Preferred citation: R. Farnood. Review: Optical properties of paper: theory and practice.

In Advances
in Pulp and Paper Research, Oxford 2009, Trans. of the XIVth Fund. Res. Symp. Oxford, 2009,
(S.J. I’Anson, ed.), pp 273–352, FRC, Manchester, 2018. DOI: 10.15376/frc.2009.1.273.

OPTICAL PROPERTIES OF PAPER:

THEORY AND PRACTICE

Ramin Farnood
Department of Chemical Engineering and Applied Chemistry
University of Toronto, February 25, 2009, Revised: May 22, 2009

ABSTRACT

The perceived value of paper products depends not only upon

their performance but also upon their visual appeal. The optical
properties of paper, including whiteness, brightness, opacity, and
gloss, affect its visual perception and appeal. From a practical
point of view, it is important to quantify these optical properties
by means of reliable and repeatable measurement methods, and
furthermore, to relate these measured values to the structure of
paper and characteristics of its constituents. This would allow
papermakers to design new products with improved quality and
reduced cost. In recent years, significant progress has been made
in terms of the fundamental understanding of light-paper inter-
action and its effect on paper’s appearance. The introduction of
digital imaging technology has led to the emergence of a new
category of optical testing methods and has provided fresh
insights into the relationship between paper’s structure and its
optical properties. These developments were complemented by
advances in the theoretical treatment of light propagation in
paper. In particular, wave scattering theories in random media are
finding increasing applicability in gaining a better understanding
of the optical properties of paper. In this document, a review of
these advancements is presented.

14th Fundamental Research Symposium, Oxford, September 2009 273

Ramin Farnood

INTRODUCTION

The perceived value of paper products depends on both their performance

and their appearance. The appearance of paper, like that of any other
material, is dependent on its interaction with visible light, and therefore is not
only a function of paper characteristics, but is also influenced by the illumin-
ation conditions and observer’s response. The growing demand for a higher
quality paper together with reduced fibre supply and market pressures have
led to significant changes in papermaking practices. To address these chal-
lenges, it is even more important to develop a fundamental understanding of
factors that influence the optical properties and appearance of paper. In this
document, a review of recent advances in characterization and theoretical
analysis of paper optical properties is presented with a focus on whiteness,
brightness, opacity and gloss.
Commission Internationale de I’Éclairage (CIE) defines brightness as “the
reflectance of blue light with a specified spectral distribution peaking at 457
nm compared to that of a perfectly reflecting, perfectly diffusing surface” [1].
This is a useful and sensitive measure of the progression of the bleaching
process [2]. Historically, brightness has been used as a measure of “white-
ness” of paper. However, the illuminant prescribed for brightness measure-
ment differs significantly from typical illumination conditions that form the
basis for common perceptions of paper quality. In recent years, the introduc-
tion of new paper products containing optical brightening agents (OBA) has
created further ambiguity in evaluating brightness and whiteness [2]. There-
fore, a great deal of effort has been dedicated to a better quantification of the
whiteness of paper through the use of suitable illuminants that mimic indoor
lighting conditions, and to provide papermakers with improved testing
methods to assess the whiteness of paper [3–5]. Whiteness measured in this
way differs fundamentally from paper brightness in that whiteness includes
the entire visible spectrum in its assessment; therefore, it is a better predictor
of the subjective ranking of paper than brightness [3].
Opacity is the attribute of paper that makes it impermeable to rays of light.
This property can be quantified as the ratio of the reflectance of a single sheet
of paper with a black backing to that of a pad of the same sample [6].
Although opacity is determined through reflectance tests, it is related to the
ability of paper to hide what is printed behind it. Over the past decade, with
the escalating cost and reduced availability of fibre resources, there is an
increasing desire to lower paper grammage while maintaining adequate paper
opacity.
Gloss refers to the degree of paper’s “shininess” and is related to the ability
of paper to reflect light in the specular direction. Gloss is a psychophysical

274 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

property that significantly influences the visual perception of the observer.

With the rapid penetration of information technology and electronic media
into the market, there has been a pressing demand to develop high quality
products with improved paper and print gloss. This has been achieved largely
through the implementation of modern coating and finishing technologies.
Therefore, fundamental understanding of gloss and factors that contribute to
it have attracted increasing attention in recent years.
In the sections that follow, an overview of optical testing of paper will be
presented and theoretical background for light reflection and scattering from
surfaces will be discussed. Following these sections, a detailed account of
recent advances in our understanding of the optical properties of paper
including whiteness, brightness, opacity and gloss will be provided. Finally,
effects of papermaking furnish – particularly pigments and fillers – on
paper’s optical properties will be reviewed.

OPTICAL TESTING OF PAPER

Quantitative characterization of the interaction of light with paper is

important not only for papermaking practitioners, but also for researchers
who study the optical properties of pulp, paper and print. Instruments used
for the measurement of optical properties of paper can be classified as
spectrophotometers, goniophotometers, microdensitometers, glossmeters,
colorimeters, and image analysis systems. Great care has been taken in the
design and construction of these pieces of equipment in terms of optical
geometry and spectral characteristics to ensure relevant and stable quantifica-
tion of optical properties. An overview of conventional measurement devices
for the optical properties of paper can be found elsewhere [7–9]. In the follow-
ing sections, the role of the illuminant in the perceived paper appearance is
discussed. In addition, brief overviews of standard methods, as well as new
laboratory techniques for the optical testing of paper, are presented.

Luminosity function and illuminant

The standard measurement techniques discussed earlier were developed to
“quantify” the response of a human observer in terms of paper appearance.
This response depends on the sensitivity of human eye to light as character-
ized by the luminosity function. The luminosity function represents the aver-
age sensitivity of human eye to various wavelengths of light. In principle, a
standard luminosity function should be applicable over a wide range in terms
of luminance and chromaticity. However, in practice, the luminosity function

14th Fundamental Research Symposium, Oxford, September 2009 275

Ramin Farnood

is only accurate for the test condition under which it is determined [10]. The
most commonly used luminosity function was originally recommended by
the CIE in 1924 and was later modified by D.B. Judd in 1951 and J.J. Vos in
1978 [11]. Most recently, Sharpe et al. revisited this topic and obtained an
improved relationship for standard daylight conditions [10]. The modified
CIE luminosity function and Sharpe’s improved function are given in Figure
1. It is particularly important to notice the difference between the functions in
the blue region that, as discussed later in this article, can have a significant
impact on the perceived whiteness of paper.
Since the sensitivity of human eye is wavelength dependent, illumination
conditions – and in particular the spectral output of the light source – are
important factors in defining the appearance of paper. Figure 2 underscores

Figure 1. A) Sharpe’s luminosity function (solid line) and Judd-Vos modified CIE
luminosity function (dashed curve). Diamonds represent the experimental data. (B)
Differences between Sharpe’s function and Judd-Vos’ function [10].

276 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 2. (a) Direct (solid curve) and diffuse (dashed curve) solar radiation at noon
on July 16 at 42°58′ N and 144°22′E [12], and (b) spectral outputs of typical
commercial incandescent (dashed curves) and fluorescent (solid curves) lamps at
100% and 30% outputs [13].

differences among the spectral irradiance of sunlight, a commercial fluor-

escent lamp, and a commercial incandescent light source. Such differences
can readily yield variations in the perceived assessment of paper optical
properties.

14th Fundamental Research Symposium, Oxford, September 2009 277

Ramin Farnood

Standard testing methods

In principle, light impinging on a sheet of paper may be reflected from,
absorbed by, or transmitted through the sheet. The relative magnitude of
each of these three components of the incident light depends not only on the
paper characteristics themselves, but also on the illumination characteristics,
such as the emission spectrum of the light source, illumination and detection
geometry, and backing material. Therefore, several standard methods have
been developed to provide guidelines for the testing of optical properties of
pulp and paper products. It is evident that the reported values for opacity,
gloss, brightness, whiteness and other optical properties of paper are
meaningful only in the context of the standard method used for the particular
measurement. Table 1 provides a list of such standard methods. A
comprehensive discussion on these methods can be found elsewhere [7, 8].

Table 1. ISO and TAPPI standard methods for optical properties of paper.

Property Standard Method Property Standard Method

Opacity CPPA E.2 Brightness CPPA E.1

ISO 2471 ISO 2470
TAPPI T 425 TAPPI T 452 (45°/0°)
TAPPI T 519 TAPPI T 525 (d/0°)
Specular Gloss CPPA E.3 Whiteness CPPA E.5
ISO/DIS 8254–1 (75°) ISO 11475 (D65/10°)
TAPPI T 480 (75°) TAPPI T 560 (d/0°)
TAPPI T 653 (20°) TAPPI T 562 (45°/0°)
Color CPPA E.5
ISO 5631 (D50/2°)
TAPPI T 527 (d/0°)
TAPPI T 524 (45°/0°)

Other laboratory methods

In addition to conventional testing methods, there has been a surge of new
laboratory techniques to investigate the optical properties of paper. This
trend is mostly driven by rapid advances in computing technology, access
to more affordable high-quality digital imaging, and the availability of laser
light sources. In particular, characterization of light-paper interactions based
on charged couple devices (CCD) has become increasingly popular. These
imaging techniques allow for the high-resolution examination of optical

278 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

properties of paper and their variations at a microscale level. Several

researchers reported using CCD imaging to capture the reflected light from
the surface of a sample mounted on a cylindrical sample holder [14–16]. This
technique is capable of providing angularly resolved information concerning
light reflection. Arney et al. studied the gloss of prints using this method [16].
Their device, schematically depicted in Figure 3, consisted of a light source
collinear with a cylindrical sample holder. The reflected light was detected
using a CCD camera. The illumination was s-polarized and a second polar-
izer was used as analyzer in front of the camera. By analyzing the digital
images, the bidirectional reflectance distribution function (BRDF) of the
sample was obtained. This device was later calibrated to determine absolute
reflectance values at different gloss angles, effective optical constants of
paper, and subsurface reflections [17]. Using a similar setup, Lindstrand
obtained angularly and spatially resolved information about reflectance and
apparent surface inclination. This was achieved by capturing images of the
sample while rotating the sample holder. In this way, reflectance information
was obtained as a function of the inclination angle for any given spot on the
sample. Based on this technique Lindstrand introduced a “gloss angle
smoothness” parameter that was found to correlate well with the perceived
gloss ranking of paper [15].
Variations of the above experimental setups have been used by other
researchers where the sample was mounted on a flat backing [18, 19]. Such a
configuration is advantageous in that it can provide information regarding
gloss variations over larger areas (several square centimetres). However,
compared to the previous setup, it does not provide any angularly resolved
information unless the illumination and detection angles are adjusted. Digital
imaging technology presents interesting new opportunities in optical testing
of paper; however, care must be taken to eliminate potential measurement
artefacts. The ultimate spatial resolution of such a system depends on the size
of the analysis area, the resolution of the camera, and the optical configur-
ation of the setup. With today’s CCD technology, however, it is possible to
achieve a spatial resolution of about one micron.
In a recent work, Elton and Preston reported the development of an
imaging reflectometer [20]. This device uses a polarized laser beam at a fixed
angle of incidence (75°) and a photometric imaging detector to examine the
forward scattered light within ±10° about the specular direction. This
imaging reflectometer uses two laser beams of wavelengths 635 nm and 670
nm to measure reflectance at various polarizations. A reference beam is used
to correct for fluctuations in the light source and the equipment is calibrated
using a highly polished glass reference. Using this apparatus, the authors
reported measurements of the angular distribution of intensity of reflected

14th Fundamental Research Symposium, Oxford, September 2009 279

Ramin Farnood

Figure 3. (a) Imaging micro-goniophotometer, and (b) a typical bidirectional

reflectance distribution function. [Scan length is 5 mm [16].]

280 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 4. (a) The microgloss imaging set up showing the CCD camera and the
ORIEL lamp housing for producing collimated beam of light, and (b) microgloss
image of a commercial MWC sample. [The image area is 1.8×1.8 cm [18].]

light (reflectogram) and effective refractive index, as well as optical surface

slope distribution and microroughness.
Sung and Keller developed a novel glossmeter based on laser triangulation
to obtain simultaneous measurements of local gloss and local roughness of
paper [21]. Gloss was determined over regions of about 20 × 20 μm2 using a

14th Fundamental Research Symposium, Oxford, September 2009 281

Ramin Farnood

Figure 5. Schematic diagram showing the imaging reflectometer [20].

laser diode (780 nm wavelength) and a CCD array detector. The CCD array
had 512 × 64 pixels, providing an imaging range of 150 m. Using this set up,
authors measured the local gloss variation, surface topography, and macro-
facet map of various commercial paper samples. The measured gloss values
were in good agreement with the standard 75° TAPPI gloss values. This
device allows for point-by-point examination and correlation between the
local gloss map and local macro-facet map of paper samples.
Another method for the analysis of optical properties of paper and prints
is based on the use of diffractive optical elements, or DOEs [22–24]. DOEs
provide information regarding both the amplitude and the phase of scat-
tered light. In this device, the paper sample is illuminated using a He-Ne laser
beam (632.8 nm wavelength). Both reflected and transmitted light is collected
and guided through a DOE sensor to form a 4 × 4 light spot on the focal

Figure 6. Schematic diagram of laser gloss-roughness measurement device [21].

282 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

plane that is captured by a CCD camera. These light spots are typically 30 μm
in diameter and are spaced 125 m apart on the focal plane. The difference in
the phase of the reflected or transmitted light appears as a speckle pattern
that provides information regarding local curvature, defects, and surface
anisotropy of paper. This method was recently refined to determine the paper
gloss as well as the differential gloss of printed papers [25, 26].
The techniques discussed so far are based on the far field measurements of
optical properties of paper. However, Apostol et al. argued that the far field
measurements of complex particulate media such as paper “. . . depend not
only on the bulk heterogeneities (composition and sizes of particulates, for
instance) but also on the surface roughness generated by the particles
themselves” [27]. In order to decouple these effects, Apostol and co-workers
suggested studying the optical phenomena in the proximity of the physical
interface using a near-field scattering measurement technique [27, 28]. This
device consists of a near field optical microscope with a spatial resolution of
50 nm that allows for simultaneous topographical and optical measurements
(Figure 7). The sample is illuminated by a fibre probe with combined propa-
gating and evanescent waves. This probe is positioned at a constant distance
and very near the surface (1/50 of wavelength) by the aid of an atomic force
microscope. Due to the scattering process, conversion of the evanescent waves
into homogeneous waves occurs and is detected by an avalanche photodiode.
Using this technique, the authors obtained near-field optical images and the
corresponding topography maps of coating layers. In addition, the group
developed a discrete random walk model to explain the near-field intensity
fluctuations due to local variations in the structure and dielectric properties
of the paper coating.

LIGHT REFLECTION AND SCATTERING

Fresnel equations
An analytical model for the reflection of light at the interface of a perfectly
smooth dielectric material was first formulated by Augustin Jean Fresnel
(1788–1827). Fresnel’s equations for the reflectance of s- and p- polarized
incident light are given by:

14th Fundamental Research Symposium, Oxford, September 2009 283

Ramin Farnood

Figure 7. (a) Near-field scanning optical microscopy system [28], and (b) Near-field
intensity distribution of a coating layer containing kaolin and CaCO3 with r.m.s
roughness of 22.7 nm [27].

where Rs and Rp are the reflection coefficients for perpendicularly (or s-

polarized) and parallel (or p-polarized) light, respectively. εr is the ratio of
refractive indices (= n2/n1), and m is defined as cos(θi) /cos(θt) where θi and θt

284 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

are incidence and transmittance angles with respect to the axis normal to the
surface. The Fresnel reflectance, RF, for a non-polarized beam of light can be
calculated from:

Based on Fresnel’s equations, the intensity of reflected light is only a function

of the refractive index of the material and the angle of incidence of light.
However, for a complex particulate medium, such as paper, light interaction
with randomly distributed particles affects the reflected light intensity
and determines its optical properties. Therefore, Fresnel’s model is insuffi-
cient to describe light reflection from paper thoroughly. However, before
addressing this problem, the interaction of light with a single particle will be
discussed.

Single particle scattering

Theories dealing with the interaction of light with a single particle can be
divided into three categories depending on the particle size, listed below:

1. Particles are much smaller than the wavelength of incident light where
Rayleigh-like scattering is dominant. In this case, the scattering cross
section of particles increases with the sixth power of particle size. Hence
for a given total mass of particles, coarser particles will have a more
intense scattering effect.
2. Particles are much larger than the wavelength of incident light and radi-
ation is reflected in specular manner (Fresnel reflection). In this case, the
scattering cross section is proportional to particle size to the second
power. Hence, for a given total mass, finer particles will have a higher
scattering power.
3. Particle size is comparable to the wavelength of light where Mie theory
can be applied, depending on the medium density (see below).

