Optimization of Pigment Dispersion Preparation for Lithographic Printing Ink

DOI : 10.17577/IJERTV5IS120222

Download Full-Text PDF Cite this Publication

Text Only Version

Optimization of Pigment Dispersion Preparation for Lithographic Printing Ink

Hoang Thi Kieu Nguyen, Tran Van Thang

Department of Printing Engineering Hanoi University of Science and Technology

Hanoi, Vietnam

Ho My Thanh

Technical Department National Banknote Printing Plant

Hanoi, Vietnam

AbstractIn this study optimization of parameters of the pigment dispersion for lithographic inks was carried out. The aim is to minimize the size of dispersed pigment particles. A factorial design was used to evaluate the effects and interactions of three factors, that is, varnish concentration, pigment concentration, and enhanced degree of roller pressure in each stage of the milling process, on the pigment dispersion quality. The optimal conditions obtained from the desirable response are varnish concentration of 62.27 % by weight, pigment concentration of 15.33 % by weight and pressure enhancement of 20 N/mm2. The validity of the statistical experimental strategies was verified by the pigment dispersion samples prepared under the optimized conditions.

Keywords Lithographic Ink; Printing Ink Production; Pigment Dispersion; Experimental Design

  1. INTRODUCTION

    The manufacture of printing ink is a technologically advanced, highly specialized and complex process. The basic formulation of ink involves the grinding of pigment in a vehicle to form the pigment dispersion, then letting down the pigment dispersion with suitable additives to meet rheological and functional properties.

    The dispersion of pigments in printing inks is very important. The flow behavior of the ink is controlled to some extent by the dispersion properties of the material. To achieve the optimum benefits of a pigment, it is necessary to obtain as full a reduction as possible to the primary particle size [1]. The color strength of a pigment depends on its exposed surface area, and the smaller the particle the higher the surface area and thus stronger the color [2]. Increasing demands on quality printing inks regarding the optical characteristics such as gloss, transparency or color strength requires the use of more effective dispersing techniques [3].

    The pigment dispersion process involves the breakdown of associated particles into smaller particles and their distribution in a fluid, leading to a colloidal suspension. The steps involved in the pigment dispersion process as follows: Wetting of the pigment particles, breakdown of the pigment particles and stabilization of the dispersion [3].

    The quality of the final dispersion is dependent on the optimization of many influencing factors such as pigment volume concentration, pigment to vehicle ratio, grinding media, grinding time and pressure, temperature,[1,3,4].

    In order to optimize the parameters of pigment dispersion for lithographic inks, in this study the 2k experimental design

    [5] was applied for evaluation of the individual contribution of selected variables to the final size of dispersed pigment

    particles. The investigated factors were varnish (vehicle resin) concentration, pigment concentration, and grinding pressure.

  2. EXPERIMENTAL

    1. Materials

      Pigment red 146 (Naphtol Carmine FBB) was used for dispersion supplied by DIC Corporation, Japan. The physical form is a red powder with the average particle size of 0.5 µm.

      High density varnish (Phenolic resin/solvent) from DIC was used as adispersing agent for pigment dispersion preparation. Solvent DIC was added to adjust the rheological properties of the system.

    2. Three-roll mill Bühler

      Bühler three-roll mill was used in the grinding state. Through mechanical energy (impact and shear forces), the pigment agglomerates are broken up and disrupted into smaller units and dispersed (uniformly distributed).

      The schematic diagram of this machine is given in Fig. 1. The speed ratio of three rolls is 1:3:9. The pressure between the rollers can be set up in a wide range from almost 0 N/mm2 to 20 N/mm2. The machine has high roll speeds and excellent cooling properties.

    3. Pigment dispersion preparation

      The pigment powder was mixed in the varnish and the solvent under low shear for 20 min. This suspension was then added to the mill. The constant pressures between rollers (P1-2

      = 9 N/mm2 and P2-3 = 8 N/mm2) were applied and the roller speedset to 200 rpm. The temperature was kept at 25 10C.

      Fig. 1. Schematic diagram of three-roll mill

      The milling process lasted in 15 minutes. Again the same milling process was performed 3 more times, but the roller pressures were increased an adequate amount for each time.

