PRINTABILITY OF HIGHLY STRUCTURED COATINGS DEVELOPED FOR INK JET PRINTING PAPERS Jean-Philippe Boisvert1,*, Jacques Persello2 and Aurélien Guyard1 1

CRPP, Université du Québec à Trois-Rivières,C.P.500,Trois-Rivières,Québec, G9A5H7 2 LCMI, Université de Franche-Comté, 16 route de Gray, 25000 Besançon, France.

Abstract This work reports the development of nanoporous silica aggregates designed for non-contact printing. Various characteristics such as the pore size, the pores size distribution and the porous volume of the nanostructured aggregates were adjusted and measured with different techniques (BET and CTAB surface area, Hg porometry). The aggregates were thereafter mixed with a binder (polyvinylalcool, PVA) and coated on a non-coated common copier paper. The coating structure was investigated by electron microscopy. Non-impact printing (ink jet) on this structured coating was performed. Black and white image analysis of printed typographic characters was correlated to the structure of the nanoporous aggregates, the structure of the coating and the binder/silica compatibility. The results show that printing properties such as smoothness of contours, gray level, contrast, definition, surface area of the typographic characters all depend on the structure properties of the silica and its compatibility with the binder. Large molecular weight PVA molecules presumably cannot penetrate into small size pores and thus most of the porous volume remains accessible for the capture of an ink droplet. The printing properties are at best in this situation. The printing properties fall when the binder fills the porous volume, which makes it no more accessible to ink fluids.

Keywords: polyvinylalcool, nanoporous, silica, structure, ink absorption, printability, ink jet.

(*) To whom correspondence should be addressed. Tel: +(819) 376-5075, Fax: +(819) 376-5148, E-mail: [email protected]

1/9

Introduction A very strong demand exists presently for the development of new highperformance printing papers. This demand follows the development of new digital multimedia devices (scanners, photo and video cameras, etc) now affordable for all and already installed on computers found in many homes and businesses. In most of these situations, color pictures are printed on ink-jet printers. The highly specialized papers used for this purpose must meet various requirements concerning image quality: longitudinal and lateral diffusion of colored pigments must be avoided, but on the other hand, the ink must set quickly, i.e. the solvent must be evacuated rapidly from the surface by diffusion as the ink droplet touch the surface. The later requirement is now much more acute since new generation of ink-jet printers can print both sides of a sheet. Some waiting time is then needed before the other side can be printed, which slows down the printing job. In general, two types of ink-jet papers are found: the polymer-rich and the mineral-rich papers (1). As described in ref. (1), the former type adsorbs the ink solvent by swelling (2) and filtrates the colored pigments on the top of the paper surface, keeping them at the contact area where the droplet touched the surface first. On the other hand, the mineral-rich papers use the porous structure of coatings to drain the solvent by capillary effects, still keeping the pigments on top. Two types of porosities may possibly be involved in this draining process: 1-the interporosity i.e. the porosity created by the more or less dense stacking of mineral particles and 2the intraporosity i.e. the porosity created within porous mineral particles. When conventional coating pigments i.e. calcium carbonate and clays are used, only interporosity can be created because these minerals are non-porous by nature. The size distribution of coating pigments, the binder (latex) content and type of binder are known to influence the ink transfer

process and printability properties, mainly by modifying the pore volume accessible to ink fluids (3-8). In the last decade, new porous aggregates with high intraporous volumes, like amorphous silica aggregates have been specialized and used to enhance the draining capability of the mineral coating, combining both the inter and intraporosity (9,10). Of course, the porous structure of these later coatings has dramatic effects on the diffusion of ink fluids (9). The main purpose of this work is to identify some of the important parameters involved in the ink holding process of nonimpact printing, namely ink-jet printing. Two different types of porous ink holders SiO2 aggregates were prepared and characterized. They were mixed with PVA at different silica contents and pHs. The PVA was also characterized as well as the PVA/SiO2 interaction through adsorption isotherms. The mixture was allowed to dry in order to recover a thin film. The film structure was characterized by electron microscopy. The printability was tested through image analysis of ink jet printed patterns on papers coated with the PVA/SiO2 mixture. The experimental procedure will allow making correlations between surface properties of SiO2 aggregates, film structure and ink absorbency. Material General All experiments were conducted in distilled and deionized water. The chemical reagents are all of analytical grade and were used without further purification. The pH of solutions and suspensions were adjusted to pH 5 and 9 with NaOH. Synthesis and characterization of silica First, primary silica particles were grown from aqueous silicate solutions neutralized by nitric acid. Sodium silicate was produced by dissolving a pyrogenic silica into a concentrated NaOH solution in order