Mie theory describes the scattering of a plane electromagnetic wave by an

isotropic homogeneous spherical object. Based on this theory, the extinction
(τt), scattering (τs), and absorption (τa), cross sections of such a spherical
object are given by [29, 30]:

14th Fundamental Research Symposium, Oxford, September 2009 285

Ramin Farnood

where an and bn are complex functions of particle size, the refractive indices of
the particles and of the medium, and the wavelength of incident light. The
scattering (absorption) cross-section is the ratio of scattered (absorbed)
power divided by the incident power.
Mie theory has been developed for single particle scattering; however, it is
also applicable to random particulate systems with low particle volume con-
centrations (optical path < 0.1) where single particle scattering dominates.
For dense particle systems, an analytical modification of this theory has been
recently proposed by Flesia and Schwendimann [31].
In a typical papermaking furnish, the size of filler and pigment particles is
of the order of the wavelength of light; hence, Mie theory may be applied. In
contrast, fibre dimensions are significantly larger than the wavelength of vis-
ible light; however, the surface texture of fibres and the surface fibrillation
effects due to refining action may complicate the interaction of light with
individual fibres.

Light propagation in random particulate media

Paper may be considered as a random assembly of individual particles,
namely fibres and pigments. Propagation of light through a random medium
is a complex phenomenon. However, in the case of random media with a low
particle concentration, double and multiple scattering events may be ignored,
and light scattering can be regarded as single scattering by a particle. Hence,
single particle scattering theories may be applied to analyze such systems.
As will be discussed later, such an approach has been used to estimate the
scattering coefficients of paper coatings.
At high concentrations of particles, multiple scattering is no longer neg-
ligible. Theoretical treatments of light propagation in dense media can be
classified into two categories: transport (or radiative transfer) theories and
multiple wave scattering (or analytical) theories. Transport theories are based
on the transport and conservation of energy through a particulate medium
while wave scattering theories are based on Maxwell’s wave equations. In the
following section, a brief introduction to these theoretical treatments is
presented.

286 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Radiative transfer equation

The radiative transfer equation (RTE) has been widely used in models for
marine biology, underwater visibility, photographic emulsions, and propaga-
tion of radiant energy in the atmospheres of planets and stars, as well as
optical properties of materials such as papers and coatings. In contrast to
wave scattering theories, this theory is approximate and heuristic in the
sense that it is not based on a rigorous mathematical foundation. Radiative
transfer theory deals with the addition and conservation of power. A general
formulation of RTE may be expressed as [29]

Equation (7) is known as the “equation of transfer”. Here, I (r, ŝ) represents
the average power flux density at location r in the direction ŝ within a unit
frequency band centred at frequency u over a unit solid angle, np is the num-
ber of scattering particles in the unit volume, p (ŝ, ŝ′) is related to the incident
power in the direction ŝ′ that is scattered in the direction ŝ, and ε(r, ŝ)
represents the source term.
Unfortunately, there is no known general solution for the radiative transfer
equation. However, several useful approximate solutions have been developed
for RTE including two-flux models or Kubelka-Munk (KM) theory and the
diffusion approximation. These methods will be discussed in subsequent
sections.

Light scattering from rough random surfaces

Light scattering from rough surfaces is of interest in many applications
including remote sensing, medicine, and materials testing, to name a few.
Numerous analytical formulations have been proposed in the literature to
examine this problem [29, 32–35]. These models may be classified based on
the degree of surface roughness. In surface scattering, the degree of surface
roughness depends on the wavelength and direction of incident light as well
as surface topography. The criterion that is used to distinguish a smooth
surface from a rough one is expressed in terms of the Rayleigh parameter:

Here, σ is root-mean-squared (rms) roughness, θi is the angle of incidence,

14th Fundamental Research Symposium, Oxford, September 2009 287

Ramin Farnood

and λ is the wavelength of incident light. A surface is considered smooth if

gR Ⰶ 1 and rough if gR Ⰷ 1.
The two most widely used approaches for the analysis of light scattering
by random rough surfaces are the small perturbation method (SPM) and
Kirchhoff approximation (KA). The small perturbation method or Bragg
theory is based on approximating the scattering amplitude using a Taylor-
Volterra expansion of surface heights. In contrast, the Kirchhoff approxima-
tion considers a rough surface to behave optically similar to an infinite plane
tangent to the surface at the point of interest, such that Fresnel equations can
be locally applied.
The Kirchhoff approximation is the oldest and most widely used approxi-
mate method. This method is also known as tangent plane approximation
(TPA) and the physical optics theory. Strictly speaking, KA is valid for locally
smooth surfaces with large local radii of curvature and negligible multiple
scattering and shadowing effects. In the high frequency regime, this method
reduces to the geometrical optics (GO) or ray optics approximation. The
simplest and perhaps most popular form of Kirchhoff approximation con-
siders scattering of a plane, monochromatic, scalar wave by a rough surface
with invariant dielectric properties. This method, which is called scalar
Kirchhoff theory, is often further simplified by assuming isotropic random
surfaces with known statistics, leading to analytical expressions for the coher-
ent (specular) and incoherent (diffuse) field intensities, Ic and Id. In the case of
a Gaussian surface with a Gaussian correlation function, it can be shown that
the overall scattered intensity is given by [32]:

where:

Here, Io is the intensity of the coherent field scattered from a smooth surface
with the same dimensions, k is the incident wavenumber (= 2π/λ), Lc is the

288 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

correlation length, A and B depend on the angles of incidence and scattering,

and F is an angular factor that depends on the boundary conditions [32].
In addition to the above approaches, a number of unifying schemes have
been proposed in an attempt to combine KA and SPM. An example of such a
theory is the small slope approximation (or SSA) developed for surfaces with
small roughness slopes [36]. Figure 8 provides an overview of the domain
of validity of these methods in terms of dimensionless roughness and
correlation length. A critical review of these methods can be found elsewhere
[37, 38].
In addition to the above approximate methods, a wide range of rigorous
numerical solutions has been proposed for scattering from random surfaces.
A review of these methods can be found in [39].
In recent years, the Kirchhoff approximation has been used by several
researchers to examine the optical properties of paper [40, 41]. This topic will
be discussed in more detail in later sections.

Figure 8. The validity domain of various approximate solution methods for

scattering of light from rough random surfaces: kσ vs. kLc where k is the
wavenumber [38].

14th Fundamental Research Symposium, Oxford, September 2009 289

Ramin Farnood

KUBELKA-MUNK THEORY

The Kubelka-Munk theory is a two-flux simplification of the radiative trans-

fer model. It is based on treating a turbid random medium, such as paper, as a
continuum in the sense that the size of optical inhomogenities is considered
small compared to the thickness of the medium. Attenuation of light in this
medium is represented by two fluxes propagating in opposite directions, i.e. a
forward scattering flux and a backward scattering flux, that are estimated
using absorption and scattering coefficients, K and S. This theory provides
good predictions for diffuse illumination of dull materials that intensely
scatter light. KM theory is not suitable for collimated illuminations and it
underestimates the transmittance by several percent if K/S > 0.01 [29, 42]. In
such cases, a multi-flux model would provide a better prediction of the
optical properties of the medium. A rigorous derivation of KM theory can be
found elsewhere [43]. In this section, several key results of this model with
respect to the optical properties of paper are presented.
Using KM theory, it can be shown that the reflectance of an opaque pile of
homogeneous sheet of paper, R∞, is given by [43]:

For a single sheet of paper with grammage W, the reflectance of the sheet
backed by a non-reflective black cavity, Ro, and the transmittance, T, are
obtained from:

The above equations are applicable to a single homogeneous layer. However,

they may be extended to multi-layer materials such as coated papers. For a
two-layer structure (e.g. coated paper), where layer 1 (i.e. the coating layer) is

290 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

backed by layer 2 (i.e. base paper), the reflectance and transmittance, Ro,12
and T12, may be estimated from [8]:

Here, subscripts 1 and 2 refer to layers 1 and 2, respectively.

In the case of a large enough pile of such a two-layer material, the reflect-
ance, R∞,12, can be estimated from:

Absorption and scattering cross sections

It has been shown that K and S are related to radiative transfer theory’s
scattering and absorption cross sections, i.e. σs and σa. In their simplest forms,
these relationships may be expressed by [44]:

In this equation, σ s′ = σ (1 − g), where g is the asymmetry factor. This factor

is the average cosine of the scattering angle that can be determined from:

p(θ) is the distribution of the scattering angle, also known as the phase
function.
Several other models have been proposed in the literature, leading to
expressions similar to those of KM theory. In particular, Silvy used a

14th Fundamental Research Symposium, Oxford, September 2009 291

Ramin Farnood

corpuscular approach to study the interaction of light with non-

homogeneous non-emitting dense media that absorb much less light than
they scatter. By applying statistical laws for scattering and absorption
(represented by elastic and inelastic collisions) and using the diffusion
approximation, he arrived at an equation similar to KM theory with the
absorption and scattering cross sections are given by [45, 46]:

Mudgett and Richards compared the predictions of KM theory with those of

a many-flux solution to RTE and obtained empirical expressions for the
scattering and absorption of the medium [42]:

where ao and a1 are Legendre polynomial coefficients of the phase function.

In a more recent study, Jacques proposed a semi-empirical model for the
reflectance of a semi-infinite medium based on the diffusion theory of light
[47]. This model, similar to the Beer-Lambert law, is expressed by:

where R∞ is the total diffuse reflectance, Co is a constant, and δ is the penetra-

tion depth that depends on the transport scattering and absorption cross
sections, i.e. σs and σa, and is given by:

By comparing Equation (26) with the numerical solution of RTE, Thennadil

showed that Equation (25) is accurate with better than 1% error for σa/
σ s′ < 0.1 [48]. He also found that for diffuse illumination, Co = 4.0, while for
collimated illumination, Co is dependent on the anisotropy factor and may be
empirically described as a second order polynomial:

The transport scattering and absorption cross sections in Thennadil’s

292 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

equation are related to the absorption and scattering coefficients of KM

theory according to:

A comparison between predictions from the solution to RTE and those of

KM theory based on Thennadil’s model for collimated illumination is shown
in Figure 9. For comparison, results of KM theory for diffuse illumination
(Co = 4.0) are also plotted in Figure 9.
There are several other theoretical models that relate RTE cross sections to
the KM absorption and scattering coefficients. These results may be
expressed by the following general equations:

where m1 and m2 are constants given in Table 2.

Table 2. Values for parameters m1 and m2 in Equation (31).

m1 m2 Ref.

3 1 [46], [49]
4
3 0 [44]
4
3 1 [50]
4 4
12 0 [48]
C 2o

14th Fundamental Research Symposium, Oxford, September 2009 293

Ramin Farnood

Figure 9. Comparison between RTE solution (x) for collimated illumination and
KM prediction using Thennadil’s model for two values of g. Solid curve: Co is given
by equation (27), dashed curve: Co = 4.0 [48].

Factors affecting K and S

Absorption and scattering coefficients of paper depend on furnish and
papermaking operations, such as fibre type, pulping, bleaching, beating, wet
pressing, and calendering, and on filler and pigment type and amount [9]. In
practice, the values of K and S are estimated from diffuse reflectance
measurements. The scattering coefficient is related to the specific surface area

294 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

and pore structure of the scattering medium. In studying thermomechanical

pulp (TMP) handsheets containing a variety of fines and fillers, Alince et al.
found a poor correlation between the specific light scattering coefficient and
the total porosity or the total surface area as estimated by mercury intrusion
porosimetry [51]. In contrast, the authors observed a good correlation when
pores smaller than 100 nm were excluded from the measurement, and hence
suggested that the pore structure smaller than this threshold does not con-
tribute to light scattering. However, by comparing Brunauer-Emmett-Teller
(BET) adsorption analysis and mercury intrusion porosimetry measurement,
Lehtonen and Dyer suggested that the concept of optically active pores could
be a measurement artefact caused by the destruction of paper’s pore
structure during mercury intrusion porosimetry [52].
Theoretically, K and S are expected to be independent parameters; how-
ever, it has long been known that S decreases for a highly absorbing medium.
Foote found the scattering coefficient of a dyed sample with a reflectance of
0.10 was two-thirds of that of an identical but undyed sample with a reflect-
ance of 0.71 [53]. Later, it was shown that the reduction of the scattering
coefficient spans over the absorption spectrum of the dyestuff [54]. This effect
was also observed by varying the amount of fines in mechanical pulp hand-
sheets [52]. To rationalize this anomaly, known as the “Foote effect”, two
possible explanations have been suggested in the literature: firstly, that dye
addition changes optical properties of the cell wall; i.e. the surface reflectance
of and internal absorption [55]; and secondly, that this anomaly is caused by
an inherent error in the KM theory [56].
For mechanical pulp-based papers, where the absorption coefficient may
affect the scattering coefficient and the pore structure can be quite variable,
the relationship between the specific surface area and scattering coefficient is
not trivial. In this case, however, it was found that the scattering coefficient is
a measure of specific surface area as long as the wavelength of light used in
measuring the scattering is outside the range of the “Foote effect” [52].
In addition to the above consideration, it is important to note that meas-
urements of K and S are based on reflectance measurements that may be
affected by both surface and bulk scattering effects. In fact, Granberg and co-
workers found that the estimated absorption and scattering coefficients are
not only dependent on the sheet structure but also can be significantly
affected by nonuniformity in surface reflectance due to surface topography
effects [57].
Over the years, several extensions for the KM theory have been proposed
to improve its domain of applicability [58, 59]. Despite these attempts, a key
issue concerning the KM theory remains: the absorption and scattering coef-
ficients, K and S, are merely model parameters that do not represent intrinsic

14th Fundamental Research Symposium, Oxford, September 2009 295

Ramin Farnood

properties of the material [60]. However, KM theory remains as one of the

most widely used theories for the interaction of light with paper and coatings,
thanks to its simplicity. A detailed discussion on the applications and
modifications of KM theory can be found elsewhere [61].

OPTICAL PROPERTIES

As discussed earlier, the perceived value of paper products depends on their

appearance. Words such as whiteness, brightness, opacity, and gloss are often
used to describe the visual perception of paper. From a practical point of
view it is important to quantify these “optical properties” by means of reli-
able and repeatable measurement methods, and furthermore, to relate these
measured values to furnish characteristics and paper structure. In what
follows, some of the recent advances in our fundamental understanding of
the aforementioned optical properties of paper will be presented.

Whiteness
Ganz described whiteness as “an attribute of colors of high luminous reflect-
ance and low purity situated in a relatively narrow region of the color space
along dominant wavelengths of 570 nm and 470 nm approximately” [62].
Whiteness is a colorimetric parameter that can be influenced by the lighting
conditions. According to CIE, whiteness can be quantified based on the
tristimulus values [63]:

Similarly, tint is defined as:

Here xn and yn depend on the choice of illuminant, and x and y are functions
of the tristimulus values, i.e. X, Y, and Z, where:

296 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Under D65/10° illumination, a specimen is to be considered white if

40 < W10 < (5 Y10 − 280) and −3 < TW,10 < 3. This definition of whiteness cor-
relates well with the visual ranking tests and it is easy to interpret. However,
the choice of illuminant is an important consideration in the measurement of
whiteness. According to ISO, there are two standards for the illuminant: D65,
that is suitable for daylight, and illuminant C that is applicable to the indoor
conditions. The main difference between these two illuminants is the UV
component of the emission.
Whiteness of a near white object is affected by the so-called “lightness” but
is not equal to it. In studying the whiteness of 35 commercial office papers,
Bonham found that changes in the brightness alone explained much of the
variability in the whiteness with the lightness (L*) being responsible for the
residual error and the scattering of data [64]. He found that whiteness could
be expressed by a linear combination of lightness and reflectance factor,
Rz = Z/Zn:

Traditionally, whiteness of paper has been controlled by bleaching and bluing

the substrate and the addition of purifying pigments [62]. Based on the defin-
ition of whiteness, a bluish shade has a positive effect on whiteness and tint.
Therefore, by introducing blue dyes to the papermaking furnish, one could
adjust the whiteness and tint of paper products [65–67]. Liu et al. reported
that the yellow tint of high yield pulp could be minimized or eliminated by
the addition of a suitable dye. They also found that with the addition of
about 4 ppm dye (based on pulp), the CIE whiteness and lightness of the pulp
reached those of market kraft pulp [65]. The addition of dye to the coating
formulation has also proved to be an effective method to adjust the whiteness
of coated papers. Table 3 shows the effect of addition of a blue dye on the
CIE whiteness and tristimulus values of a coated paper. This result shows

Table 3. Brightness, CIE whiteness and tristimulus values for various levels of
addition of a blue dye to the coating formulation in terms of % in dry dye weight [68].

