine by merely dividing the easily measured force applied to
the total resilient press roll by the easily measured nip length.
However, the nip width is more difficult to measure. A widely
20 accepted formula which is believed to provide a satisfactory
approximation of nip width for many common installations is the
Hertzian equation set forth by Narayan V. Deshpande, in Calculation
of Nip Width, Penetration and Pressure for Contact Between Cylinders
With Elastorneric Covering, TAPPI October 1978, Vol. 61, No. 10, pp.
115-118.
The formula is:
CH = [4FR (l-cr2)/(~E)]2
where:
CH = one-half the nip width;
F = force per unit length of the nip
R = the equivalent radius determined by the radii of the
heated drum (R1) and the resilient roll (R2) [R = R1R2/(R1+R2)]
a= the Poisson ratio for the resilient roll cover (0.5 for the
type of covers used in this invention); and
E = Young modulus for the resilient cover of the roll.
The Young modulus will depend upon the hardness of the resilient roll
cover. For example, roll covers having a P.~J. hardness of '1-5 at
operating temperature will have a modulus of about 517,000 KN/M2
5 (75,000 psi). The modulus changes significantly with temperature
changes in the roll cover.
The following is a guide for determining average nip pressures in
KN/M2 for various common nip loads and roll diameters (expressed in
terms of equivalent radius, R) and where the modulus of the roll cover
10 is one of the following:
MODULUS OF ROLL COVER
Designation KN/M2 x 1000 PSI x 10G0
A 3,447 500
B 1,379 200
C 689 100
D 517 75
E 345 50
F 172 25
NIP LOAD
43.75 KN/M (250 PLI) 87.5 KN/M (500 PLI)
E R R
4 6 8 10 12 4 6 8 10 12
A 20 16 14 12 11 28 23 20 18 16
B 12 10 9 8 7 18 14 12 11 10
C 9 7 6 6 5 12 10 9 8 7
D 8 6 5 5 4 11 9 8 7 6
E 6 5 4 4 4 9 7 6 6 5
F 4 4 3 3 3 6 5 4 4 4
-27-
NIP LOAD
131.25 KN/M (750 PLI) 175KN/M (1000 PLI)
E R R
4 6 8 10 12 4 6 8 10 12
_
A 33 27 23 21 19 39 32 28 25 23
B 21 17 15 13 12 25 20 18 16 14
C 15 12 10 9 9 18 14 12 11 10
D 13 10 9 8 7 15 12 11 10 9
E 10 9 7 7 6 12 10 9 8 7
F 7 6 5 5 4 9 7 6 6 5
N I P LOAD
262.5 KN/M (1500 PLI) 350 KN/M (2000PLI)
E R R
4 6 _ 10 12 4 _ _ 10 12
A 47 38 33 30 27 56 46 39 35 32
B 30 24 21 19 17 35 29 25 22 20
C 21 17 15 13 12 25 20 18 16 14
D 18 15 13 11 10 22 18 15 14 12
E 15 12 10 9 9 18 14 12 11 10
F 10 g 7 7 6 12 10 9 8 7
Although the most valuable use for the invention is to produce
supercalender quality coated paper, the principles of the invention are
believed to be applicable to any type of web of papermaking fibers,
whether coated or uncoated, groundwood or woodfree. The invention
is valuable for woodfree papers (which will be defined herein
as having at least 80% of its papermaking fibers provided by
chemical pulp), and groundwood papers (which will be defined herein
as having at least 50% of its papermaking fiber provided by
groundwood pulp) and those in between, which will comprise from S0%
to 80% chemical pulp fibers and from 20% to 50% groundwood fibers.
Any of these may be coated. Coatings for woodfree sheets preferably
will be in an amount of at least 7.5 g/m2 and those for the other
sheets preferably will be in an amount of at least 4.5 g/m2. The
invention is believed to be applicable to all conventional basis weights,
77`
-28-
including the heavy weight board products. The invention is capable
of producing, at least with the coated woodfree sheets, gloss higher
than 50 and even 70, and Parker Print-Surfs better than 1.4 and even
better than 1Ø
Although the invention is believed to provide similar advantages
to all papermaking fibers, groundwood is believed to provide an
additional result because of the large amount of lignin in the web. N.
L. Salmen has described lignin as having a static Tg at 115C (239F)
or dynamic Tg of 127C (260F) for moisture content of 2.5% and
above. (See previously cited Salmen and Beck references and also
Thermal Softening of the Components of Paper and its Effects on
Mechanical Properties, N. L. Salmen, C.P.P.A. 65th Annual Meeting,
Feb., 1979, pp. Bll-B17.) This value is equivalent to the Tg for
Cellulose at a moisture content of 4.7-0. A typical groundwood web
would have about 30% lignin, causing a similar but perhaps smaller rise
in gloss and smoothness when its Tg was reached as with cellulose. A
second and probably larger rise would occur when the Tg of the
cellulose was reached, which could be at a higher or lower temperature
than the Tg of the lignin, depending upon moisture content.
Therefore, the invention is also subjecting a groundwood web (at least
50% groundwood) to a drum surface temperature which is at least as
high as that calculated by the formula using a moisture content of
4.7%.