2/9

to reach a molar ratio x = SiO2/Na2O = 3.40. The resulting silicate solution was thereafter diluted to a silica concentration of 0.57M. The precipitation of silica was initiated by diluting an initial batch of silicate solution with water to a concentration of 0.004M; this dilution lowered the pH to 9 and initiates the formation of silica nuclei. They were thereafter allowed to grow and ripen, as described elsewhere (11, 12). In order to remove the salt, which was a by-product of the reaction, the silica sol was circulated in a tangential ultrafiltration module. During this stage the concentration of the sol was maintained constant by addition of deionized water. The pH of the sol was also kept constant at pH 9. A further ultrafiltration stage without replacement of water was used to concentrate the sol up to a given silica volume fraction. Thereafter, the aggregated products were allowed to coalesce under controlled ionic strength and temperature conditions. In order to obtain the SNP1 and SNP2 products, the silica volume fractions of the suspension were adjusted to 0.025 and 0.045 and the ionic strengths were 5.10-2 M and 3.10-1 M, respectively. The characterization methods used to characterize the porous structure of both products are described just below. The porosity has been measured by mercury porometry with a Quantachrome instrument (Autosorb 60 model). The specific surface areas were measured by low temperature N2 BET adsorption isotherms also with a Quantachrome Autosorb Instrument. The external surface areas were probed by hexadecyltrimethylammonium Bromide (CTAB) adsorption. This method is fully described elsewhere (13). The aggregate size distribution was measured by laser diffraction using a Malvern Mastersizer model E instrument. The slurries were diluted in aqueous electrolyte solutions with the corresponding ionic strength, stirred and ultrasonicated in situ five minutes before measurements. The particle concentration

was imposed by the instrument ensuring optimum operational conditions. Polymer characterization The polyvinylalcool (PVA) used in this work was purchased from Fluka (cat. number 9002-89-5). According to the supplier, this polymer has a molecular weight (MW) of 100Kg/mol and a degree of purity of 99%. In the present study the polymer was used without further purification. Viscosity measurements showed a MW of 107 Kg/mol and a gyration radius of 15 nm at pH 5 and 9. The overlap concentration for this polymer is 8 g/l in water. The mass of the elastic chains was estimated from swelling experiments in water. This yields for the unfilled sample to a MW of 7400±400 g/mol. Adsorption isotherms SiO2 suspensions at initial pH 5 or 9 were mixed with increasing concentrations of PVA solutions at same pHs. Five days were allowed for equilibrium at the end of which the suspensions were centrifuged and the total organic carbon (TOC) of the supernatant was measured with a Carbon analyzer (Dohrmann instrument). Only experimental data with (initial TOCequilibrium TOC)/initial TOC >0.2 were taken as significant. Calibration curve was established with known concentrations of PVA. The standard deviation for 3 consecutive measurements was lower than 5%. In order to compare the obtained saturation values at full surface coverage with the maximum possible surface coverage on silica, the adsorption isotherms were also performed on nanometric spherical silica particles. The reader is referred elsewhere for full details on the preparation of these particles (11,12). The particles are 27nm large and the polydispersity index of the suspension is 1.07. In addition to adsorption isotherms, the hydrodynamic thickness d

3/9

for saturated layers of PVA on the silica spheres was measured at pH 5 by viscosimetry (results not shown here) and computed through a relation adapted from the Einstein’s relation (14):

h /h0 = 1+ 2.5(f (1+ d /R) 3 ) where h/h0 is the viscosity relative to the bulk, f is the volume fraction and R is the † particle radius. The thickness of the adsorbed layer was found to be 10nm at saturation and pH 5. This value is fairly comparable to what has been reported elsewhere for a very similar system (14). This technique could not be used at pH 9 because the surface coverage was too low to allow reasonable significance in the results. Film formation and coating The films were prepared in nonadhesive moulds by mixing SiO2 suspensions (f=0.02, pH 5 or 9) with the corresponding PVA solution (5%dwb, pH 5 or 9) in appropriate proportion in order to finish with dry basis volume fractions (SiO2/(SiO2+PVA)) ranging from 0.05 to 0.35 once the solvent has been evaporated. Prior to evaporation, the PVA surface coverage is well above saturation. At the end of the evaporation process, the films were removed from the moulds and analyzed by electron microscopy. The PVA/SiO2 slurries were also coated on a common non-coated paper (Domtar copier paper), which rates an ISO brightness of 80. The coated device was a lab-coater with striped rods (rod #6). The sheets were thereafter dried and printed in black and white on a Cannon BJC4400 ink jet printer with the ink recommended by the manufacturer. Electron and optical microscopy Transmission electron microscopy (TEM) sample preparation consisted of immobilizing the PVA/SiO2 films in a resin and slicing the film with an