Dye % CIE Whiteness Tappi X Y Z

Brightness

0 44 71 71 74 72
0.25 58 68 67 69 72
0.50 71 66 62 64 70

14th Fundamental Research Symposium, Oxford, September 2009 297

Ramin Farnood

that whiteness continuously increases while brightness decreases with the

addition of a blue dye [68].
Improving optical properties of uncoated paper by dye addition has been
the subject of several recent theoretical studies. Dye addition forms color by
subtraction, i.e. by absorbing and reducing/eliminating certain wavelengths
from the light. Using KM theory, Mäkinen and co-workers studied the effect
of dye addition on whiteness of uncoated papers. Assuming a Gaussian
absorption model for the dye, the total absorption coefficient of dyed paper,
K, was calculated as below [69]:

where Kp, and Kd are absorption coefficients of paper and dye, respectively, λ
is the wavelength of light, λc is the wavelength of maximum absorption of the
dye, and β is half peak absorption of the dye. Hence, the tristimulus value,
X10, was calculated from:

Similar equations can be written for Y10 and Z10 simply by replacing X10 with
Y10 or Z10, and x̄10 with ȳ10 or z̄10, where x̄10, ȳ10, and z̄10 are the CIE color
matching functions. Hence, whiteness can be calculated from Equation (32)
(see Figure 10). Based on this approach, Mäkinen et al. concluded that the
required amount blue dye for a modest gain in whiteness is a function of the
dye distribution; however, the maximum achievable increase in whiteness is
independent of the dye distribution.
Since whiteness is determined by measuring the reflection of light from
paper, the surface structure of paper is expected to influence the perceived
whiteness values. Makarenko and Shaykevich studied this effect by the
numerical modeling of the scattering of light from a specimen under stand-
ard illumination geometries [70]. They suggested that the resultant intensity
of the light scattered by the specimen Ir may be approximated by adding the
intensities of the volume scattering Iv and the surface scattering, Is, of the
specimen. Therefore, the relative spectral scattering coefficient of the
specimen at wavelength λ, i.e. R(λ), may be calculated from:

298 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

In this equation, Io is the intensity of scattered light from a reference speci-

men. To estimate the volume scattering, authors considered scattering from
an infinitely thick layer with spectral photon revival probability represented
by Λ. Assuming a uniform angular distribution of scattered light, the index
of scattering in direction γ can be estimated based on:

In this equation η, and ξ are the cosines of angles of observation and inci-
dence of light, and φ is an auxiliary function that characterizes the intensity
of luminous flux near the surface:

The relationships for non-spherical scattering indicatrix are more compli-

cated; more detail on this subject may be found in [70]. To account for surface
scattering, the authors considered surfaces with irregularities at the scale of

Figure 10. Effect of dye addition on ISO whiteness based on KM theory [69].

14th Fundamental Research Symposium, Oxford, September 2009 299

Ramin Farnood

the incident light wavelength to be either a) small and gently sloped, or; b)
smooth. Small and gentle irregularities mean that the root-mean-square sur-
face height is much smaller than both the wavelength of light and the size of
surface irregularities, i.e. the correlation length Lc. In this case, the scattering
cross section can be estimated from Bass and Fuks formulae:

For smooth surfaces (case b), Beckman-Spizzichino’s model may be used:

Here, RF is the Fresnel reflectance, and the angles φ, χ, and ψ are defined in
Figure 11.
Using the above equations, the intensity of reflected light from the sample
was determined by integrating the scattered flux over both the specimen and
the surface of the integrating sphere (Figure 11). This calculation was
repeated for the reference material and the relative spectral scattering
coefficient of the specimen at wavelength λ, i.e. R(λ), was obtained from
Equation (40). Based on this approach, the effects of surface roughness on
the tristimulus coordinate Y10, whiteness and tint of weakly absorbing papers
with high volume scattering were predicted.
Figure 12 shows the results of this study under three measurement condi-
tions: 1) ISO illumination with only diffuse reflectance detected, 2) ISO
illumination with diffuse reflectance and some specular reflection detected,
and 3) illuminant modeled as a cloudy sky. It was found that whiteness and
tint were dependent both on the measurement/illumination conditions and
on the surface roughness. However, the effect of roughness on whiteness was
detectable only when the specular reflectance was also taken into account. In
contrast, the chromaticity coordinates of paper were found to be independent
of both roughness and the measurement scheme.
As discussed before, a bluish shade increases the measured value of white-
ness. This fact has been exploited by papermakers in the recent years through
the addition of optical brightening agents (OBA) in the production of vari-
ous grades of papers to enhance brightness and whiteness [71]. Optical

300 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 11. Top: angles φ, χ, and ψ in equations(43) and (44). 1 and 2 are planes of
incident and scattering; bottom: reflectometer geometry showing 1- integrating
sphere, 2- light source, 3- specimen, 4- point of observation, 5- black ring, and
6- glossy trap [70].

brightening agents or fluorescent whitening agents (FWA) can shade the yel-
low tint of paper and improve the whiteness of the product. Optical whiteners
function by absorbing ultraviolet radiation and re-emitting blue light. A
comparison of various types of OBAs that are currently used in the paper
industry can be found elsewhere [72]. The addition of OBAs is particularly of
interest since there is an increasing trend towards using mechanical pulps in
traditionally wood-free grades. Liu et al. reported that by the addition of
OBAs, the CIE whiteness of laboratory handsheets made of high yield pulp
was significantly enhanced; however, OBA addition was less cost effective

14th Fundamental Research Symposium, Oxford, September 2009 301

Ramin Farnood

302 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

compared to the addition of dyes in improving whiteness. Furthermore, it has

been shown that heat can completely destroy OBAs creating optical mottle
[73], and that the optical properties of papers containing OBAs degrade
faster than papers without OBA addition both under the influence of light
and in dark storage [74]. Beside wet-end applications, OBA could be also
added to the coating formulation. One of the key issues in such applications
is the compatibility of coating formulations with OBAs. Ma et al. investi-
gated the distribution of fluorescent whitening agents in coating layers con-
taining ground calcium carbonate (GCC) and clay [75]. The results of their
study suggested that a higher OBA concentration near to the coating surface
produced a higher whiteness.
The mathematical models presented so far do not account for the fluor-
escence effects due to the addition of OBAs. In order to estimate such effects,
one may use the modified KM theory proposed by Bonham for colorant
formulation of newsprint [76]. According to this model, the reflectance of a
thick pile of paper containing a fluorescent dye at a given wavelength,
R∞,OBA(λ), may be estimated from:

Here, R∞(λ) is the KM reflectance at wavelength λ given by Equation (13),

Q(λ) is the quantum yield of fluorescent dye per unit wavelength, Io(λ) is the
intensity of incident light at wavelength λ, and F is an auxiliary function that
is defined as:

where λ1 and λ2 are limits of the absorption band of the dye, K(λ) is the
absorption coefficient at wavelength λ for paper with OBA, Ko(λ) is the
absorption coefficient at wavelength λ for paper without OBA, and A(λ) is
equal to 冪K2(λ) + 2 K(λ) S(λ). Bonham used the above model to back-

Figure 12 (opposite). Effect of surface roughness on the tristimulus coordinate, Y10,

whiteness, and tint for a smooth (left) and rough (right) surface under three model
illumination schemes: 1- ISO illumination with only diffuse reflectance detected,
2- ISO illumination with diffuse and some specular reflectance detected, and
3- illuminant is modeled as cloudy sky [70].

14th Fundamental Research Symposium, Oxford, September 2009 303

Ramin Farnood

calculate the absorption and emission spectra of the fluorescent dye in a

newsprint sample, and applied the results to color prediction. Once the tri-
stimulus color values are determined, one can readily calculate the whiteness
value.

Brightness
Brightness is defined as the reflectance of blue light with a spectral distribu-
tion peaking at 457 nm compared to that of a perfectly reflecting, perfectly
diffusive surface [1]. Brightness has been used for more than sixty years to
monitor the pulp bleaching process. The wide application of brightness as a
measure of quality is due to three reasons: 1) brightness measurement is
simpler than measuring color (which requires 3 values); 2) this measurement
is especially sensitive to the blue region of the spectrum where pulp bleaching
has its main effect, and; 3) the brightness value is related to the perceived
whiteness [5]. There is no strict relationship between brightness and CIE
variables; however, brightness values are very similar to the reflectance factor,
Rz, defined by [64]:

Here, Z is the tristimulus coordinate, and the value of Zn depends on the

illumination characteristics [63]. Brightness can be also expressed in terms of
the CIELAB color coordinates, L*a*b*, that are defined as:

Therefore,

However, since the measurement of brightness is limited to the blue region of

the spectrum, it is not an adequate representation of perceived whiteness as
assessed under indoor or outdoor illumination. Furthermore, brightness is
not sensitive to papers treated by the deinking process; any ink residues
mostly absorb larger wavelengths [5].

304 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Based on its definition, brightness may be estimated from Equation (13)

with the scattering and absorption coefficients at 457 nm. However, as
pointed out earlier, OBAs are frequently added to pulp and paper products to
improve the brightness. In such cases, the modified KM theory has to be used
to account for fluorescence effects as in Equation (45).
Paper brightness is affected by bleaching, degree of beating, wet pressing,
type and amount of fillers and dye and OBA addition, as well as coating and
calendering. Addition of fillers improves the brightness of paper by increas-
ing the scattering coefficient due to increased specific surface area. On the
other hand, as discussed earlier, dyes and OBAs improve brightness by
imparting a bluish shade to the paper. The pore structure of paper influences
the brightness as well, such that higher calendering levels generally cause a
reduction in brightness due to the collapse of fibres and pores. In addition, it
has been reported that brightness uniformity is adversely affected by poor
formation in OBA-containing papers [77].
In the case of coated papers, the coating layer significantly affects the
product brightness. Pigment type, average size and size distribution, and
shape affect the ability of coated papers to scatter light and hence in-
fluence brightness and opacity of the paper. Light scattering efficiency of a
coating layer is determined by its pore structure. In studying GCC and pre-
cipitated calcium carbonate (PCC) pigments, Hallam and Hiorns demon-
strated that the coating scattering coefficient closely followed the Mie theory:
light scattering per void increased by increasing the average distance between
TiO2 particles as well as the pore diameter and peaked when the pore size was
about 400 to 450 nm [78]. For mechanical pulp containing papers, a coating
layer not only improves brightness but also enhances the brightness’ stability
[79].

Opacity
Opacity is a measure of perceived covering ability of paper. Although this
quality of paper is related to its non-transparency (low transmittance), in
practice, it is quantified through reflectance measurements.
There are two main standard methods to measure opacity: TAPPI 425 and
ISO 2471 (TAPPI 519 is another method similar to ISO 2471), where opacity
is defined as:

14th Fundamental Research Symposium, Oxford, September 2009 305

Ramin Farnood

Here, Ro is the reflectance of a single sheet of paper backed by a black

backing, R∞ is the reflectance of an opaque pile of paper, and R0.89 is the
reflectance of a single sheet of paper over a standard backing with a reflect-
ance of 89%. It is important to keep in mind that the ISO standard is deter-
mined at diffuse illumination and 0° detection angle, while the TAPPI T425
standard is based on 15°/diffuse geometry. Both methods are based on the
human vision model and use the CIE 1931 color matching function, ȳ(λ).
Opacity may be estimated based on KM coefficients and using Equations
(13) to (15) and recognizing that R0.89 can be expressed by [80]:

The above opacity standards were developed to quantify the degree by which
the printed areas appear darker once viewed from the reverse (unprinted) side
of a single sheet of paper. By definition, opacity standards are based on
lightness (or equivalently tristimulus Y) only. This creates a problem with the
perceived opacity of coloured papers [81]. For coloured papers, the black
print on the back of paper changes hue and saturation as well as lightness.
Therefore, the standard opacity measurements are restricted to white or near
white samples [82]. To address this issue, Mäkinen et al. proposed the use of
the CIE 1994 standard for color difference to quantify the covering ability of
papers:

where ΔL*, ΔC*, ΔH* are differences in the lightness, chroma, and hue,
respectively:

Here, C*i (i = 1, 2) is the chroma defined as 冪a*i + b*i , and kL, kC, and kH may be
used to correct for deviations from CIE 1994 reference viewing conditions.
Mäkinen et al. analyzed the opacity of three sets of printed samples: black

306 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

prints on white papers, black prints on color newsprint, and color prints on
white paper. They found that the standard opacity; i.e. that described in
Equation (52), was unable to predict the hiding ability of coloured papers,
while CIE 1994 color difference (Equation (55)) adequately ranked and quan-
tified the perceived opacity of all samples.
Using KM theory presented by Equations (37) to (39), Mäkinen et al.
studied the effect of dye addition and distribution on paper opacity [69].
They found that it was challenging to achieve both a good opacity and a high
whiteness by dye addition. Furthermore, since opacity depends on light
transmittance, dye distribution was found to have little or no effect on opa-
city. However, if paper whiteness was limiting, adding dye to the centre of
paper was found to give a higher opacity while maintaining a good level of
whiteness.
It is worthwhile to point out that for a rough scattering medium such as
paper theoretical analysis has shown that the transmitted light is affected by
the surface roughness [83]. Although this effect may not be significant for
diffuse illumination measurements, it could play a role in directional opacity
tests, i.e. TAPPI T425 15°/diffuse geometry.

Figure 13. Angular dependence of the transmitted flux through a medium with a
refractive index of 1.5 relative to the ambient as a function of Lc/σ. The incident angle
is 20° [83].

14th Fundamental Research Symposium, Oxford, September 2009 307

Ramin Farnood

Specular gloss
Gloss, as previously mentioned, is an attribute of paper that relates to its
“shininess”. Gloss not only depends on material structure and characteristics
but also on measurement geometry and illumination conditions. It is quanti-
fied as the ratio of the specular reflection of a sample, R(θ), to that of a
reference material, Rref (θ):

The traditional reference material used for gloss measurements is a polished

black glass with refractive index of nref = 1.527. However, recent studies have
shown that black glass standards are prone to aging, contamination, and
damage. Therefore, a new primary standard has been recently developed by
NIST that is a highly pure BaK50 glass with a refractive index of 1.5677 [84].
According to Fresnel’s equations, gloss is expected to depend on the angle
of incidence and refractive index of the material. For rough materials, how-
ever, experimental evidence suggests that gloss is also a function of the sur-
face texture of paper.
By studying high-gloss coating layers applied on smooth films (i.e. gR Ⰶ 1),
Gate et al. found that gloss could be predicted based on the rms roughness of
the coating layer according to the Beckman-Spizzichino relationship [85]:

where Gm is the maximum gloss achievable by a given coating layer as rms

surface roughness approaches zero, i.e. σ → 0. The absolute value of the
argument in the exponential term, gR, is the Rayleigh parameter; as defined
by Equation (9), that signifies the degree of light scattering at the surface.
Oittinen argued that roughness of coated paper arises from two sources:
the coating itself and the base paper [86]. Generally, roughness of paper is
two orders of magnitude larger than the wavelength of light while the rough-
ness due to coating pigments is of the order of wavelength of light. Accord-
ingly, a two-scale model was proposed where the total light reflection was
assumed to be the product of the contribution due to the optical roughness
(originating from the coating layer) and that of microscopic roughness (ori-
ginating from the base paper nonuniformity). In addition to surface rough-
ness, Oittinen showed that the local gradient of surface heights also affects
the gloss of paper. In a subsequent study, Lipschitz and co-workers used an
optical surface profilometer and reported that the distribution of local slopes,

308 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

or facet angle distribution, correlates well with changes in the gloss values of
coated papers [87]. Furthermore, it was found that facet angle could explain
78% of the measured gloss variation [88], and that there is a good correlation
between mean facet angle and visual ranking [89]. Chinga explored the rela-
tionship among various surface descriptors and gloss using multivariate
analysis [90]. He reported that gloss correlated well with the median facet
angle as well as surface pore volume while the conventional Parker Print-Surf
(PPS) roughness was only weakly related to the gloss (Figure 14). Chinga and
co-workers later examined this observation and proposed that PPS roughness
is mostly affected by surface features on the order of 80–160 μm, while gloss
is dominated by micro-structures below 12 μm [91]. Hence, by representing
the surface of paper as a collection of micro-facets, several researchers
developed mathematical models to estimate the intensity of scattered field.
For a polarized collimated incident beam of light Io, and assuming that light
reflection from each fact follows the laws of geometrical optics, one can show
that the intensity of scattered light, Is(θs, φs, over the solid angle is dωs [20]:

Figure 14. Relationship between paper and print gloss with the facet angle of
unprinted SC papers [91].