      After 4 passes a pigment suspension samples were taken to observe the particle size reduction.

    4. Particle size determination

      The degree of dispersion was measured using PD-1510 Grindometer 25. This is a stainless steel block containing a

      The experiments were evaluated in order to fit a regression model

      = 0 + 11 + 22 + 33 + 1212 + 1313 +

      2323 + 123123 (1)

      Where, yj (j = 1 ÷ N, N = 8) is the response variable to be modeled; xj (i = 1 ÷ 3) is the independent variable which influence y; b0, bi (i = 1 ÷ 3), biu (i = 1 ÷ 3, u = 1 ÷ 3) are model terms, that are estimated by

      shallow groove which is graduated in depth from zero to 25

      = 1

      (2)

      microns. A sample of ink was placed at the deep end, draw

      =1

      down with a steel blade towards shallower part, the degree of

      = 1

      (3)

      dispersion was taken as the point at which the continuous ink film breaks down across the groove.

    5. Experimental design và data analysis

    In this work, a factorial design in which the influences of three experimental factors, e.g. varnish concentration, pigment concentration, and enhanced degree of roller pressure in each stage of the milling process, on the response,

    i.e. the dispersed particle size, was investigated. Two different levels were assigned to each factor. The factorial design is shown in Table 1. The levels of the factors are given by (minus) for low level and + (plus) for high level. A zero- level is also included, a centre, in which all variables are set at their mid value.

    =1

    Analysis of variances (ANOVA) was used for graphical analyses of the data to obtain the interaction between the process variables and the responses. The quality of the fitted model was expressed with the coefficient of determination, R2, and its statistical significance was checked by the F-test. Model terms were selected or rejected based on the p value (probability) with 95% confidence level.

    The regression model for real variables (z) describing the relationship between the investigated factors was determined from (1) by replacing variables x with z:

    2(+0)

    A sign table, or design matrix, used to calculate the main

    =

    (4)

    effects and the interaction effects from the factorial design is

    Where = + ; +, 0, are values of the ith

    constructed in Table 2.

    variable at high,

    an m l

    el, espectively.

    low d id ev r

    Exp. No

    Variable

    Response

    x1

    x2

    x3

    1

    y1

    2

    +

    y2

    3

    +

    y3

    4

    +

    +

    y4

    5

    +

    y5

    6

    +

    +

    y6

    7

    +

    +

    y7

    8

    +

    +

    +

    y8

    Exp. No

    Variable

    Response

    x1

    x2

    x3

    1

    y1

    2

    +

    y2

    3

    +

    y3

    4

    +

    +

    y4

    5

    +

    y5

    6

    +

    +

    y6

    7

    +

    +

    y7

    8

    +

    +

    +

    y8

    TABLE I. FACTORIAL DESIGN

    TABLE II. MATRIX OF FACTORIAL DESIGN

    The optimum values of selected variables were obtained by using MATLAB 7.0. The interactive effects of the independent variables on the dependent ones were illustrated by three dimensional plots. Finally, two additional experiments were conducted to verify the validity of the statistical experimental strategies.

  3. RESULTS AND DISCUSSION

    Two different levels were assigned to each factor. These levels were experimentally determined to assure that the system has viscosity and tack value in the range of lithographic printing technology. The investigated results are reported in Table 3 5.

    Pigment (g)

    Varnish (g)

    Solvent (g)

    Varnish concentra- tion

    (% wt.)

    Viscosity, 300C

    (Pa.s)

    Tack value, 250C, 100 m/s (T.U)

    100

    370

    30

    74

    80

    230

    100

    360

    40

    72

    74

    218

    100

    350

    50

    70

    72

    210

    100

    340

    60

    68

    65

    192

    100

    330

    70

    66

    56

    179

    100

    320

    80

    64

    48

    169

    100

    310

    90

    62

    42

    150

    100

    300

    100

    60

    39

    145

    Pigment (g)

    Varnish (g)

    Solvent (g)

    Varnish concentra- tion

    (% wt.)