ultramicrotome. The cross section was fixed onto a coated carbon grid. TEM examination was performed using a Philips instrument operating at 120 kV. Optical microscopy was used to magnify and image the ink jet printed typographic characters. Images were digitized and processed through an image analysis software (Image SXM, NIH) in order to quantify the gray level and the contour of the printed patterns. Results and discussion Two porous SiO2 materials were synthesized. Their morphological differences can be qualitatively estimated from the TEM micrographs shown in Figure 1. However, great care must be taken when interpreting two-dimension images resulting from the projection of a three-dimension object. This is especially true when high silica contents are involved and when the objects observed are smaller than the thickness of the ultramicrotome slice. Moreover, the micrographs in Figure 1 show SiO2 aggregates embedded into PVA rather than SiO2 aggregates alone. This being said, it is clear from Figure 1 that the SNP1 and SNP2 aggregates have different structures. Obviously, the SNP1 aggregates seem much more connected than the SNP2 product. Many isolated objects (small linear clusters, primary particles) are present in SNP2, while these are almost inexistent in SNP1. The main physical characteristics of the products are grouped together in Table 1. Both methods of surface area measurement using different probes (N2 and CTAB) are concordant, indicating that the entire surface is external. The presence of isolated primary particles and linear objects in SNP2 can explain in part the higher surface area of this product. According to mercury porometry measurements in Figure 2, the SNP1 product shows smaller pore size and lower porous volume than SNP2. This is consistent with the surface area

4/9

Figure 1. MET micrographs of the two nanoporous products (SNP1/ PVA film (top) and SNP2/PVA film (bottom)) used in this study. The silica content of these composite materials is 25% on a dry weight basis. The scale on the micrographs represents 200nm.

Table 1. Main characteristics of the porous materials used in this study BET surface area (m2/g) SNP1

113

CTAB surface area (m2/g) 133

SNP2

196

194

Total porous volume (ml/g)

Intra-porous volume (Vp) (ml/g)

Effective density8 (g/ml)

Aggregate size (mm)

0.6

0.6

0.94

1.5

0.9

0.8

0.73

2

* The effective density is calculated according to: 1/(1/dSiO2+Vp) where dSiO2 is the silica density (2.16 g/ml)

5/9

0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 0,001

Monomer/nm2

Porous volume (ml/g)

2,5 2 1,5 1 0,5 0 0,01

0,1

1

10

100

1000

Pore size (mm)

Figure 2. Porogram of the SNP1 and SNP2 products. Keys: Filled diamonds: SNP1, Open triangle: SNP2.

measurements and TEM reported above. The higher connectivity of SNP1 leads to lower surface area, lower pore size, lower porous volume and higher effective density. As can be seen, the products are greatly different and significant differences in printability are expected. Adsorption isotherms As a first step, the PVA was adsorbed on spherical SiO2 particles at pH 5 and 9. On these particles, one hundred percent of the surface is available. Accordingly, the surface density at saturation is the maximum possible surface density that can be reached on this material with the PVA used in this study. This value is considered as a reference value to which the saturation limits of the porous aggregates will be compared through adsorption isotherms. The dashed horizontal lines in Figure 3 indicate the reference surface densities at saturation for the silica spheres. The upper line refers to the maximum surface density at pH 5, while the lower line is for pH 9. The great difference in the saturation values for the spheres is expected because of the great differences in the surface properties of silica at these two pHs. Indeed, earlier works on adsorption of neutral polymers on silica have shown that neutral polymers like PVA adsorb on silica through hydrogen bonding with silanol surface

0

0,002

0,004

0,006

0,008

0,01

[PVA]eq (monomol/L)

Figure 3. Adsorption isotherms of 100kg/mol PVA on SNP1 and SNP2 at pHs 5 and 9. Squares: SNP2, Diamonds: SNP1. Open symbols: pH 5, Filled symbols: pH 9. The dashed horizontal lines indicate the PVA surface densities at saturation for the silica spheres (reference). The upper line refers to the maximum surface density at pH 5, while the lower line is for pH 9.