14th Fundamental Research Symposium, Oxford, September 2009 309

Ramin Farnood

where ff(θf, φf) is the distribution of facet angles and RF(θ, n) is the Fresnel
coefficient based on the local angle of incident θ. Following Oittinen’s model,
Elton and Preston suggested that the overall light reflection is the result of
two contributions: macroroughness and microroughness of paper. Hence, the
s-polarized light intensity, Is, was suggested to be :

The term dIv represents the contribution of subsurface multiple scattering. A

similar equation may be obtained for the p-polarized illumination. Using the
above equation, the authors reported a good correlation between the meas-
ured and predicted gloss values. In their case, gloss was measured using an
imaging reflectometer (Figure 5). Elton and Preston also examined the effect-
ive refractive index of coated papers and found that for a given pigment type
the refractive index decreases linearly with the mean pore diameter [92].
Hansson used an angle resolved light scattering measurement method to
analyze a wide range of commercial papers [14]. This device used a 2 mm2
He-Ne laser beam as illuminant and a rotating sample holder to detect the
angular distribution of scattered light. Using geometrical optics to estimate
the surface reflection and Lambert’s law for the bulk (subsurface) reflection
of paper, he proposed the following expression for the intensity of scattered
light:

where Rd and Rs are the diffuse and specular reflectance coefficients, θi is the
illumination angle, and σs is the standard deviation of surface slopes. Figure
15 shows the experimental and predicted microgloss images of a light-weight
coated paper sample based on Equation (60).
In studying laboratory coated papers with gloss values ranging from about
20% to above 80%, Caner et al. reported that Rayleigh parameter, gR, varied
from 5 to more than 800 [94]. They demonstrated that under such conditions,
as expected, the smooth surface approximation described by Equation (57)
was no longer applicable. Instead, gloss was a function of both rms roughness
and correlation length, Lc, of the surface topography. This is consistent with
the earlier findings of Alexander-Katz and Barrera who reported that the
surface height correlation cannot be neglected in the prediction of gloss [93].
To better illustrate the significance of correlation length on the topography,
Figure 16 illustrates two Gaussian random surfaces with the same rms
roughness but different correlation lengths and their corresponding diffuse

310 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 15. Experimental (left) and simulated (right) microgloss images of a light
weight coated paper [14].

scattered field intensities. Caner et al. found that gloss of coated papers is a
function of surface texture parameter, Lc/σ2, according to the following
semi-theoretical expression:

Values of b and n for various standard gloss angles, θ, are given in Table 4.
Despite wide variations in the surface roughness and coating formulations
(including fourteen clay and GCC pigments), the above equation was able to

Table 4. Values of b and n for equation (61).

Gloss Angle, θ B n

20° 0.5 0.41

60° 4.1 0.52
75° 19.3 0.31

14th Fundamental Research Symposium, Oxford, September 2009 311

Ramin Farnood

Figure 16. Top: simulated random surface profiles for a Gaussian distribution of
heights and an exponential height correlation function for σ/λ = 0.4. λ is the
wavelength of light, bottom: effect of correlation length on the diffuse scattering field
[93].

predict the standard gloss with a regression coefficient ranging from 0.61 to
0.78 (Figure 17).
It should be pointed out that the values of rms roughness and correlation
length in Equation (61) were measured using an interferometric profilometer

312 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 17. Relationship between standard gloss and surface texture parameter, Lc/
σ2, at three gloss angles [94].

(WYKO NT2000) over an area of 460 × 680 μm2 with a horizontal resolution
of 0.82 μm. However, the surface statistics of paper are dependent on the
measurement resolution and sampling area. To illustrate this effect, Figure 18
shows that by increasing the measurement size up to 6 mm, the rms rough-
ness for a coated paper sample increased and reached a near-plateau after
about 2 mm. In contrast, the correlation length continued to increase with the
sample size and at 6 mm reached nearly 500 μm; three orders of magnitude
greater than the wavelength of visible light. A similar trend has been reported
elsewhere for several office papers over a measurement length of 1.5 mm [95,
96]. These observations have significant implications in terms of applicability
of various wave scattering theories and they also illustrate challenges
involved in the measurement of surface topography of paper. In an attempt
to identify the suitable resolution that is optically responsible for gloss,
Vernhes et al. compared the experimental scattering indicatrix with the
predicted values using a virtual goniometer [97]. Their study showed that
the best fit between experimental and measured values was obtained if the
surface was measured at a high magnification.
With recent advancements in computational speed, structure-based model-
ing of paper optical properties is becoming increasingly affordable. For
example, Green et al. studied the angle-resolved light reflectance from paper

14th Fundamental Research Symposium, Oxford, September 2009 313

Ramin Farnood

Figure 18. Dependence of rms roughness (a) and correlation length (b) on the
sample size, L, measured by an interferometric profilometer. Sample is a kaolin coated
papers with 10 pph SB latex and lateral resolution is 0.82 μm [99].

based on geometrical optics (ray tracing) techniques and Monte Carlo simu-
lation. They modeled paper as a three-dimensional network of randomly
distributed cylindrical fibres with varying dimensions (i.e. length, diameter,
fibre wall thickness) and orientations [98]. Although the geometry used in this

314 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

study was a crude representation of paper structure, their study was a signifi-
cant step towards the prediction of paper’s optical properties based on paper
structure and furnish.
A fact that is commonly overlooked in gloss measurements is the depend-
ency of the measured gloss value on the angle of acceptance of detector, δθ.
Oittinen found that the measured gloss values for a polished black glass
increased linearly and reached a plateau by increasing δθ. However, it is to be
expected that this dependency would be significantly different for rough sur-
faces such as commercial papers. To study this effect, Arney and co-workers
proposed that gloss may be estimated using a circular BRDF function [100]:

Here, w is the half width of BRDF, and K is the area under the BRDF for the
reference material. The angle of acceptance of the gloss meter was estimated
by reverse engineering. By ignoring shadowing effects, authors were able to
predict the experimental gloss data and show that the angle of acceptance of
the gloss meter had a strong effect on the predicted gloss values.

Modelling gloss of rough surfaces using KA

Using scalar Kirchhoff theory and assuming surface height is a stationary
Gaussian random variable with a modified exponential correlation function,
Lettieri et al. studied the reflection of light from the surface of coated papers
[40]. The authors found a good correlation between the angle-resolved light
scattering data and the model predictions. In a later study, Alexander-Katz
and Barrera developed analytical models for light reflection from stationary
Gaussian surfaces with either Gaussian or exponential autocorrelation func-
tions. With some simplifying assumptions, for the asymptotic case where
surface is rough (i.e. gR Ⰷ 1), and where surface slope (≈ σ/Lc) is large
enough to create a nearly isotropic diffuse field, they obtained the following
expression for the specular gloss [93]:

Here, δθ is the detector collecting half-angle, RF and RF,o are the Fresnel
reflectance values of the sample and that of the standard, ΩD is the solid
angle of acceptance of the detector, and yD can be expressed as:

14th Fundamental Research Symposium, Oxford, September 2009 315

Ramin Farnood

In Equation (63), m is a constant that depends on the form of the surface

height autocorrelation function. The value of m is equal to 1 for the
exponential correlation function and 2 for Gaussian correlation function.
For large values of the Rayleigh parameter, the integral term in this equa-
tion can be further reduced to give the following simple expression for gloss:

For an exponential autocorrelation function (m = 1), Equation (65) predicts

that the specular gloss is proportional to Lc/σ2. This contradicts experimental
observations that gloss is a nonlinear (power-law) function of Lc/σ2. To
resolve this issue, we recently proposed an analytical expression for the gloss
of paper based on first-order Kirchhoff approximation [41]. A detailed
derivation of this model is provided below.
As discussed earlier, in the Kirchhoff approximation, the scattered field
could be fully characterized based on the incident illumination, surface top-
ography, and the Fresnel coefficients. In this case, assuming that the scattered
field is concentrated around the specular direction, the components of
scattered field are given by [101]:

where Eo is the amplitude of incident field, Ro is the distance from the surface
to the detector, ko is the incident wave number, I1 is a function that depends on
the position and elevation, and F is a function of Fresnel coefficients and
measurement geometry.
Bourlier and Berginc estimated the elevation difference between two
adjacent points using zi − zi ≈ sj . (ri − rj), where sj = sjx→
x + sjy→
y is the local
surface slope and (see Figure 19):

Based on this approximation and by ignoring the shadowing effects, the

scattered intensity becomes:

316 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 19. Top, first order Kirchhoff (or tangent plane) approximation, and bottom
incident and scattered fields geometry [101].

where A is the illumination area and fs is the probability distribution function

of slopes. For a given glossmeter, the incident intensity and geometrical
parameters remain constant. Hence, recalling that gloss is the ratio of the
specular reflection of the sample to that of a reference material, the specular
gloss measured by a receiver with angles of acceptance of (±δθ, ±δφ), is given
by:

14th Fundamental Research Symposium, Oxford, September 2009 317

Ramin Farnood

Theoretically, Go is the limiting value of gloss for a perfectly smooth surface

and hence can be determined from the Fresnel equation. To evaluate the
above equation, it is assumed that surface slopes in x and y directions are
independent and identically distributed exponential random variables with
the standard deviation of σs, i.e.:

The rms of the surface slopes may be either calculated directly from
topographical data, or estimated recognizing that σs = 2 σ/Lc [102].
Integration of Equation (71) provides an estimation of the gloss. However,
further simplification can be considered to derive closed-form analytical
expressions. In the case of specular gloss, the angle of acceptance of the
receiver is relatively small. Hence, based on Taylor expansion and neglecting
higher order terms, one can obtain:

Finally, from (71) and (72), the following approximation for the specular
gloss is obtained:

Although Equation (73) is derived for rectangular apertures, numerical

studies have shown that it provides a good estimation of gloss for circular
detector geometries by assuming δθ = δφ.
As was pointed out earlier, theoretically, Go is the limiting value of gloss for
a perfectly smooth surface (i.e. σs = 0). However, here, this value was treated
as a constant value and was determined by fitting the experimental data.
Figure 20 shows the comparison between experimental data and predicted
gloss values for 47 coated and uncoated samples with standard 75° Tappi

318 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 20. Comparison between experimental and predicted 75° Tappi gloss values
for laboratory and commercial coated papers based on Equation (73) [41].

gloss values of about 6 to 87. Equation (73) provides a good fit to experi-

mental data (r2 = 0.94).
The above theoretical and experimental studies emphasize the importance
of material characteristics as well as measurement conditions on the gloss
value. Moreover, it is clear that the intensity of reflected light in the specular
direction is a nonlinear function of surface parameters; i.e. roughness, correl-
ation length, and facet angle. At first, this latter conclusion may appear to
disagree with the recent results of Järnström et al. who found the gloss of
coated films varied linearly with the rms roughness [103]. However, the
roughness values in Järnström’s study were in the range of nanometers and
hence their samples may be considered optically smooth. Although the study
of gloss of optically smooth coated films with roughness values of the order
of wavelength of light has been used traditionally to gain useful fundamental
information [85], in practice, the surface roughness of coated paper is modu-
lated by base sheet nonuniformities that are several orders of magnitude
larger than the wavelength of visible light.

14th Fundamental Research Symposium, Oxford, September 2009 319

Ramin Farnood

Bulk scattering effects on gloss

Conventionally, it is believed that gloss is created at the uppermost layer of
paper, i.e. within a thickness smaller than half the wavelength of light [104].
Previous studies have shown that the gloss of coated papers is predominantly
a surface-dependent parameter. Therefore, in theoretical analysis of gloss, the
effect of bulk properties of coating and paper are commonly ignored.
Although this may be a good first approximation, evidence suggests that the
choice of base paper and pigment characteristics can also affect gloss and
local gloss variability through bulk scattering effects. Experimentally, the
influence of bulk and surface scattering on gloss may be isolated using polar-
ized light reflectrometry [105, 106]. Based on this method and using the set-up
described earlier (Figure 4), we recently measured the surface and bulk
microgloss for a wide range of laboratory coated papers. These samples were
coated on one side with six different formulations (including three kaolin
pigments and three GCC pigments) and were calendered to various degrees
to achieve a range of Tappi gloss from 10% to about 65%. Figure 21 shows
the measured contributions of bulk and surface scattering to the total meas-
ured microgloss. Interestingly, regardless of coating formulations and calend-
ering treatment, all samples followed the same trend lines. Surface microgloss
increased linearly as total microgloss increased while the bulk microgloss
remained nearly constant at about 100 grey level unit (corresponding to a
contribution of about 10% to the total Tappi gloss). At microgloss values
below about 250 grey level unit (GL), corresponding to a Tappi gloss of
about 30%, the measured total gloss values were dominated by the bulk
scattering effects. Further work is currently being undertaken to better
understand these phenomena.

Print gloss
Print gloss has a direct impact on the perception of print quality. Print gloss
is measured similar to paper gloss. This is a complex parameter that is
affected by paper properties, including pore structure, roughness, and surface
chemistry, and printing conditions, such as ink type and printing method.
Moreover, ink-paper interactions play a significant role in determining the
final gloss of printed papers. In the case of coated papers, it is generally
believed that the ink film covers small irregularities and hence enhances the
gloss. However, it is also known that ink film can add to irregularities through
filament patterns produced during ink splitting [108] or due to ink pigments
that protrude from the smooth ink layer [109].
MacGregor and Johansson analyzed a wide range of unprinted and
printed coated papers and reported that there was virtually no correlation

320 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 21. (a) schematic illustration of bulk and surface scatteing effects on the total
scattered field, and (b) surface and bulk microgloss of coated paper samples in grey
level units (GL) [107].

between print gloss and gloss of the base paper [110]. In contrast, Ström et al.
found a strong correlation between the print gloss and the roughness of
unprinted paper measured in PPS [111]. In a different study, Donigian et al.
observed that print gloss and paper gloss exhibited opposing trends with

14th Fundamental Research Symposium, Oxford, September 2009 321

Ramin Farnood

decreasing the size of calcite precipitated calcium carbonate (PCC) [112].

They also reported that fewer larger pores resulted in higher print gloss, and
that coating absorbancy could overwhelmingly affect the print gloss by mask-
ing any gain in the paper gloss. Similarly, for offset printed coated papers, it
has been reported that print gloss is affected by the surface pore structure
[113] as well as ink levelling and resin deletion [114]. These sometimes con-
tradictory observations emphasize the importance of printing conditions and
ink-paper interactions to the gloss of printed papers.
Similar to paper gloss, print gloss is a function of the surface roughness.
Using frequency domain analysis, Matsuda et al. suggested that the surface of
printed paper could be considered as a cluster of sine waves with various fre-
quencies and amplitudes [115]. They reported that surface profiles with spatial
frequencies more than 30 or 40 mm−1 can significantly affect paper gloss. They
also found that gloss of printed papers was affected by a larger range of spatial
frequencies than that of unprinted papers, and hence suggested that by increas-
ing the high frequency component and reducing the low frequency contribu-
tion to roughness it may be possible to increase the delta-gloss of matte coated
grades. Ström and co-workers studied the coated offset prints and concluded
that print gloss had a strong correlation (r2 = 0.82 − 0.90) with both PPS and
the “sub-macro” roughness, i.e. roughness measured over 100 × 100 μm2 using
AFM [111]. Furthermore, using Fourier analysis they reported that base paper
roughness at the scale of 50 − 250 μm was predominantly responsible for the
observed print gloss. In a later study, Ström and Karathanasis found that the
relationship between print gloss and micro-roughness of unprinted paper vir-
tually disappeared after 3 − 4 g/m2 of ink was applied on the surface of coated
paper, but a strong linear relationship was found between print gloss and
micro-roughness of prints [116]. They also observed that increasing the
amount of ink applied on a fast setting substrate increased the roughness and
created a non-linear relationship between print gloss and the applied amount
of ink due to the formation of ink filament patterns.
Using the microgloss setup described earlier (see Figure 4), we studied the
effect of the surface texture of printed papers on the print gloss. Figure 22
shows the micro-gloss of laboratory-made and commercial unprinted coated
paper (lower line) and commercial offset printed coated papers (upper line).
Similarly to paper gloss, print gloss was found to exhibit a power-law rela-
tionship with the surface texture parameter, Lc/σ2. It is also important to note
that the gloss of printed samples is consistently higher than that of unprinted
papers at the same value of Lc/σ2.
Arney et al. used their imaging micro-goniophotometer (Figure 3.a) to
study the gloss of paper and prints, and found that the reflectance of black
electro-photographic prints closely followed the Fresnel equation with an

322 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 22. Microgloss of laboratory and commercial coated papers (lower line) and
commercial offset prints (upper line) as a function of Lc/ σ2 [117].

extinction coefficient of 0 [17]. Using the same setup, Arney and co-workers
studied the effect of color on gloss. Conventionally, gloss of prints is con-
sidered to be the same color as the illuminant. However, by measuring the
relative area under the bidirectional reflectance distribution function, Arney
et al. found that the relative reflectance of color inks was not only a function
of the illumination color but also depended on the ink type (Table 5) [16]. In

Table 5. Relative area under BRDF for various inks under different illumination
color [16].