    Viscosity, 300C

    (Pa.s)

    Tack value, 250C, 100 m/s (T.U)

    100

    370

    30

    74

    80

    230

    100

    360

    40

    72

    74

    218

    100

    350

    50

    70

    72

    210

    100

    340

    60

    68

    65

    192

    100

    330

    70

    66

    56

    179

    100

    320

    80

    64

    48

    169

    100

    310

    90

    62

    42

    150

    100

    300

    100

    60

    39

    145

    TABLE III. EFFECT OF VARNISH CONCENTRATION ON RHEOLOGICAL PROPERTIES OF THE PIGMENT DISPERSION

    Exp.No

    I

    x1

    x2

    x3

    x1x2

    x1x3

    x2x3

    x1x2x3

    Response

    1

    +1

    -1

    -1

    -1

    +1

    +1

    +1

    -1

    y1

    2

    +1

    +1

    -1

    -1

    -1

    -1

    +1

    +1

    y2

    3

    +1

    -1

    +1

    -1

    -1

    +1

    -1

    +1

    y3

    4

    +1

    +1

    +1

    -1

    +1

    -1

    -1

    -1

    y4

    5

    +1

    -1

    -1

    +1

    +1

    -1

    -1

    +1

    y5

    6

    +1

    +1

    -1

    +1

    -1

    +1

    -1

    -1

    y6

    7

    +1

    -1

    +1

    +1

    -1

    -1

    +1

    -1

    y7

    8

    +1

    +1

    +1

    +1

    +1

    +1

    +1

    +1

    y8

    TABLE IV. EFFECT OF PIGMENT CONCENTRATION ON RHEOLOGICAL PROPERTIES OF THE PIGMENT DISPERSION

    Pigment (g)

    Varnish (g)

    Solvent (g)

    Pigment concentra- tion

    (% wt.)

    Viscosity, 300C

    (Pa.s)

    Tack value, 250C, 100 m/s (T.U)

    70

    350

    80

    14

    70

    150

    77

    350

    73

    15

    65

    180

    79

    350

    71

    16

    62

    189

    100

    350

    50

    20

    60

    210

    119

    350

    31

    24

    58

    215

    123

    350

    27

    25

    55

    217

    137

    350

    13

    27

    49

    220

    TABLE V. EFFECT OF ENHANCED DGREE OF ROLLER PRESSURE ON RHEOLOGICAL PROPERTIES OF THE PIGMENT DISPERSION

    Enhance- ment of roller pressure (N/mm2)

    Roller pressure, P1-2/P2-3

    (N/mm2)

    Dispersed particle size (µm)

    Viscosity, 300C

    (Pa.s)

    Tack value, 250C, 100 m/s (T.U)

    9.0/8.0

    0.6

    9.6/8.6

    11.2/10.2

    10

    50

    220

    11.8/10.8

    9.0/8.0

    1.0

    10/9.0

    11/10

    6

    57

    200

    12/11

    9.0/8.0

    1.5

    10.5/9.5

    11.5/10.5

    4

    60

    195

    13/12

    9.0/8.0

    2.0

    11/10

    12/11

    4

    70

    187

    14/13

    9.0/8.0

    2.5

    11.5/10.5

    14/13

    4

    80

    170

    16.5/15.5 *

    1. (* ): Limited pressure of the machine

      As required, the viscosity is from 42 to 72 Pa.s and the tack value is from 150 to 220 T.U. Corresponding to these ranges, the experimental domains of three investigated factors were determined (see Table 6).

      Eight experiments in the factorial design and three experiments at the center point were simultaneously performed. All the experiment parameters are reported in Table 7 and the model matrix is given in Table 8.

      Coefficient values and statistical parameters obtained for the model are given in Table 9. For assessing the statistical significance of the result, a t-test (t-Student) was carried to the 95% confidence level.

      TABLE VI. INVESTIGATED FACTORS: LEVELS AND CONDITIONS

      Factor

      Experimental domain

      Level (-)

      Level (0)

      Level (+)

      z1:Pigment concentration (% wt.)

      15

      20

      25

      z2:Varnish concentration (% wt.)