sites (15,16). Since the surface density of these sites is decreasing as the pH rises from pH 2 to higher values (11), decreasing surface coverage is expected with PVA as the pH increases from 5 to 9. At pH 5, the large majority of surface sites are non-ionized silanol groups while at pH 9 more than half of them are ionized (17). Adsorption of PVA on porous structures reveals fine details about the dimensional parameters involved in the adsorption with such material. When the polymer adsorbs, two main and opposing effects are involved: the loss of entropy due to the deformation (unfavorable to adsorption) and the gain of enthalpy due to the creation of H-bonds (favorable). A correlation can be established between pore size (Figure 2) and surface saturation (Figure 3). Indeed, most of the SNP1 surface area is located inside pores having a diameter lower than 10nm. On the other hand, the SNP2 has most of its pores ranging between 20 to 40 nm, and consequently the pores can more easily accommodate the polymer coils. It is recalled that the coils have a radius of gyration of 15 nm and cannot penetrate into pores smaller than about 40 nm without being deformed. However, as reported previously in Material and 6/9

Methods, some deformation is expected, even upon unrestricted adsorption. Indeed, on the curved surface of spherical silica, the polymer layer thickness is only 10 nm, i.e. one third of the free polymer coil size. According to the Figure 3, only half of the SNP2 surface is accessible to the coils at pH 5, if one takes the saturation coverage of spherical particles as a reference. The situation is even worst with the SNP1 product: only 20% of the surface is accessible at the same pH. Such a low surface coverage at saturation, compared to the reference, as well as the difference between both products can be explained by the degree of openness of the two porous products. The higher degree of openness of the SNP2 aggregates leads to a higher surface fraction accessible to the nondeformed polymer coils. The lower surface coverage observed with the SNP1 product in Fig. 3 could presumably be explained by the energetic considerations described above, i.e. the polymer coils cannot penetrate deep inside the pores without being subjected to an important and energetically unfavorable deformation. The pores remains then free of any polymer and might readily adsorb ink (see below). The situation is less clear at pH 9 where the PVA affinity for the surface is very low (filled symbols in Fig. 3), and consequently no comparative information between both products can be extracted from adsorption isotherms. The results are nonetheless presented in order to illustrate how poor is the affinity between both constituents. At this pH, the gain in enthalpy would be too low to overcome the loss of entropy due to extensive deformation, as a result that almost no adsorption occurs. The experimental conditions during the preparation of the slurries were somewhat different to that occurring in adsorption experiments because the polymer concentration was much higher in the slurries with the consequence that the polymer coils are entangled (semi-dilute regime), as described by deGennes (18).

However, the above arguments regarding the energetic cost of polymer deformation and its limiting effect on adsorption remain fully appropriate. Structure vs printability An uncoated paper was used as a coating support and comparative reference for ink jet printability testing. This uncoated paper has thereafter been coated with a pure PVA solution. The uncoated paper has also been coated with the PVASNP1 or PVA-SNP2 mixture. All four different types of papers were ink-jet printed. As one can see from the results reported in Table 2, large differences on the printing properties of typographic characters between the four papers are observed and the results can be grouped together in three different categories that can be qualitatively compared, at least on a relative point of view: 1-papers with an important normal ink diffusion, but low lateral diffusion. These should lead to low perimeter, surface area and gray level. Compared to the others, the uncoated paper falls into this first category. 2-papers with an important lateral diffusion and low normal diffusion. This category can be characterized by high perimeter and low gray level. The gray level depends mainly on the extent of the ink smearing on an impermeable surface. According to the print surface area and particularly to the perimeter length, it seems clear that lateral diffusion occurs to a much higher degree with SNP2 and pure PVA than with SNP1 or the uncoated reference. Because of the higher degree of ink smearing, the gray level is lower with SNP2 and PVA. These two papers fall into the second category. 3papers with low normal and lateral diffusion. In this last category, low perimeters and high gray level are expected. According to the results presented in Table 2, the SNP1-coated paper simultaneously prevents both the normal and lateral diffusions more efficiently than the other papers.

7/9

Table 2. Gray level and area of a character (capital T ) printed with a commercial ink jet printer on an uncoated paper and on the same paper rod-coated with PVA alone, PVA/SNP1 and PVA/SNP2 slurries at pH 5 and SiO2 dry volume fraction = 0.35. Gray level * Printed surface area (mm2) Perimeter (%) s=±3% (mm) s=±3% Uncoated paper 84 (s** =±3%) 1.80 15.0 Paper coated with PVA

81 (s=±9%)

1.83

19.3

Paper coated with SNP1

90 (s=±2%)

2.17

15.4

Paper coated with SNP2

81 (s=±7%)

2.53

23.2

* 100% = black, 0% = white. Gray level of background = 40% ** STD Dev of the gray level from pixel to pixel along the vertical bar of the letter T. The std dev. from letter T to letter T for gray level is 1%, well within the STD Dev from pixel to pixel.