Color of Type of Ink

Incident Light
Cyan Magenta Yellow CMY

Red 1.2 1.7 1.4 1.0

Green 1.4 1.2 1.3 1.0
Blue 1.5 1.2 1.2 1.0

14th Fundamental Research Symposium, Oxford, September 2009 323

Ramin Farnood

addition, the amount of specular reflection was found to be a strong function

of the optical density of ink, such that more transparent inks were found to
be more reflective than absorbing inks. This seems to be counterintuitive since
a highly absorbing material is expected to have a higher refractive index, and
according to Fresnel’s law, will cause a higher reflectance. The authors
explained their results by suggesting that specular reflection occurs in more
than one interface and that reflection from the substrate had a significant
contribution to the overall gloss. They proposed a three-layer model where
the ink layer is in optical contact with air (instead of direct contact with
paper), and hence the overall reflectance is the sum of reflectances from the
air-ink (Ri), ink-air, and air-paper (Rp) interfaces. Defining the transmittance
of the ink layer as T, and assigning T = 0 for cyan-magenta-yellow (CMY)
ink, the relative reflectance with respect to the black ink, r, was estimated to
be [16]:

where FA is an attenuation factor that is treated as a fitting parameter.

According to Equation (74), the relative reflectance of printed samples is a
linear function of the square of the ink’s transmittance. This relationship was
confirmed by the experimental data presented by the authors.

Gloss nonuniformity
Both the average gloss and the gloss nonuniformity affect the visual percep-
tion of paper and prints. Using an x-y scanner with 400 μm spot size,
Fujiwara et al. analyzed a wide range of commercial coated papers and
showed that the gloss uniformity had the strongest impact on the appearance
of unprinted coated samples [118]. In their study of commercially printed
coated papers, MacGregor and Johansson found that the visual ranking of
printed papers correlated well with gloss variations in the range of 200 μm to
1600 μm [110]. However, evidence suggests that the scale of gloss variation
that is primarily responsible for the visual ranking may depend on the type of
samples and the visual assessment method. For example, Béland et al.
reported that gloss variations in the 3.3 mm to 6.6 mm range correlated best
with the visual ranking of 32 commercial offset prints of matte-coated papers
[89].
It has been shown that the gloss nonuniformity of printed paper corres-
ponds to variations in the local roughness of print [119]. Using atomic force
microscopy, Béland and Bennett found that low gloss areas of a printed
sample were 3 to 4 times rougher than the high gloss areas. They also

324 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

identified low gloss regions that were either smooth but tilted or flat but
rough. MacGregor suggested that up to potentially 90% of the gloss vari-
ation was accounted for by topography, and proposed that gloss evaluation
methods need to improve in order to capture the texture effects on the visual
perception of gloss variability [120].
In a recent study, we investigated the microgloss variation of coated
samples as a function of pigment type, as well as calendering load and
temperature [117]. It was found that for a given coating formulation and base
sheet, microgloss variation was governed by the surface texture parameter,
Lc/σ2. However, as seen in Figure 23, this relationship was also dependent on
the pigment type. Figure 22 and Figure 23 suggest that by appropriate selec-
tion of coating pigment, it is possible to achieve the same level of gloss with
lower micro-gloss variability.

Figure 23. Log-log relationship between the standard deviation of microgloss, σGμ,
and surface texture parameter, Lc/ σ2, for three kaolin pigments. Pigment size and
morphology are K7 : 0.22μm, undelaminated; K4: 1.18 μm, delaminated, K3N: 2.16,
delaminated [121].

FILLERS AND PIGMENTS

Papermaking furnish, together with finishing and converting processes,

determines the optical performance of paper products. A recent review of
papermaking practices that affect the appearance of paper can be found in

14th Fundamental Research Symposium, Oxford, September 2009 325

Ramin Farnood

[122]. An effective method to enhance the appearance of paper is through

the use of mineral pigments for coating and internal filler applications. Fill-
ers enhance the optical properties of paper by improving the light scattering
capacity through increasing the specific surface area. A review of pigment
characteristics and applications in papermaking can be found elsewhere
[123]. Due to higher cost and reduced availability of raw materials, there has
been a tendency to substitute fibres with fillers. Fillers are traditionally
introduced to the wet end of papermaking and are retained in the sheet
during the dewatering process. Increasing the filler content adversely affects
the mechanical properties of paper. An alternative method of incorporating
fillers in the sheet is through lumen loading [124–126]. Compared to
traditional methods for filler addition, lumen loading allows for a higher
mechanical strength at the same amount of filler content; however, this
method demonstrates a lesser degree of improvement in optical properties
(Figure 24).

Figure 24. SEM cross section showing the calcium carbonate distribution in, A)
conventionally filled handsheet with 13% filler, and B) lumen loaded sheet with 22%
filler [125].

326 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Pigment coating, on the other hand, improves the surface finish of paper.
In recent years, higher demand for improved brightness and opacity has
resulted in the increasing introduction of calcium carbonate as a coating
pigment. At the same time, as with fillers, there is a trend towards greater
pigment to fibre ratios [127]. The main pigment characteristics that affect the
appearance of paper are pigment particle size and particle size distribution
(PSD), as well as pigment shape, brightness, color, refractive index, scatter-
ing, and morphology. Significant advances have been made in designing novel
pigments with improved brightness, opacity and gloss by optimizing pigment
size, morphology and structure. Proper choice of pigment and calendering
conditions is of the utmost importance for the trouble-free production of
paper with desirable attributes.
Table 6 provides a summary of recent studies related to the effect of coat-
ing and calendering conditions on the optical properties of coated papers.
This table shows that both calcium carbonate and kaolin clay are widely used
in today’s papermaking. Kaolin clay remains a popular mineral for filler and
coating applications. Clay has a good color (white or near white) and a fine
particle size. It is inert, non-abrasive, and readily dispersible in water. Optical
properties of clay can be significantly enhanced by engineering clay products
through chemical and thermal treatment methods [128, 129]. A recent shift
towards alkaline papermaking has resulted in an increased utilization of cal-
cium carbonate in the wet-end and as coating pigment. Calcium carbonate
used in papermaking is either produced by grinding natural limestone, chalk,
or marble (GCC) or through a precipitation process (PCC). GCC has
rhombohedral morphology; however, the morphology of PCC can be
adjusted. Scalenohedral PCC is the most widely used morphology as filler
[130]. The higher brightness, whiteness, and light scattering coefficient of
calcium carbonate when compared to clay may produce a base paper with
better optical properties. However, this is not always true, since, for example,
the use of calcium carbonate as filler may lead to a more porous and rougher
base sheet that will offset any gain in the optical properties of finished prod-
uct [131]. Figure 26 illustrates the range of brightness values for commercial
papermaking fillers.

Scattering coefficient of fillers and coatings

The ability to scatter visible light is an important attribute of fillers used in
papermaking. A high scattering coefficient is desirable for the optical
properties of paper such as opacity, brightness, and whiteness. The scattering
coefficient of pigments/fillers is not only a function of their size, size distribu-
tion, morphology, and refractive index, but is also affected by the packing

14th Fundamental Research Symposium, Oxford, September 2009 327

328
Table 6. Summary of recent studies related to the effect of coating formulation and calendering conditions on the optical
properties of coated papers.
Ramin Farnood

Ref. Pigment Variable Summary of Results

[132] Clay, PCC Pigment type Adding prismatic PCC to platy clay and decreasing latex content
lowered gloss.
[133] Aragonite PCC Pigment Particle Size Brightness, paper gloss, and print gloss decreased and opacity
increased with particle size.
[134] GCC or PCC + Calendering Moisture Gloss increased with sheet moisture.
Ultra Fine Clay
[112] Calcite PCC, Clay Particle size distn., Binder Reducing particle size, increased paper gloss but decreased print gloss.
content Latex content did not have a consistent trend on gloss. Adding 50%
clay to formulation improved gloss.
[135] Unknown Coating method Gloss changed in the following order:
Blade Coating > Film coating > Curtain Coating.
[136] TiO2 Rutile & Pigment type and loading At the same target brightness, opacity increases as: Rutile > ZnS >
Anatase, ZnS Anatase.
[78] TiO2 Coating Pore Structure Light scattering increased by increasing the pore diameter and peaked
at about 400–450nm.
[137] GCC + Clay Base paper, Filler content, Film coating had lower gloss than blade coating. Higher filler content
Pre-calendering, Coating and better formation had lower gloss variability. PGW had higher
method opacity than TMP, and precalendering reduced opacity.
[138] GCC, Clay Pigment type, Calendering Pigment size and shape have strong influence on gloss. The broadness
conditions parameter had no significant influence on the gloss.

Session 2: Fibre Suspensions and Forming

 [121] GCC, Clay Pigment Characteristics, Pigment size distribution affected both microgloss and microgloss
Calendering Conditions nonuniformity. Higher gloss results in higher microgloss
nonuniformity, and this relationship depends on the pigment type.
Gloss nonuniformity of GCC is more sensitive than kaolin. Higher
calendering temperature increased gloss and gloss nonuniformity.
[139] Clay+ GCC + Pigment type Increased hollow sphere pigment increased reflectance and gloss.
Hollow/ filled Hollow sphere increased brightness and scattering coefficient more
plastic pigment effectively than filled spherical pigments.
[140] GCC, PCC, Clay Pigment type Aragonite PCC outperformed GCC. At the same target gloss, PCC
had higher opacity & brightness. PCC coated sheet had higher gloss
at lower calendering levels
[141] GCC Latex dispersion Better dispersion of latex in the coating layer improved gloss.
[142] Scalenohedral Pigment morphology High light scattering, and high opacity and brightness were achieved
PCC, porous PCC, using porous PCC compared to GCC. Increasing coat weight from 8
GCC to 16 gsm increased brightness of PCC but not GCC.
[143] PCC + Fine Clay Pre-calendering temperature Increasing pre-calendering temperature increased gloss due to better
coverage.
[144] GCC Latex type, Calendering Latex with higher gel content gave higher gloss. Higher web

14th Fundamental Research Symposium, Oxford, September 2009

speed temperature increased the gloss. Higher calendering speeds lower the
gloss.
[145] GCC, Clay Coating structure, Latex Gloss increased linearly with the reduction in pore diameter.
content Pigment type Correlation between gloss and the reduction in pore volume was
poorer.
Higher latex content decreased gloss of precalendered sheet, but had
no effect on the gloss of calendered paper.
Clay has higher gloss than GCC (with similar PSD) before and after
calendering.

329
Optical Properties of Paper: Theory and Practice
 Table 6. Continued

330
Ref. Pigment Variable Summary of Results

[146] GCC+Clay blends Pigment blend Gloss increased linearly by increasing clay content.
Light scattering coefficient increased by clay addition and peaked
Ramin Farnood

about 80–90% clay for calendered and uncalendered sheets.

[147] Aragonite PCC Pigment shape factor Gloss is mostly affected by pigment size than the shape factor.
Clay Light scattering coefficient was the same for all clay coatings, but
much higher for aragonite PCC.
[131] GCC + Clay or Filler content Increasing filler content from 5% to 15% had the same effect on
Engineered Clay brightness as replacing clay with engineered clay, but with a higher
paper and print gloss.
Compared to kaolin coating, GCC had a more open structure hence
higher light scattering but rougher surface and hence lower gloss.
[75] GCC + Clay Pigment blend Pigment size GCC with narrow particle size distribution had the highest
distn. contribution to the brightness, while GCC with broad particle size
distribution had highest contribution to whiteness.
[148] Delaminated Clay Pigment Blend Gloss was reduced with increase in talc or calcined clay content. At
Calcined Clay Talc equal calendering and equal gloss conditions, brightness increased
with delaminated clay and decreased with talc. At equal gloss levels,
opacity improvement was quite significant for delaminated clay over
talc
[149] Clay+GCC + Talc Pigment blend Replacing clay # 1 with talk in a matte coated art paper, increased
print gloss but decreased paper gloss.
[150] Narrow PSD Pigment type and binder Higher latex decreased gloss, whiteness and brightness of a coated
GCC, PCC, Fine content board.
GCC However, print gloss increased with latex content. Gloss, whiteness

Session 2: Fibre Suspensions and Forming

and brightness decreased as: PCC>Narrow PSD GCC >Fine GCC
 [151] GCC, PCC, Kaolin Pigment blend, calendering Brightness increased by increasing the coating pore size while print
load gloss and delta gloss require a balance between pore size and pore
volume.
[152] GCC, Kaolin Latex type, Calendering load Higher calendering temperature had a higher paper and print gloss.
and temperature Extended nip calendering improved gloss compared to soft nip. Gloss
of paper and print varied as: low Tg Styrene-Butadiene (SB) latex:
High paper gloss & good print gloss, high Tg SB: low paper gloss,
Styrene Acrylate (SA) latex: good paper and print gloss, Poly(vinyl
acetate) (PVAc): Low paper and print gloss.
[153] GCC + Kaolin Calendering conditions Calendering at lower temperature and higher line loads decreased
light scattering of ability of coating and reduced gloss.

14th Fundamental Research Symposium, Oxford, September 2009

331
Optical Properties of Paper: Theory and Practice
Ramin Farnood

Figure 25. SEM micrographs of a) coarse clay with mean particle size: 2.16 μm;, b)
fine clay with mean particle size: 0.26 μm, and c) ground calcium carbonate with mean
particle size: : 0.77 μm [154].

Figure 26. Brightness of common papermaking fillers. 1) European clay, 2) US clay,

3) US delaminated clay, 4) US calcined clay, 5) European talc, 6) Chinese talc, 7)
European chalk, 8) GCC, 9) PCC, and 10) TiO2 [130].

characteristics and pore structure of the coating/paper. Table 7 and Table 8

provide typical values of the light scattering coefficient and the refractive
index for common fillers used in the paper industry.
KM theory may be used to estimate the scattering efficiency – and hence

332 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Table 7. Light scattering coefficient of common fillers and virgin pulps [155].

Material Light Material Light

Scattering Scattering
Coeff. (m2/kg) Coeff. (m2/kg)

Kaolin hydrous 70–120 Titanium dioxide 450–650

GCC 140–190 Silicate/silica 280–340
PCC 210–270 Chemical pulp 20–45
Calcined clay 200–300 Mechanical pulp 50–70

Table 8. Physical characteristics of coating pigments [156, 157].

Pigment Size (μm) Shape Density Refractive ISO

(kg/m3) Index Brightness

Kaolin clay 0.3–5 Hexagonal 2.58 1.56 80–90

platy
GCC 0.7–2 Cubic, 2.7 1.56–1.65 87–97
prismatic, platy
PCC 0.1–1.0 Usually rod- 2.7 1.59 96–99
like
Talcum 0.3–5 Platy 2.7 1.57 85–90
Gypsum 0.2–2 Roundish 2.3 1.52 92–94
Plastic pigments
-Solid 0.1–0.5 Spherical 1.05 1.59 93–94
-Hollow 0.4–1.0 Spherical 0.6–0.9 1.59 93–94
Calcined kaolin 0.7 (median) Aggregated 2.69 1.56 93
plates
Titanium dioxide
-Anatase 0.2–0.5 Rod-like 3.9 2.55 98–99
-Rutile 0.2–0.5 Roundish 4.2 2.70 98–99
Aluminum 0.2–2 Platy 2.42 1.57 98–100
trihydrate

opacity and brightness – of filled and coated papers. In the case of filled
papers, the additivity rule may be applied to determine the effective scattering
and absorption coefficients [158]:

14th Fundamental Research Symposium, Oxford, September 2009 333

Ramin Farnood

For coated papers, the layered model described by Equations (16) to (18) can
be used. In practice, the reflectance of a single sheet of coated paper, Ro,12,
may be obtained without any transmittance measurements using:

Based on the above approach, the reflectance and scattering of the coating
layer can be estimated based on the measurement of reflectance values of
coated papers with two different backings [139, 159]:

where :

Here, R∞,c is the reflectance of an infinitely thick coating layer, R1 and R2 are
the reflectances of a single sheet of coated paper with different backings
(R1 < R2), Rg1 is the reflectance of substrate 1 (black cavity), and Rg2 is the
reflectance of substrate 2 (a sufficiently large number of sheets of base
paper). Using this method, Hamada et al. found a linear relationship between
the light scattering coefficient and brightness of coating layer (i.e. R∞,c).
Experimental studies have shown that the scattering coefficient and reflect-
ance of coating layer determined in this way depend on the binder content,
degree of calendering, and the base sheet properties. In particular, a
non-absorptive base sheet is known to have a substantially lower scattering
coefficient [159].
As discussed earlier, KM theory is an approximation of the more general
radiative transfer model, and the absorption and scattering coefficients in
KM theory, K and S, are not material properties. Instead, these coefficients
should be regarded as model parameters that fit the experimental observa-
tions for a given set of materials. However, it would be advantageous to
predict the scattering efficiency of filled papers and coating layers based on
characteristic of paper constituents, i.e. fibres, fillers, and pigments. This
could be addressed by resorting to theories that describe light-particle
interaction.