      62

      66

      70

      z3:Enhancement of roller pressure (N/mm2)

      0.6

      1.3

      2

      TABLE VII. EXPERIMENT PARAMETERS

      Exp

      No

      Varnish concentration (% wt.)

      Pigment concentration (% wt.)

      Solvent concentration (% wt.)

      Enhance- ment of roller pressure (N/mm2)

      Dispersed particle size

      (µm)

      1

      62

      15

      23

      6

      3

      2

      70

      15

      15

      6

      9

      3

      62

      25

      13

      6

      7

      4

      70

      25

      5

      6

      10

      5

      62

      15

      23

      20

      2

      6

      70

      15

      15

      20

      6

      7

      62

      25

      13

      20

      6

      8

      70

      25

      5

      20

      5

      9

      66

      20

      14

      13

      5

      10

      66

      20

      14

      13

      6

      11

      66

      20

      14

      13

      6

      TABLE VIII. MODEL MATRIX AND RESPONSE

      Exp. No

      I

      x1

      x2

      x3

      x1x2

      x1x3

      x2x3

      x1x2x3

      Particle size (µm)

      1

      +1

      -1

      -1

      -1

      +1

      +1

      +1

      -1

      3

      2

      +1

      +1

      -1

      -1

      -1

      -1

      +1

      +1

      9

      3

      +1

      -1

      +1

      -1

      -1

      +1

      -1

      +1

      7

      4

      +1

      +1

      +1

      -1

      +1

      -1

      -1

      -1

      10

      5

      +1

      -1

      -1

      +1

      +1

      -1

      -1

      +1

      2

      6

      +1

      +1

      -1

      +1

      -1

      +1

      -1

      -1

      6

      7

      +1

      -1

      +1

      +1

      -1

      -1

      +1

      -1

      6

      8

      +1

      +1

      +1

      +1

      +1

      +1

      +1

      +1

      5

      Coefficient

      Coefficient value

      Standard deviation

      p-value

      b0

      6

      0.20

      < 0.05

      b1

      1.5

      0.20

      < 0.05

      b2

      1

      0.20

      < 0.05

      b3

      -1.25

      0.20

      < 0.05

      b12

      -1

      0.20

      < 0.05

      b13

      -0.75

      0.20

      > 0.05

      b23

      -0.25

      0.20

      > 0.05

      b123

      -0.25

      0.20

      > 0.05

      Coefficient

      Coefficient value

      Standard deviation

      p-value

      b0

      6

      0.20

      < 0.05

      b1

      1.5

      0.20

      < 0.05

      b2

      1

      0.20

      < 0.05

      b3

      -1.25

      0.20

      < 0.05

      b12

      -1

      0.20

      < 0.05

      b13

      -0.75

      0.20

      > 0.05

      b23

      -0.25

      0.20

      > 0.05

      b123

      -0.25

      0.20

      > 0.05

      TABLE IX. COEFFICIENT VALUES AND STATISTICAL PARAMETERS OBTAINED FOR THE MODEL

      As the results shown in Table 9, with the confidence value < 95%, the coefficients b13, b23 and b123 are not significant and are rejected from the model. The obtained equation is as follows.

      = 6 + 1.51 + 2 1.253 12 (5)

      This model was then analyzed by F- statistical test for analysis of variance (ANOVA) to assess the goodness of fit. The analysis results are presented in Table 10.

      TABLE X. STATISTICAL PARAMETERS OBTAINED FROM THE ANOVA TEST PERFORMED FOR THE MODEL

      Source of variation

      Sum of square (SS)

      Degree of freedom (ddl)

      Average square

      Fisher number

      Signification

      R2

      Regression

      25.5

      4

      6.38

      5.91

      4.53

      0.86

      Residues

      6.5

      6

      1.08

      Lack of fit

      5.83

      4

      1.46

      4.29

      19.25

      Pure error

      0.67

      2

      0.34

      The value of F-statistic (the ratio of mean square due to regression to mean square to real error) of 5.91 is larger than F0.05,4,6 (4.53) so the model is significant at the chosen probability level and it is correct [6]. In addition, the lack of fit error was used to test whether the model can fit the data well. The ratio between lack of fit (SSlof) and pure experimental error (SSpo) is much smaller the critical F0.05,4,2 (19.25). This result confirms that the model adequately fits the data. The R2 of 0.86 also indicates that only 14% of the total variation could not be explained by the empirical model [6]. Clearly, at that significance level, it is acceptable to use the obtained model that does not include the rejected terms.