The relationship between the structure and the PVA/SiO2 can explain most of the results presented in Table 2. Indeed, the higher free porous volume i.e. intra-porosity not filled by PVA, would be responsible for the higher ink absorbency of SNP1. Because of the higher void volume, the ink would stay more likely at the touching contact area where the ink droplet hits the surface. Accordingly, the ink holding capacity is higher with the SNP1-coated surface than with any other surfaces, including the uncoated surface. In the latter case, the void volume comes only from inter-porosity between the nonporous mineral fillers used in the manufacturing of this paper (mainly CaCO3) and its void volume is generally about 0.15 ml/g, at most (3). This value is four times lower than the total porosity of the SNP1 product, with the result that the gray level is lower with the uncoated paper. As explained above, the small pore size of the porous SNP1 product is expected to prevent pore clogging by extensive PVA adsorption. Then, most of the 0,6ml/g of intra-porosity is expected to remain accessible to ink fluids. On the other hand, the high pore size of SNP2 allows the PVA molecules to adsorb on most of the surface area and to occupy most of the voids. The resulting material can be assimilated to SiO2 aggregates filled with PVA and embedded in a PVA matrix. When the ink hits the

surface, this non-porous (impermeable) surface prevents the normal diffusion but not the lateral smearing of ink. Accordingly, the ink holding capacity of SNP2 is poor and comparable to that of the pure PVA coating. The structure and surface properties of SNP1 and SNP2 coating could explain most of the differences between the two products as well as the differences between SNP2 and the reference. Conclusion This study showed that not only the porous volume of silica aggregates is important regarding the ink absorbency but also how accessible the volume is to the binding polymer. This accessibility depends on the aggregate structure (intraporosity) but also on the free energy costs of polymer deformation, which in turn depends on the polymer affinity for the surface (pH). The combination of both effects determines how important the remaining free porous volume accessible to the ink fluids will be, and ultimately how the printing properties such as shade and pixel resolution will be. Of course, pores filled with polymer lose their absorbency capacity. The result of this is an impermeable PVA film with silica aggregates embedded into it, with printability properties similar to that of pure PVA film.

8/9

Acknowledgments The authors would like to acknowledge Agnès Lejeune from UQTR for the supply of the TEM micrographs. The Ministère des Relations Internationales du Québec as well as NSERC-International Opportunity Fund are also acknowledged for financial support through research grants. References 1. Chapman, D.M., and Michos, D., J. Imag. Sci. Technol. 44 (2000) 418422. 2 . Poeschke R., and Stumpf, A., Proceedings of the Imag. Sci. T e c h n o l . , NIP conference, Springfield, VA (1998), 266. 3. Xiang, Y., and Bousfield D.W., J. Pulp Paper Sci. 26 (2000) 221-227. 4 . Kowalczyk G.E., and Trksak, R.M., TAPPI J., 81 (1998) 181190. 5 . Arai, Y., Nojima,K., Proceedings of the TAPPI Coating Conference (1997) 133-142. 6. Zang Y.H., Aspler J.S., J. Pulp and Paper Sci. 24 (1998) 141-145. 7. Xiang Y., Bousfield D.W., J. Pulp and Paper Sci. 26 (2000) 221-227. 8 . Desjumaux D., Bousfield D.W., Glatter T.P., Donigan D.W., Ishley

J.N., Wise K.J., J. Pulp and Paper Sci. 24 (1998) 150-155. 9. Chapman, D.M., Proceedings of the TAPPI Coating Conference (1997) 73-93. 1 0 .Takahashi, M., Sato, T., and Ogawa, M., Japan TAPPI J. 4 2 (1988) 923-931. 11. Iler, R.K., The chemistry of Silica, Wiley Ed., New-York (1979). 12. Persello, J., Magnin, A., Chang, J., Piau, J.M., and Cabane, B., J. Rheol. 38 (1994) 1845-1870. 13. Jansen, J., and Kraus, G., Rubber Chem. Technol. 44 (1971) 1287. 14. Lafuma, F., Wong, K., and Cabane, B., J. Colloid Interface Sci. 1 4 3 (1991) 9-21. 15. Van der Beek, G.P., and CohenStuart, M.A., J. Phys. (France) 4 9 (1988) 1449. 1 6 .Cohen-Stuart, M.A., Fleer, G.J., Scheutjen, J.M.H.M., J. Colloid Interface Sci. 97 (1989) 526. 17. Foissy A., Persello J., “The surface properties of Silica”, John Wiley and Sons Ltd., Ed. A.P. Legrand, pp365-414, (1998). 18. deGennes, P.G., “Scaling concepts of polymer physics”, Cornell University Press, (1979).

9/9