334 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

The typical size of papermaking pigments is of the order of wavelength of

light; hence, Mie theory may be applied to study the interaction of light with
pigment particles. Based on Mie theory, scattering and absorption of a
particle depend on particle size, dielectric properties, and the wavelength of
incident light [29, 30]. This theory has been originally developed for light
interaction with a single particle; however, it may be also used for loosely
packed particle systems [160]. Mie theory has been applied in combination
with the KM model to estimate the optical properties of coatings, paints and
polymer films. For dilute particle systems, where inter-particle interactions
may be ignored, KM absorption and scattering coefficients may be estimated
based on the Mie’s absorption and scattering cross sections of a single pig-
ment particle and by using Equations (30) and (31). Although paints and
coatings are densely packed systems, this approximation has been often
used to examine the relative changes of K and S with variations in pigment
characteristics [161].
To account for particle interactions in densely packed systems, various
corrections have been proposed to improve the predictions of Mie theory
[162–166]. Based on Van de Hulst’s approach, Brinkworth assumed that the
scattering cross section of a suspension of monodispersed spherical particles
can be expressed in terms of the volume concentration of particles Vp, par-
ticle diameter D, and the scattering efficiency of a single particle Qsca [166]:

where g is the asymmetry factor that is defined as the average cosine of the
scattering angle (see Equation (21)). If the incident energy were totally for-
ward scattered then g = +1, but if it were fully backscattered, then g = −1.
Scattering efficiency is the ratio of the scattering cross section and the
geometrical cross section of the particle (i.e. 4σs /πD2). For cases where the
refractive index of particle is close to that of medium, the scattering efficiency
is given by [166]:

where εr is the ratio of the refractive index of the particle to that of the
surroundings.
Using an approach similar to Brinkworth and by implementing Gate’s
equation for scattering cross section (Equation (20)), Ross proposed the fol-
lowing model that takes into account the volume concentration of particles,
anisotropy of the scattered field, and illumination conditions [163]:

14th Fundamental Research Symposium, Oxford, September 2009 335

Ramin Farnood

The parameter c in the above equation is a constant that depends on the

illumination geometry and surface reflection effects. Assuming that there is
no boundary surface to complicate the path of light by refraction and reflec-
tion, i.e. pigment particles are dispersed in air, two cases were considered:
diffuse illumination for which c = 1, and normal collimated illumination for
which c = 0.8. Note that S has the unit of inverse length. Using Ross’ model,
Borch and Lepoutre estimated the variations in the light scattering coef-
ficients of coatings containing spherical pigments as a function of pigment
size and refractive index [167].
Strictly speaking, neither KM nor Mie theory are suitable for the predic-
tion of optical properties of paper coatings because: 1) pigment size is close
to the optical wavelengths of light, and 2) packing density of typical coatings
is above 50% while Mie theory and RTE are only applicable for low packing
densities (<10%). It is well-known that a high concentration of pigment par-
ticles results in inter-particle interactions and reduces the overall scattering
efficiency [164]. To address this problem, various analytical and numerical
methods have been suggested that do not suffer from the limitations of Mie
and KM theories. In particular, an elegant extension of Mie theory has been
suggested by Flesia and Schwendimann to account for multiple scattering
effects [31]. Others have resorted to numerical solutions to study the crowding
and particle-particle interaction effects at higher pigment volume concentra-
tions. Figure 27 shows one such simulation that illustrates the near field
scattering for two adjacent titanium dioxide particles in a polymer matrix
based on the finite element solution to Maxwell’s equation [168]. The far field
scattering coefficient of these two particles reduced from ∼24 μm−1 to about
22 μm−1 as particle-particle separation decreased from 500 nm to zero.
More recently, Penttilä et al. applied the discrete dipole approximation
(DDA) to predict the reflectance and transmittance of densely packed coat-
ing layers [169]. DDA is an approximate solution to the Maxwell equations
for studying the scattering and absorption of radiation by particles with arbi-
trary geometry. It is based on partitioning the medium into small polarized
cells called dipoles that acquire dipole moments in response to the local
electric field [170]. Using DDA it is possible to account for changes in scatter-
ing efficiency of the coating layer due to variations in the packing density and
pigment characteristics; such as size, shape, and refractive index. This is a
useful approach that helps to design better pigments with enhanced optical
performance by optimizing size, geometry, and refractive index. Based on this
method, Penttilä et al. studied the scattering coefficient, reflectivity and

336 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 27. Simulated near field scattering for two anisotropic rutile titanium dioxide
particles in a resin with n = 1.514. Particle-particle distances are 0.2 μm (top) and 0
(bottom), incident light is normal to the surface of paper and wavelength of light is
560 nm [168].

transmittance of model coating layers composed of starch acetate pigments

as a function of packing density and pigment particle size (Figure 28.a). They
found that at a refractive index of 1.47, the scattering coefficient decreased
from ∼10 μm−1 for a 0.2 μm pigment to a minimum value of 6.5 μm−1 for a
pigment size of about 0.7–0.8 μm (Figure 28.b). In contrast, at n = 1.67 (not
shown here), the light scattering coefficient increased slightly before decreas-
ing at above a pigment size of 0.5 μm. This approach is currently being
extended to predict the effects of pigment modification on the optical proper-
ties of coatings [171].
In practice, the reduction in the scattering coefficient caused by pigment
particle interactions is detrimental to the optical properties of coating layers.
Figure 29 illustrates the effect of volume concentration of titanium dioxide
on the KM scattering coefficient. It is evident that by increasing TiO2 concen-
tration beyond ∼25 v/v%, there is little gain in the scattering ability of the
coating [164]. Such effects, however, may be negated by the introduction of a
co-pigment [172, 173].

14th Fundamental Research Symposium, Oxford, September 2009 337

Ramin Farnood

Figure 28. (a) A model cylindrical coating structure containing 1400 pigment
particles with a packing density of ∼50%, and (b) the predicted light scattering
coefficient of the model coating layer [169].

An alternative method to enhance light scattering of coating layers is to

introduce microvoids into the coating’s structure. It has long been known that
microvoids are able to increase the “hiding” power of coatings and paints.
Ross reported that the theoretical light scattering coefficient of micro-bubbles
in a latex resin is about 12% of that of rutile titanium dioxide (Figure 30)
[163]. In an experimental study concerning latex coatings, Pierce et al. found
that microvoids have a scattering coefficient of about 0.064 to 0.070 μm−1
[174]. Furthermore, they reported that it is possible to achieve the same over-
all scattering coefficient by introducing microvoids and significantly reducing

338 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Figure 29. Effect of pigment volume concentration on the KM scattering coefficient

for titanium dioxide [164].

the amount of TiO2. Most recently, using the numerical model described
earlier, Penttilä et al. investigated starch acetate coating layers containing
microvoids and found that the scattering coefficient of these layers decreased
from 10 μm−1 to about 7 μm−1 as the void diameter increased from 200 nm to
800 nm (Figure 28.b). These values, however, are significantly higher than the
experimental values of Pierce et al. and theoretical predictions of Ross.
One way to introduce “voids” in coating structure is through the addition
of hollow pigments. Hollow plastic pigments are commonly used to improve
the optical performance of the coating layer, the coated paper’s reflectance
and its gloss [139, 175, 176]. Recently, Enomae and Tsujino demonstrated
that similar improvements may be realized by introducing hollow inorganic
pigments into the coating formulations. They synthesized hollow calcium
carbonate pigments with a primary particle size of 50 nm and examined their
performance both as filler and as coating pigment [177]. They observed a
positive effect on the coating brightness and light scattering coefficient.

14th Fundamental Research Symposium, Oxford, September 2009 339

Ramin Farnood

Figure 30. Theoretical KM scattering coefficients of uniform spherical rutile

pigments in resin (a) and spherical microvoids in resin (b) as a function of sphere
diameter [161].

CONCLUDING REMARKS

Significant progress has been made in our understanding of the interaction of

light with paper and its effect on paper’s appearance. Advancements in com-
puters, digital imaging, and laser light sources have resulted in the develop-
ment of a variety of modern instrumental assemblies to examine the optical
properties of paper in novel ways, at speeds that were not possible a few years

340 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

ago. These advancements have enabled us to map optical properties of paper

at a high resolution over relatively large areas. This is advantageous since our
perception of paper quality is not only affected by the average values of
optical characteristics of paper but also by their local variations at relatively
small scales, i.e. at the scale of fibre length and below. However, despite these
technological advancements, typical standard testing methods only provide
“average” measurements of optical properties over relatively large areas.
Development of standardized methods that take advantage of these
advancements would be beneficial not only to practitioners but also to
researchers in this field.
Progress in our theoretical understanding of propagation and surface scat-
tering of waves in random media driven by remote sensing, material testing,
and medical imaging applications, has provided us with new insights into the
relationship between paper’s structure and its optical properties. In spite of
this progress, our quest for developing physical models to predict the optical
properties of paper based on its structure and furnish characteristics is still
ongoing. Numerical simulations of light-paper interaction, similar to those
discussed in this article, offer a promising possibility to achieve this goal.
However, further work is required to improve these models while taking into
account the effects of base sheet and coating structure, as well as fibre and
pigment characteristics.
The surface structure of paper plays an important role in terms of its
optical properties, especially gloss. Today, confocal laser scanning micro-
scopy, atomic force microscopy, and a variety of surface topography meas-
urement techniques are available to characterize paper surface structure at
high resolutions. However, we still lack a fundamental understanding of
effect of paper furnish, forming and finishing conditions on paper surface
statistics. This is further complicated by the fact that rms roughness and
correlation length of surface heights depend both on the sample size and
measurement resolution. Further fundamental work is required to better
understand these effects.
With escalating cost and reduced availability of fibre resources, there is an
increasing desire to increase the amount of filler and/or reduce grammage of
paper, while maintaining adequate optical properties for printing and writing
papers. On the other hand, rapid growth of digital printing technology has
introduced new demands on the paper performance and characteristics.
These competing factors stress the need for a better fundamental understand-
ing of the relationships among structure, furnish, and optical properties of
paper.
Over the next decade, rapid reduction in the consumption of printing and
writing papers in mature market regions, such as North America, combined

14th Fundamental Research Symposium, Oxford, September 2009 341

Ramin Farnood

with increased environmental awareness, will likely transform the face of pulp
and paper industry as we know it. The emergence of novel green paper-based
products for new applications that take advantage of wood fibres as a bio-
degradable and renewable resource is imminent. Inevitably, these new prod-
ucts will impose different demands and new challenges in terms of the optical
properties of paper, but they also present us with an opportunity to further
improve our fundamental understanding of these properties and to
contribute to the advancement of optical science and technology.

ACKNOWLEDGEMENTS

The author wishes to thank Mr. Duoyin Lu for his help with the literature
search and Mr. Peter Angelo for his editing work on this document.

REFERENCES

1. PAPTAC Standard E.1. Brightness of Pulp, Paper, and Paperboard. In Pulp and
Paper Technical Association of Canada Standard Testing Methods, 1990.
2. J.A. Bristow. What Is ISO Brightness? TAPPI J. 77(5):174–178, 1994.
3. B.D. Jordan and M.A. O’neill. The Whiteness of Paper- Caolorimetry and
Visual Ranking. TAPPI J. 74(5) 1991.
4. J.A. Bristow and C. Karipidis. ISO Brightness of Fluorescent Papers and Indoor
Whiteness-Proposal for Illuminant. TAPPI J. 82(1):183–193, 1999.
5. S.J. Popson, D.D. Malthouse and R.C. Robertson. Applying Brightness, White-
ness, and Color Measurements to Color Removal. Tappi J. 80(9):137–147, 1997.
6. PAPTAC Standard E.2. Opacity of Paper. In Pulp and Paper Technical Associ-
ation of Canada Standard Methods, 1993.
7. R. Silvennoinen, K.-E. Peiponen and K. Myller. Traditional Glossmeters. Ch.7
in Specular Gloss, pp. 267–314, Elsevier, Amsterdam, 2008b.
8. J. Borch. Optical and Appearance Properties. Ch.4 in Handbook of Physical
Testing of Paper, vol. 2, (ed. R.E. Mark, M.B., Lyne and J. Borch), pp. 95–148,
Marcel Dekker, New York, 2002.
9. M. Leskelä. Optical Properties. Ch. 4 in Paper Physics, (ed. K. Niskanen),
pp. 117–137, Finish Paper Engineers’ Association and Tappi Press, Helsinki,
1998.
10. L.T. Sharpe, A. Stockman, W. Jagla and H. Jagle. A Luminous Efficiency
Function, V*(), for Daylight Adaptation. J. Vision 5(11):577–585, 2005.
11. A. Stockman and L.T. Sharpe. The Spectral Sensitivities of the Middle- and
Long-Wavelength-Sensitive Cones Derived from Measurements in Observers of
Known Genotype. Vision Res. 40(13):1711–1737, 2000.
12. M. Baneshi, S. Maruyama, H. Nakai and A. Komiya. A New Approach to

342 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Optimizing Pigmented Coatings Considering Both Thermal and Aesthetic

Effects. J. Quant. Spectrosc. Radiat. Transfer 110(3):192–204, 2009.
13. X. Hu and K.W. Houser. Spectral and Electrical Performance of Screw-Based
Dimmable Cmpact Fluorescent Lamps. Lighting Res. Technol. 35(4):331–342,
2003.
14. P. Hansson. Simulation of Small Scale Print Gloss Variations Based on
Topographic Data. NPPRJ 19(1):112–115, 2004.
15. M. Lindstrand. An Angularly and Spatially Resolved Reflectometry for a
Perceptually Adequate Characterization of Gloss. JIST 49(1):71–84, 2005.
16. J.S. Arney, P.G. Anderson, G. Franz and W. Pfeister. Color Properties of Specular
Reflections. JIST 50(3):228–232, 2006.
17. J.S. Arney and D. Nilosek. Analysis of Print Gloss with a Calibrated Microgonio-
photometer. JIST 51(6):509–513, 2007.
18. G. Bernard, R. Farnood and N. Yan. Customized Setup to Assess Microgloss
Uniformity of Paper. Pulp & Paper Can. 106(9):T190–193, 2005.
19. M.A. Macgregor and P.-A. Johansson. Submillimeter Gloss Variations in Coated
Paper: Part I. The Gloss Imaging Equipment and Analytical Techniques. Tappi J.
73(12):161–168, 1990.
20. N.J. Elton and J.S. Preston. Polarized Light Reflectometry for Studies of Paper
Coating Structure. Part I. Method and Instrumentation. Tappi J. 5(7):8–16,
2006.
21. Y.-J. Sung and D.S. Keller. Evaluation of Gloss Variation with a Novel Method. J.
Korea Tappi 34(2):73–83, 2002.
22. M. Sorjonen, A. Jääskeläinen, K.-E. Peiponen and R. Silvennoinen. On the
Assessment of the Surface Quality of Black Print Paper by Use of a Diffractive
Optical-Element-Based Sensor. Meas. Sci. Technol. 11(6):N85–N88, 2000.
23. J. Palviainen, M. Sorjonen, R. Silvennoinen and K.-E. Peiponnen. Optical Sens-
ing of Colour Print on Paper by Diffractive Optical Element. Meas. Sci. Technol.
13(4):N31–N37, 2002.
24. R. Silvennoinen, et al. Diffractive Element in Optical Inspection of Paper. Opt.
Eng. 37(5):1482–1487, 1998.
25. M. Juuti, V. Kalima, T.T. Pakkanen and K.-E. Peiponen. A Novel Method to
Measure and Analyse Delta Gloss by Diffractive Glossmeter. Meas. Sci. Technol.
18(5):L5–L8, 2007.
26. M. Jutti, H. Koivula, M. Toivakka and K.-E. Peiponen. A Diffractive Gloss
Meter for Local Gloss Measurements of Papers and Prints. Tappi J. 7(4):27–32,
2008.
27. A. Apostol, D. Haefner and A. Dogariu. Near-Field Characterization of Effective
Optical Interfaces. Physical Rev. E. 74(6):066603, 2006.
28. D. Ma, R.D. Carter, D. Haefner and A. Dogariu. Simultaneous Characterization
of Coated Paper Topography and Optical Contrast by near-Field Scanning
Optical Microscopy (Nsom). NPPRJ 23(4):438–443, 2008.
29. A. Ishimaru. Wave Propagation and Scattering in Random Media, IEEE, New
York 1997.

