Molecular Weight and Rheology: A Sketch of a Resin.
This paper will hopefully shed some light on looking at resin properties in relation to molecular weight. We will show how some tests performed in the lab can correlate on a broad basis to molecular weight as provided by the GPC (gel permeation chromatograph) process. We will also look at the effect solvent or oil has on determining the apparent nature of the resin. Finally, we will provide some ways to differentiate the terms listed above and offer definitions which may clarify conversations when discussing these polymers.
Molecular Weight vs. Resin Analysis
Molecular weight determination on complex polymers like rosin-based resins can be performed many ways. Since these polymers do not have a single molecular weight number, a relative value representing the molecular weight distribution must be determined. The most common reported value is Mw (weight-average molecular weight)
"Weight-average molecular weight is determined from experiments in which each molecule or chain makes a contribution to the measured result. This average is more dependent on the number of heavier molecules than is the number-average molecular weight (Mn), which is dependent simply on the total number of particles. Bulk properties associated with large deformations, such as viscosity and toughness, are particularly affected by Mw values. 
Though many customers are interested in this value, few have the capability or equipment to accumulate this data. We looked at several test methods over a wide range of resins to determine common test methods which can at least correlate to molecular weight. Here is the list of test methods we used:
We collected resin samples (several lots from each product if possible) of the entire North American, South American (Ascona) and European Akzo Nobel product line, as well as many other resins. We ran each resin according to the same procedures using each of the above test methods. The solution of choice was 45 percent resin in TXIB. A detailed explanation for the solution choice is covered in an Ink World article.  Since there arc tests that depend on analyzing a resin solution from one shear rate to another (ex. Duke Viscosity) or from one temperature to another (ex. Carrimed Oscillatory measurements), all resins in the database had to be infinitely compatible with the solvent to be applicable for this study.
Below are the results of this comparison. We found that resins from the same location hold better correlation to molecular weight than the entire database. This phenomenon is due probably to the fact that the resins produced in the same location are generally developed with similar formula and process styles. We will offer examples of both to illustrate.
The 1 percent intrinsic viscosity had the best correlation to molecular weight. This test is used at Akzo Nobel Resins on many of the high molecular weight resins to ensure that the molecular weight is sufficient, as well as apparent viscosity, for consistency. We make a 1 percent resin solution in toluene and determine the intrinsic viscosity in an automatic capillary (Cannon-Fenske) viscometer. This test allows the individual polymers to stretch to their fullest extent and have minimal intramolecular forces thereby giving the minimal viscosity possible in the absence other influences. The lower the concentration, theoretically the better the correlation.
Normally in the academic world, a true intrinsic number is found by the "extrapolation of the reduced viscosity to zero concentration."  Obviously this procedure would be too time consuming for production applications. The slope or change of viscosity per concentration can be very enlightening on the resin's character.
The next method that appears to have a good correlation to molecular weight is tan delta. By measuring the solution in a precise oscillatory manner, we can produce stress and viscosity values relative to stored or lost energy.
G' -- storage modulus
G" -- loss modulus
tan delta - G"/G' 
The lower the tan delta value is for a given temperature, the more elastic or gelled the solution appears. We will discuss the impact solvent choice has on this value in the next segment.
Tan delta 40/20 is a concept from Akzo Nobel to look at the heat stability and elastic behavior of the resin at room temperature and temperatures more indicative of press conditions (40[degrees]C). This ratio reveals that some resins may appear to be elastic (low tan delta) at room temperature but when you increase temperature while maintaining the same strain requirements, the tan delta increases significantly. (Both G' and G" decreases as temperature increases, but the tan delta change is what we are monitoring in this method.) This can explain why a formulator can make ink in the lab, which appears to have sufficient rheology to run on the high speed presses, the ink appears to fall apart.
In this TXIB solution where the resin is infinitely soluble at both temperatures, the correlation to molecular weight is very good. If the system becomes more compatible at the higher temperature, the 40/20 increases in value. In actual application conditions this increase is expected to some extent. The formulator uses limited incompatibility to his advantage for dry or set speed.
The final method, which appears to show some indication of molecular weight correlation, is the Duke Viscometer measured Shortness Ratio. While the viscosity at both high and low shear rates show poor correlation to molecular weight, the slope of this change from one energy level to another does correlate. Again, this correlation to molecular weight only works when the resin is equally compatible with the solvent at both shear rates.
Effect of Solvency
Solvent choice in a system is very important. A resin supplier cannot supply application data on every possible solvent, cosolvent and coresin combination. We can work with a customer one on one and meet his specific needs. But when presenting a resin line in such a way that we can compare apples to apples, the solvent choice must be carefully considered.