      Replacing the x variables by the z factors, the model for real variables is obtained:

      = 1.381 + 3.52 0.183 0.0512 86.43 (6)

      The function above is now describing how the experimental variables and their interactions influence the final particle size. The model shows that pigment concentration has the largest influence on the size. An increase of this factor with one scaled unit (e.g., from 1 to 2% by weight) results in an increase of the size by 3.5 µm. An opposite effect is observed with milling pressure, higher pressure smaller particle size is. Also following (6) the interaction of varnish and pigment concentration has an effect on the dispersed particle size, but not much meanwhile the interactions of these factors with the last one are not observed. These conclusions are clearly shown in Fig. 2 and Fig. 3. A curvature in the surface of varnish and pigment concentration indicates that these factors are interdependent (Fig. 2). The response surface (Fig.3) implies that the optimal conditions were exactly located inside the design boundary. The optimal conditions are as follows. z1 = 62.4% by weight; z2 = 15.5% by weight; z3 = 2.0 N/mm2. Under these conditions the minimized value of particle size is 1.97 µm.

      In order to verify the statistical analysis, the confirmation samples were prepared. The characteristics of these samples are reported in Table 11.

      Fig. 2. Surface graph of response y (particle size) showing the effect of of varnish and pigment dosage (at enhancement of pressure = 2 N/mm2)

      Fig. 3. Surface graph of response y (particle size) showing the optimal conditions

      The result indicates that under the optimized conditions the prepared pigment dispersion has a good quality. The particle size almost reaches to the primary size and no sedimentation is observed during the storage time. In addition, the rheological properties of the dispersions are fully suitable for the preparation of lithographic printing ink [3].

      TABLE XI. CHARACTERIZATION OF INK SAMPLES PREPARED UNDER THE OPTIMIZED CONDITIONS

      Characteristic

      Sample 1

      Sample 2

      Viscosity (300C), Pa.s

      40

      39

      Tack value (250C, 100m/s), T.U

      158

      150

      Particle size, µm

      2

      2

      Color density (ink thickness: 5g/m2)

      1.5

      1.5

      Stability (250C, 1 month)

      No sedimentation

      No sedimentation

  4. CONCLUSIONS

The results showed that the three factors considered in this study play an important role in the pigment dispersion process. The optimal conditions obtained for varnish concentration, pigment concentration, and enhancement of milling pressure are 62.4 % by weight, 15.5% by weight and

2.0 N/mm2, respectively. Under these conditions, about 2 m of particle size is obtained. The conducted pigment dispersions showed rheological properties and dispersion stability suitable for lithographic printing ink.

REFERENCES

  1. L. Pal, P. D. Fleming, "The Study of Ink Pigment Dispersion Parameters, The Hilltop Review: Vol. 2: Iss. 1,pp.60-70, 2006.

  2. A. Frimova, A. Pekarovicova, P. D. Fleming,Ink Stability During Printing, TAGA J., 2, pp.121-122, 2006

  3. L.H. Leach, L. J. Pierce, E. P. Hickman, The printing ink manual, 5th ed., Springer, 2007

  4. S. Bastani, M. Pishvaei, M. Jalili, Sh.Sorooshnia, The Effect of Pigment Concentration and Particle Size Distribution on the Rheological Behavior of Lithography Inks, Journal of Color science and Technology, 2010

  5. T. Lundstedt, E. Seifert, L. Abramo, Experimental design and optimization, Chemometrics and Intelligent Laboratory Systems, vol. 42, pp.340, 1998

  6. I. Fendri1, L. Khannous, A. Timoumi, Optimization of coagulation- flocculation process for printing ink industrial wastewater treatment using response surface methodology, African Journal of Biotechnology Vol. 12(30), pp. 4819-4826, 2013

Leave a Reply