14th Fundamental Research Symposium, Oxford, September 2009 343

Ramin Farnood

30. G. Mie. Contributions to the Optics of Turbid Media, Particularly of Colloidal

Metal Solutions. Annal. Phys. 25(3):377–445, 1908.
31. C. Flesia and P. Schwendimann. Analytical Multiple Scattering Extension of
the Mie Theory – Part I: General Discussion. Appl. Phys. B 56(3):157–163,
1993.
32. J.A. Ogilvy. Theory of Wave Scattering from Random Rough Surfaces, Bristol:
Adam Hilger 1991.
33. A.K. Fung. Microwave Scattering and Emission Models and Their Applications,
Artech House, Boston 1994.
34. O.M. Johannessen, V.Yu. Alexandrov, I.Ye. Frolov, S. Sandven, L.H. Pettersson
and L.P. Bobylev. Remote Sensing of Sea Ice in the Northern Sea Route, Springer,
Berlin 2007.
35. P. Beckmann and A. Spizzichino. The Scattering of Electromagnetic Waves from
Rough Surfaces, Oxford: Pergamon, 1963.
36. G. Berginc. Small-Slope Approximation Method: A Further Study of Vector
Wave Scattering from Two-Dimensional Surfaces and Comparison with
Experimental Data. PIER 37(2002):251–287, 2002.
37. T.M. Elfouhaily and C.-A. Guérin. A Critical Survey of Approximate Scattering
Wave Theories from Random Rough Surfaces. Waves in Random Media
14(4):R1–R40, 2004.
38. C.-A. Guérin, G. Soriano and T. Elfouhaily. Weighted Curvature Approximation:
Numerical Tests for 2 D Dielectric Surfaces. Waves in Random Media
14(3):349–363, 2004.
39. K.F. Warnick and W.C. Chew. Numerical Simulation Methods for Rough Surface
Scattering. Waves Random Media 11(1):R1–R30, 2001.
40. T.R. Lettieri, E. Marx, J.-F. Song and T.V. Vorburger. Light Scattering from
Glossy Coatings on Paper. Appl. Opt. 30(30):4439–4447, 1991.
41. N. Dechamps and R. Farnood. Modeling of Paper Gloss Using Geometrical
Optics Approximation. Pulp & Paper Centre Internal Report. University of
Toronto, 2006.
42. P.S. Mudgett and L.W. Richards. Multiple Scattering Calculations for Technol-
ogy. Appl. Opt. 10(7):1485–1502, 1971.
43. P. Kubelka. New Contributions to the Optics of Intensely Light-Scattering
Materials. Part I. J. Opt. Soc. Am. 38(5):448–457, 1948.
44. L.F. Gate. Comparison of the Photon Diffusion Model and Kubelka-Munk
Equation with the Exact Solution of the Radiative Transport Equation. Appl.
Opt. 5(4):837–846, 1974.
45. J.-F. Bloch and R. Sève. About the Theoretical Aspect of Multiple Light Scatter-
ing: Silvy’s Theory. Color Res. Appl. 28(3):227–228, 2003.
46. J. Silvy. Contribution À L’étude De La Diffusion Multiple De La Lumière Dans
Le Cadre De L’approximation De Diffusion. Rev. Opt., Theor. Instrum.
40(10):495–517, 1961.
47. S.L. Jacques. Reflectance Spectroscopy with Optical Fibre Devices, and Transcu-
taneous Bilirubinometers. In Biomedical Optical Instrumentation and

344 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

Laser-Assisted Biotechnology, (ed. S.M. A.M. Verga Scheggi, A.N. Chester and
R. Pratesi), pp. 83–94, Kluwer Academic, Boston, 1996.
48. S.N. Thennadil. Relationship between the Kubelka-Munk Scattering and
Radiative Transfer Coefficients. J. Opt. Soc. Am. A 25(7):1480–1485, 2008.
49. B.J. Brinkworth. Interpretation of Kubelka-Munk Coefficient in Reflection
Theory. Appl. Opt. 11(6):1434–1435, 1972b.
50. W.M. Star, J.P.A. Marijnissen and M.J.C. Van Gemert. Light Dosimetry in
Optical Phantoms and in Tissues: I. Multiple Flux and Transport Theory. Phys.
Med. Biol. 33(4):437–454, 1988.
51. A. Alince, J. Porubská and T.G.M. Van De Van. Light Scattering and Microscopy
in Paper. JPPS 28(3):93–98, 2002.
52. L.K. Lehtonen and T.J. Dyer. Light-Scattering Coefficient as a Measure of
Specific Surface Area in Mechanical Pulp Laboratory Sheets. Paperi ja puu
87(8):517–524, 2005.
53. W.J. Foote. Investigation of Fundamental Scattering and Absorption Coefficients
of Dyed Handsheets. Paper Trade J. 109(25):31–38, 1939.
54. L. Nordman, P. Aaltonen and T. Makkonen. Relationship between Mechanical
and Optical Properties of Paper Affected by Web Consolidation. In Transactions
of the Symposium on Consolidation of the Paper Web, pp. 909–927. Cambridge:
Tech. Section British Paper and Board Makers’ Assoc., London, 1996.
55. A.A. Koukoulas and B.D. Jordan. Effect of Strong Absorption on the Kubelka-
Munk Scattering Coefficient. JPPS 23(5):J224–232, 1997.
56. P. Edstrom. Comparison of the Dort2002 Radiative Transfer Solution Method
and the Kubelka-Munk Model. NPPRJ 19(3):397–403, 2004.
57. H. Granberg, M. Rundlöf and L. Mattsson. Influence of Surface-Induced Non-
uniform Reflectance on the Diffuse Reflectance Factor. Part I: Model Predictions.
JPPS 29(8):247–253, 2003.
58. L. Yang and B. Kruse. Revised Kubelka-Munk Theory: I. Theory and Applica-
tion. J. Opt. Soc. Am. A 21(10):1934–1942, 2004.
59. A. Mandelis and J.P. Grossman. Perturbation Theoretical Approach to the
Generalized Kubleka-Munk Problem in Nonhomogeneous Optical Media. Appl.
Spectrosc. 46(5):737–745, 1992.
60. L.F. Gate. The Determination of Light Absorption in Diffusing Materials by a
Photon Diffusion Model. J. Phys. D: Appl. Phys. 4(7):1049–1056, 1971.
61. B. Philips-Invernizzi, D. Dupont and C. Caze. Bibliographical Review for Reflect-
ance of Diffusing Media. Opt. Eng. 40(5):1082–1092, 2001.
62. E. Ganz. Whiteness: Photometric Specification and Colorimetric Evaluation.
Appl. Opt. 15(9):2039–2058, 1976.
63. PAPTAC Standard E.5. Pulp, Paper and Paperboard- Colour Measurement with
a Diffuse/Zero Geometry Tristimulus Reflectometer. In Pulp and Paper Technical
Association of Canada Standard Methods, 1993.
64. J.S. Bonham. The Appearance of ‘White’ Papers. Appita J. 50(6):446–451, 2006.
65. H. Liu, S. Yang and Y. Ni. Using Dyes for Improving the Optical Properties of
High Yield Pulps. Pulp & Paper Canada 108(10):T157–160, 2007.

14th Fundamental Research Symposium, Oxford, September 2009 345

Ramin Farnood

66. E. Milanova and B.B. Sithole. A Simple Method for Estimation of Newsprint
Dyes in Effluents and Their Migration from Paper Samples. Tappi J. 80(5):121–
128, 1997.
67. L. Cabos, M. Berrada and J. Paris. Effect of Furnish Composition on the Color of
a Fine Three-Ply Paperboard. Tappi J. 83(10):1–16, 2000.
68. B. Aksoy, M.K. Joyce and P.D. Fleming. Comparative Study of Brightness/
Whiteness Using Various Analytical Methods on Caoted Papers Containg Color-
ants. In TAPPI Spring Technical Conference and Exhibit, pp. 371–385, Chicago,
IL. Technical Association of Pulp and Paper Industry, Atlanta, 2003.
69. M.O.A. Mäkinen, T. Jääkelainen and J. Parkkinen. Improving Optical Properties
of Printing Papers with Dyes: A Theoretical Study. NPPRJ 22(2):236–243,
2007.
70. A.V. Makarenko and I.A. Shaykevich. Dependence of the Whiteness of Paper on
Surface Roughness and Illumination Conditions. Color Res. Appl. 25(3):170–175,
2000.
71. W.J. Auhorn. Chemical Additives. Ch. 3 in Handbook of Paper and Board, (ed. H.
Holik), pp. 62–149, Wiley-VCH Verlag, 2006.
72. O. Jokinen, R. Baak, G. Traser and P. Rohringer. Optical Brighteners for High
White Coated Paper. Wochenblatt für Papierfabrikation 128(9):590–593, 2000.
73. J.-E.P. Nordström, J.P. Nordlund and J.P.L. Grön. Effect of Optical Brightening
Agents Degradation Due to Heat and Moisture- Determining Optical Mottle?
Paperi Ja Puu 80(6):449–458, 1998.
74. M. Cieplinski and E. Drzewinska. The Effect of Fluorescent Whitening Agents
and Tinting Dyes on Permanence of Paper Colour. Paper Technol. 48(5):37–40,
2007.
75. D. Ma, R.D. Carter, D. Haefner and A. Dogariu. The Influence of Fine Kaolin
and Ground Calcium Carbonates on the Efficiency and Distribution of Fluor-
escence Whitening Agents (Fwa) in Paper Coating. NPPRJ 23(3):327–332, 2008.
76. J.S. Bonham. Fluorescence and Kubelka-Munk Theory. Color Res. Appl.
11(3):223–230, 1985.
77. Y.Y. Kulikova, M.K. Schaepe and W.B. Robbins. Assessment of the Influence of
Paper Formation on the Fluorescence Contribution to Brightness. Color Res.
Appl. 20(2):87–92, 1995.
78. B.T. Hallam and A.G. Hiorns. Developing Optical Efficiency through Optimised
Coating Structure. In Tappi Advanced Coating Fundamental Symposium Proceed-
ings, pp. 330–339, Turku. 2008.
79. Z. Yuan, J. Schmidt, C. Heitner and X. Zou. Coating Improves the Brightness
Stability of Wood-Free Coated Papers Containing High-Yield Pulp. Tappi J.
5(1):9–13, 2006.
80. W.J. Hillend. Opacity Problems in Printing Papers: Kubelka-Munk Theory Gives
Good, Quick Answers. Tappi J. 49(7):41A–47A, 1966.
81. M.O.A. Mäkinen, J. Parkkinen and T. Jääskeläinen. Generalized Opacity Basedx
on Cielab Colour Coordinates and the CIE94 Colour Difference Formula. JPPS
31(1):61–67, 2005.

346 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

82. ISO2471. Determination of Opacity (Paper Backing)- Diffuse Reflectance

Method. Geneva, Switzerland, 1998.
83. A. Dogariu and G.D. Boreman. Facet Model for Photon-Flux Transmission
through Rough Dielectric Interfaces. Opt. Let. 21(10):701–703, 1996.
84. M.E. Nadal and E.A. Thompson. New Primary Standard for Specular Gloss
Measurements. J. Coatings Technol. 72(911):61–66, 2000.
85. L.F. Gate, W. Windle and M. Hine. The Relationship between Gloss and Surface
Microtexture of Coatings. Tappi J. 56(3):61–65, 1973.
86. P. Oittinen. The Surface Structure of Coated Paper and the Formation of Gloss.
In The Role of Fundamental Research in Paper Making. Transactions of the 7th
Fundamental Research Symposium, (ed. J. Brander), pp. 635–654, Cambridge.
Mechanical Engineering Publications Ltd., 1981.
87. H. Lipschitz, M. Bridger and G. Derman. Relationships between Topography and
Gloss. TAPPI J. 73(10):237–245, 1990.
88. M.A. Macgregor, P.-Å.. Johansson and M.-C. Béland. Measurement of Small-
Scale Gloss Variation in Printed Paper: Topography Explains Much of the Vari-
ation for One Paper. In International Printing and Graphic Arts Conference, pp.
33–43, Haplifax. 1994.
89. M.-C. Béland, S. Lindberg and P.-Å. Johansson. Optical Measurement and Per-
ception of Gloss Quality of Printed Matte-Coated Paper. JPPS 26(3):120–123,
2000.
90. G. Chinga. Detailed Characterization of Paper Surface Structure for Gloss
Assessment. JPPS 30(8):222–227, 2004.
91. G. Chinga, P. O. Johnsen, R. Dougherty, E.L. Berli and J. Walter. Quantification
of the 3D Microstructure of SC Surfaces. J. Microsc. 227(3):254–265, 2007.
92. N.J. Elton and J.S. Preston. Polarized Light Reflectometry for Studies of Paper
Coating Structure. Part II. Application to Coating Structure, Gloss and Porosity.
Tappi J. 5(8):10–16, 2006.
93. R. Alexander-Katz and R.G. Barrera. Surface Correlation Effects on Gloss. J.
Polymer Sci. B 36(8):1321–1334, 1998.
94. E. Caner, R. Farnood and N. Yan. Relationship between Gloss and Surface
Texture of Coated Papers. TAPPI J. 7(4):19–26, 2008.
95. P. Vernhes. Relevant Paper Properties for Digital Printing, Ph.D. Thesis, Ecole
internationale du papier, de la communication imprimée et des biomatériaux,
Grenoble, 2008.
96. P. Vernhes, J.F. Bloch, C. Mercier, A. Blayo and B. Pineaux. Statistical Analysis of
Paper Surface Microstructure: A Multi-Scale Approach. Appl. Surf. Sci.
254(22):7431–7437, 2008.
97. P. Vernhes, J.-F. Bloch, A. Blayo and B. Pineaux. Gloss Optical Elementary
Representative Surface. Appl. Opt. 47(29):5429–5438, 2008.
98. K. Green, L. Lamberg and K. Lumme. Stochastic Modeling of Paper Structure
and Monte Carlo Simulation of Light Scattering. Appl. Opt. 39(25):4669–4683,
2000.
99. R. Farnood and N. Dechamps. Use of Recent Electromagnetic Theories to

14th Fundamental Research Symposium, Oxford, September 2009 347

Ramin Farnood

Predict Paper Gloss and Microgloss Nonuniformity. Pulp & Paper Centre
Internal Report, University of Toronto, 2005.
100. J.S. Arney, L. Ye and S. Banach. Interpretation of Gloss Meter Measurements.
JIST 50(6):567–571, 2006.
101. C. Bourlier and G. Berginc. Multiple Scattering in the High-Frequency Limit
with Second Order Shadowing Function from 2D Anisotropic Rough Dielectric
Surfaces: I. Theoretical Study. Waves in Random Media 14(3):229–252, 2004.
102. V.I. Tatarskii and W.V. Tatarskii. Statistical Description of Rough-Surface Scat-
tering Using the Quasi-Small-Slope Approximation for Random Surfaces with a
Gaussian Multivariate Probability Distribution. Waves Random Complex Media
4(2):191–214, 1994.
103. J. Järnström, L. Sinervo, M. Toivakka and J. Peltonen. Topography and Gloss of
Precipitated Calcium Carbonate Coating Layers on a Model Substrate. Tappi J.
6(5):23–31, 2007.
104. J.A. Van Den Akker. Optical Properties of Paper. In The Structure and Physical
Properties of Paper, (ed. H.F. Rance), pp. 127–174, Elsevier, Amsterdam, 1982.
105. R.M. Leekley, C.W. Denzer and R.F. Tyler. Measurement of Surface Reflection
from Paper and Prints. Tappi J. 53(4):615–621, 1970.
106. J.S. Arney, L. Ye, J. Wible and T. Oswald. Analysis of Paper Gloss. JPPS
32(1):19–23, 2006.
107. T. Takahashi and R. Farnood. Contribution of Surface and Bulk Scattering to
the Gloss of Coated Papers. In Tappi Coating Conference, Salt Lake City. TAPPI
Press, Atlanta, 2009.
108. T. Glatter and D.W. Bousefield. Print Gloss Development on a Model Substrate.
Tappi J. 80(7):125–132, 1997.
109. J.S. Preston, G.C. Allen, J.J. Elton, J.C. Husband, J.S. Dalton and P.J. Heard.
Investigation into the Distribution of Ink Components Throughout Printed
Coated Paper- Part1: Optical and Roughness Considerations. Colloids Surf. A
205(3):183–198, 2002.
110. M.A. Macgregor and P.-Å. Johansson. Gloss Uniformity in Coated Paper. In
Tappi Coating Conference, pp. 495–504, Montreal. Tappi Press, Atlanta, 1991.
111. G. Ström, A. Englund and M. Karathanasis. Effect of Coating Structure on
Print Gloss after Sheet-Fed Offset Printing. NPPRJ 18(1):108–115, 2003.
112. D.W. Donigian, J.N. Ishley and K.J. Wise. Coating Pore Structure and Offset
Printed Gloss. Tappi J. 80(5):163–172, 1997.
113. Y. Arai and K. Nojima. Coating Structure for Obtaining High Print Gloss. Tappi
J. 81(5):213–221, 1998.
114. J.S. Preston, D.J. Parsons, J.C. Husband and C. Nutbeen. Ink Gloss Develop-
ment Mechanisms after Printing. Part 2-the Influence of Substrate. In Advanced
Coating Fundamentals Symposium, Chicago. TAPPI Press, Atlanta, 2003.
115. N. Matsuda, T. Ichihashi and Y. Zama. Influence of Ultrafine Surface Profile to
Sheet and Print Gloss of Coated Paper. Japan Tappi J. 54(4):25–33, 2000.
116. G. Ström and M. Karathanasis. Relationship between Ink Film Topography and
Print Gloss in Offset Prints on Coated Surfaces. NPPRJ 23(2):156–163, 2008.

348 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

117. E. Caner, R. Farnood and N. Yan. Effect of the Coating Formulation on the
Microgloss Nonuniformity of Coated Papers. In PAPTAC Annual Meeting, pp.
C9–C12, Montreal. Pulp and Paper Technical Association of Canada, Montreal,
2005.
118. H. Fujiwara, C. Kaga and I. Kano. Measurement of Gloss Profile. In Tappi
Coating Conference, pp. 209–218, Boston. 1990.
119. M.-C. Béland and J.M. Bennett. Effect of Local Microroughness on the Gloss
Uniformity of Printed Paper Surface. Appl. Opt 39(6):2719–2726, 2000.
120. M.A. Macgregor. A Review of the Topographical Causes of Gloss Variation and
the Effect on Perceived Print Quality. In Proceedings of the Hansol Research
Symposium, Seoul, Korea. 2000.
121. E. Caner, R. Farnood and N. Yan. Effect of the Coating Formulation on the
Gloss Properties of Coated Papers. In International Printing and Graphic Arts
Conference, Cincinnati. TAPPI Press, Atlanta, 2006.
122. M.A. Hubbe, J.J. Pawlak and A.A. Koukoulas. Paper’s Appearance: A Review.
bioresources 3(2):627–665, 2008.
123. R. Bown. A Review of the Influence of Pigments on Papermaking and Coating.
In The Fundamentals of Papermaking Materials. Transactions of the 11th Fun-
damental Research Symposium, (ed. C.F. Baker), pp. 83–137, Cambridge. Pira
International, Leatherhead, 1997.
124. H.V. Green, T.J. Fox and A.M. Scallan. Lumen-Loaded Paper Pulp. Pulp Paper
Can. 83(7):39–43, 1982.
125. S.R. Middleton, J. Desmeules and A.M. Scallan. Lumen Loading with Calcium
Carbonate Fillers. JPPS 29(7):241–246, 2003.
126. M.L. Miller and D.C. Paliwal. The Effects of Lumen-Loading on Strength and
Optical Peroperties of Paper. JPPS 11(3):84–88, 1985.
127. P. Burri. Properties and Performance Attributes between Coated Papers in North
America, Europe, and Asia, TAPPI Press, Toronto 2005.
128. H.H. Murray. Engineered Clay Products for the Paper Industry. Appl. Clay Sci.
29(3):199–206, 2005.
129. R.W. Wygant, R.R. Iyer and D.O. Cummings. Glossing Properties of Engin-
eered Kaolins. In Tappi Coating Conference, pp. 873–852, New Orleans. TAPPI
Press, Atlanta, 1998.
130. M. Laufmann. Fillers for Paper-a Global View. In Proceedings of the PTS-
Seminar Wet End Operations-Vorgänge in der Siebpartie, pp. 1–6, Munich. 1998.
131. M. Lorusso. Base Paper Coating Interatcion: The Importance of Filler Design.
Paper Technol. 45(2):37–41, 2004.
132. H. Al-Tuaif. The Effect of Pigment Blends on the Structure and Chemistry of
Pigmented Latex Coatings. TAPPI J. 5(8):24–30, 2006.
133. A.C. Dimmick. Influence of the Average Particle Size of Aragonitic Precipitated
Calcium Carbonate on Coated Paper Properties. In Tappi Spring Technical
Conference, pp. 1–12, Chicago. TAPPI Press, Atlanta, 2003.
134. A.C. Dimmick. Effect of Sheet Moisture and Calendering Pressure on Pcc and
Gcc Coated Papers. Tappi J. 6(10):16–22, 2007.

14th Fundamental Research Symposium, Oxford, September 2009 349

Ramin Farnood

135. I. Endres and M. Tietz. Blade, Film and Curtain Coating Techniques and Their
Influence on Paper Surface Characteristics. Tappi J. 6(11):24–32, 2007.
136. J. Hocken and M. Stefan. Titanium Dioxide and Zinc Sulphide: High-Quality
Pigments for Special Papers. Paper Technol. 49(2):31–38, 2007.
137. A. Hedman, J. Gron and M. Rigdahl. Coating Layer Uniformity as Affected by
Base Paper Characteristics and Coating Method. NPPRJ 18(3):333–343, 2003.
138. E. Caner, C. Quijano-Solis, R. Farnood and N. Yan. Multivariate Analysis of
Micro-Gloss of Coated Papers. In TAPPI Coating Conference, Miami. TAPPI
Press, Atlanta, 2007.
139. H. Hamada, T. Enomae, F. Onabe and Y. Siato. Characteristics of Hollow
Sphere Plastic Pigment (Part I) -Optical Properties of Coated Paper and
Colorimetric Control in Printing. Jappan Tappi J. 58(4):56–63, 2004.
140. A. Hiorns. Calendering Response of Calcium Carbonates in Doble Coated
Woodfree Paper. In Tappi Coating Conference, pp. 127–146, Chicago. TAPPI
Press, Atlanta, 2003.
141. J.C. Husband. Interactions between Ground Calcium Carbonate Pigments and
Polymer Latices. NPPRJ 15(5):382–386, 2000.
142. L. Järnström, M. Wikström and M. Rigdahl. Porous Mineral Particles as Coat-
ing Pigments. NPPRJ 15(2):86–95, 2000.
143. H. Koyamoto and K. Okomori. Effect of Surface Properties of Base Paper on
Print Quality. In Advanced Coating Fundamentals Symposium, pp. 217–227,
Turku. TAPPI Press, Atlanta, 2006.
144. T. Lamminmäki, U. Forsstrom, M. Tienari, E. Saharinen, J. Mikkilä and J.
Koskelainen. Calendering of Coated Paper at High Speeds and Temperatures.
NPPRJ 20(3):354–360, 2005.
145. M. Larsson, G. Engström, D. Vidal and X. Zou. Impact of Calendering on
Coating Structures. NPPRJ 22(2):267–274, 2007.
146. M. Larsson, D. Vidal, G. Engstrom and X. Zou. Paper Coating Properties as
Affected by Pigment Blending and Calendering. TAPPI J. 6(8):16–22, 2007.
147. S. Lohmander. Influence of a Shape Factor of Pigment Particles on the
Rheological Properties of Coating Colours. NPPRJ 15(3):231–136, 2000.
148. P.B. Malla, R.E. Starr, T.J. Werkin and S. Devisetti. Effects of Pigment Type and
Pore Structure on Optical Properties and Rotogravure Printability of Light
Weight Coated (Lwc) Paper – a Statistically Designed Clc Coating Study with
Mixtures. In Tappi Coating Conference, pp. 407–430, San Diego. TAPPI Press,
Atlanta, 2000.
149. Y.-S. Perng, I.-C. Wang, W.-C. Yang and M.-H. Lai. Application of Talc to
Calcium Carbonate Containing Paper Coating Formulations. Taiwan J. For. Sci.
23(1):1–11, 2008.
150. J. Preston, A. Hiorns, D.J. Parsons and P. Heard. Design of Coating Structure
for Flexographic Printing. Paper Technology 49(3):27–36, 2008.
151. Y. Wang and Z. Cao. Influence of Coating Pore Structure on Optical Properties
of Low Gloss Coated Paper. In TAPPI Advanced Coating Fundamentals
Symposium, pp. 276–285, Turku. TAPPI Press, Atlanta, 2006.

350 Session 2: Fibre Suspensions and Forming

 Optical Properties of Paper: Theory and Practice

152. M. Wikström, M. Bouveng, J. Lindholm and J.-E. Teirfolk. The Influence of

Different Coating Latices on Printing and Surface Properties of Paper
Calendered at High Temperatures. Wochenblatt für Papierfabrikation
128(19):1321–1331, 2000.
153. M. Wikström, M. Bouveng, M. Rigdahl and A.G. Hiorns. Impact of High
Temperature and Soft Nip Calendering on the Pore Structure of the Coating
Layer. NPPRJ 17(2):147–152, 2000.
154. E. Caner. Effect of the Coating Formulation on the Microgloss and the
Microgloss Nonuniformity of Coated Papers, M.A.Sc. Thesis, University of
Toronto, 2006.
155. M. Laufmann. Pigments as Fillers. In Handbook of Paper and Board, (ed. H.
Holik), Wiley-VCH Verlag, 2006.
156. E. Lehtinen. Papermaking Science and Technology. Book 11. Pigment Coating
and Surface Sizing of Paper, Finnish Paper Association and TAPPI Press,
Helsinki 1999.
157. W. Kogler, T. Gliese and W.J. Auhorn. Coating Pigments. In Handbook of Paper
and Board, (ed. H. Holik), pp. 55–61, Wiley-VCH Verlag, 2006.
158. D.A. Hughes. Methods of Obtaining the Optical Properties of Papers Contain-
ing Titanium Dioxide and Mixtures of Titanium Dioxide and Other Fillers.
Tappi J. 45(2):159A–163A, 1962.
159. G.A. Hemstock. The Effect of Clays Upon the Optical Properties of Paper.
Tappi J. 45(2):158A–159A, 1962.
160. N.C. Debnath, and D.D. Kotkar. Theoretical Studies of Ligth Scattering Power.
Euro. Coat. J. 4(1998):264–269, 1998.
161. W.D. Ross. Theoretical Light-Scattering Power of Tio2 Microvoids. Ind. Eng.
Chem. Prod. Res. Dev. 13(1):45–49, 1974.
162. M. Kerker, D.D. Cooke and W.D. Ross. Pigmented Microvoid Coatings. Theor-
etical Study of Three Models. J. Paint Technol. 47(603):33–42, 1975.
163. W.D. Ross. Theoretical Computations of Light Scattering Power. Comparison
between TiO2 and Air Bubbles. J. Paint Technol. 43(563):50–66, 1971.
164. S. Fitzwater and J.W. Hook III. Dependent Scattering Theory: A New
Approach to Predicting Scattering in Paints. J. Coatings Technol. 57(721):39–47,
1985.
165. B.R. Palmer, P. Stamatakis, C.F. Bohren and G.C. Salzman. Multiple Scattering
Model for Opacifying Particles in Polymer Films. J. Coat. Technol.
61(779):41–47, 1989.
166. B.J. Brinkworth. A Scattering Efficiency Factor for Radiative Transport. J. Phys.
D: Appl. Phys. 5(3):476–480, 1972a.
167. J. Borch and P. Lepoutre. Light Reflectance of Spherical Pigments in Paper
Coatings-a Comparison with Theory. Tappi J. 61(2):45–48, 1978.
168. E.S. Thiele and R.H. French. Light-Scatering Properties of Representative,
Morphological Rutile Titania Particles Sutdied Using a Finite-Element Method.
J. Am. Ceram. Soc. 81(3):469–479, 1998.
169. A. Penttilä, K. Lumme and L. Kuuti. Light-Scattering Efficiency of Starch

14th Fundamental Research Symposium, Oxford, September 2009 351

Ramin Farnood

Acetate Pigments as a Function of Size and Packing Density. Appl. Opt.

45(15):3501–3509, 2006.
170. B.T. Draine and P.J. Flatau. Discrete-Dipole Approximation for Scattering
Calculations. J. Opt. Soc. Am. A 11(4):1491–1499, 1994.
171. A. Penttilä. Private Communication. 2009.
172. S.-Y. Chang and W.E. Scott. An Investigation into the Mechanism of Titanium-
Dioxide Extension by Calcium Carbonate. Tappi J. 71(10):128–131, 1988.
173. L. Abrams. Using Mercury Porosimetry to Characterize Coating Pore Structure
and Its Relation to Coating Optical Performance. In Tappi Coating Conference,
pp. 19–22, Nashville. TAPPI Press, Atlanta, 1996.
174. P.E. Pierce, S. Babil and J. Blasko. Hiding Power of Microvoids in Polymer
Coatings. Ind. Eng. Cem., Prod. Res. Develop. 13(1):37–41, 1974.
175. J.T. Brown. The Relationship between Hollow Sphere Pigment Geometry and
Optical Performance of Finished Paper Coatings. In Tappi Coating Conference,
pp. 113–122, Montreal. TAPPI Press, Atlanta, 1991.
176. C.P. Hemenway, J.J. Latimer and J.E. Young. Hollow-Sphere Polymer Pigment in
Paper Coating. Tappi J. 68(5):102–105, 1985.
177. T. Enomae and K. Tsujino. Application of Spherical Hollow Calcium Carbonate
Particles as Filler and Coating Pigments. Tappi J. 3(6):31–49, 2004.

352 Session 2: Fibre Suspensions and Forming

 Transcription of Discussion

OPTICAL PROPERTIES OF PAPER:

THEORY AND PRACTICE

Ramin Farnood
Department of Chemical Engineering and Applied Chemistry,
University of Toronto

Bob Pelton McMaster University

I do not have a question, I have a comment for the fraction of the paper
physics community that does this kind of modelling. There is a real
opportunity out there right now for doing modelling in this area, and it arises
from a large number of biochemists who are doing biochemical tests on
paper. In a typical biochemical test you generate a colour dye which you put
in the spectrophotometer and use Beer’s Law to get a concentration or you
generate something that fluoresces and you measure fluorescent intensity and
get a concentration. There are large numbers of people now trying to do this
on paper surfaces. What is the role of paper roughness? What happens when
these dyes diffuse down further into the paper structure? I think there is a real
opportunity for modelling to quantify the coloured and fluorescent materials
on paper surfaces. Just to finish this off and to give you an example, in the
current issue of Analytical Chemistry there is an article where they talk about
doing analysis on paper. They did their chemistry first, which generated a
colour on paper, and then they saturated the paper with vegetable oil trying to
contrast match the scattering of paper. Then they put it in a spectro-
photometer. To me this is an act of desperation. What people really want to
do is to be able to take a piece of paper with a colour generated by some
biochemical test, point a cellphone camera at it, get a picture and send it off
somewhere for external analysis. I think there is an opportunity here for
modelling.

Ramin Farnood
Thank you for your comment, Bob, I agree with you. I think there is a lot that

14th Fundamental Research Symposium, Oxford, September 2009

Discussion

we can do to improve the predictability of light–paper interaction specifically

in the examples that you have just mentioned.

Jean-Francis Bloch University of Grenoble.

First of all, thank you very much for your very interesting presentation. I
have to say I am a little bit confused when you say that the Kubelka-Munk
model answers all technological problems. I have to say there are some limita-
tions. At least the two coefficients you obtained (k and s) are only parameters
from a given model. They are absolutely not physical parameters. Just to
underline what Bob just said, if for example you increase dyes in a paper, you
will of course modify the absorption coefficient (k), but you will also modify
the scattering coefficient (s). People from time to time are surprised that the s
coefficient only comes from the model that you have chosen and is not due to
physical properties. So there are nowadays some alternatives to use, for
example the radiative transfer equation with auxiliary functions – I will not
go into the details. You also have new perspective to do this, such as to use
diffusion models, for example. So I am a bit confused when you say, please
just use Kubelka-Munk model, it works for everything. It has worked many
times because you can use it for control, for example, and that is okay. You
can use it as a first order to control the machine and it works quite well, but
when you want to go further, it has a lot of limitations. Kubelka and Munk
wrote be careful how you use the model and people just forget. It was only a
comment, but I think that was important.

Ramin Farnood
Thank you for your comment, Jean-Francis. Actually I have read your paper
with Jaques Silvey on this topic as well and I totally agree with you. I did not
go through the details in my presentation. I have somewhere here made refer-
ence to the Foote effect which is what you pointed out, i.e. effect of dye on the
scattering coefficient. Then as you have pointed out and I also presented here,
scattering coefficient is not only a function of scattering cross-section but also
absorption cross-section. So you are absolutely right. This is a deficiency of
the Kubelka Munk model, but it is still widely used. The way I look at it, the
value of a model is partly due to its simplicity and its ability to predict what
we want to predict. As long as we are clear what the limitations are, as you
pointed out, and we know how to interpret what we get out of the model, we
are fine. The danger is when a model is used without an appreciation of the
limitations of the model. I apologize if I perhaps made you believe that
Kubelka Munk does everything. That is certainly not the case. I showed you

Session 3
 Transport and Moisture Interactions

some example of Kubelka Munk applications, but we need to be aware of its

limitations, and we need to be careful when we interpret the data.

Norayr Gurnagul FPInnovations (from the chair)

Perhaps I can just ask a simple question: can you make some comment about
the use of CIE whiteness versus brightness, especially for paper makers out
there?

Ramin Farnood
Okay, that is a controversial topic. I tried to make some comments in the
manuscript and, if you like you can have a look at it in more detail, but what I
understand is that brightness has been developed for a different purpose. It
was developed originally to quantify the progress of a bleaching process and
it does a fabulous job for that. But because it is simple and widely used for
bleaching process, which by the way is focused very much on the blue region,
it has also found application in paper. But, as you saw when I was talking
about luminosity function, sensitivities are dependent on the wavelength of
light and depending on the type of light source that was used, therefore, you
may have a totally different perception, as published by Byron Jordan more
than 10 years ago. Brightness is valuable but when we come to try to correlate
it with human perception, the choice of illuminant, which in the brightness is
limited to the blue region, would be very important as well.

14th Fundamental Research Symposium, Oxford, September 2009