As an example, we will show this table of three totally different resins. Though their ARLO viscosity is similar, the analysis in these globally popular solvents varies. Resin A is a high apparent viscosity, relatively polar resin. Resin B is what we would call the conventional high apparent viscosity resin where the aliphatic solubility is relatively low. And finally, Resin C is extremely high molecular weight with good aliphatic solubility.
The solvent systems used here are very compatible with most resins in our industry. Resin A is very polar and at lower temperatures the tan delta is very low leading us to believe it is elastic. When you look at higher temperatures, the elasticity is much less (tan delta is considerably higher). The polar attractions are weaker. This resin is not necessarily a poor product. For example, in systems where "picking" can be a problem or low tack is needed while maintaining good holdout, this resin is probably the most appropriate of the three. Resin B is not comfortable with ARLO and this incompatibility can be seen by the false low tan delta in ARLO. Resin C has similar values in all solvents due to its excellent compatibility characteristics.
For your information, we will include the data from the Duke Viscometer as well as other data on these three resins in the various solvent systems in Appendix A.
Resin manufacturers are asked constantly for more structured resins. Structure according to The American Heritage Dictionary; Collegiate Edition is defined as, "Something made up of a number of parts that are held or put together in a certain way..... The interrelation of parts or the principle of organization in a complex entity."  With this definition in mind, everything resin manufacturers make is "structured."
Let us see if we can make a definition for highly structured. Since structure talks about the complexity of the polymer, we find that property most relates to apparent viscosity. In dealing with two resins of similar chemistry a more complex polymer generally exhibits higher apparent viscosity.
Remember the definition above; "...parts that are held or put together." A structured resin does not necessarily have to have high molecular weight. Inter and intramolecular bonding forces can provide structure or apparent viscosity in a system. Resins can be engineered to have excellent viscosity and yield value but molecular weight kept to a lower level to increase varnish cycle times or reduce tack.
All rosin-based resins show some degree of viscoelasticity.
"Viscoelasticity, where the deformation of the polymer specimen is reversible but time dependent and associated (as in rubber elasticity) with the distortion of polymer chains from their equilibrium conformations through activated segment motion involving rotation around chemical bonds." 
As described in the first segment of this paper, tan delta at high and low temperature appears to monitor this characteristic best in actual ink applications. Theoretically, elasticity increases with molecular weight.
The time dependency factor for recovery to an equilibrium state is important for ink formulators when monitoring dot gain and holdout properties. The ink is drying and increasing viscosity rapidly once printed. The rapid recovery of a polymer may be essential to a high quality, fast speed print job.
We can today engineer resins with gel-like properties, reducing the need for aluminum chelates. The gel structure from an "elastic" resin can better withstand high shear and temperature conditions providing sufficient rheology at the end of the press to produce high dot fidelity. Along with this characteristic, we have been able to press the resin for better compatibility with aliphatic solvents reducing the need for coresins and cosolvents, which can detract from achieving optimal lithographic performance.
3. Molecular Weight
Molecular weight is crucial to formulating resins. Consistency in building the polymer involves not only a good formula but also a good process. Companies are becoming more and more competitive, making quality a major selling tool. Our customers are asking for tighter specifications. Understanding terms like molecular weight can increase understanding between the supplier and customer in meeting his needs.
Resins used in the lithographic industry vary in molecular weight average (Mw) from as low as 2-3000 to 300,000 daltons. In this range, resins of similar molecular weight can vary in size and shape. Some resins are very large and bulky exhibiting high apparent viscosity yet not given to transfer easily on press. Other resins are very linear with long aliphatic chains to increase compatibility in ink systems.
Polarity can also influence final properties of an ink or varnish while maintaining a certain molecular weight. Inter and intra-molecular bonding influence rheology as much as molecular weight. For example, Resin A was very polar but low in molecular weight. Even so, this resin was very high in apparent viscosity in most systems.
We found that each resin must be reviewed in light of all three of these factors: structure, elasticity and molecular weight. All three characteristics affect the other but a generalization using all three terms interchangeably is not efficient for good communication between supplier and customer. If we can agree on terminology when talking about resins, we believe lab and sales time will be reduced in finding solutions to a formulator's needs.
We have started a new program called the "Stargraph" which we feel goes a long way toward this end. We have supplied typical molecular weight averages as well as other rheo-logical and solubility information. This "fingerprint" can give the formulator a tool for comparing resins of diverse properties without the influence of different compatibility in the same solvent system.
This tool is not to take the place of quality control or final ink evaluations. The Stargraph provides basic criteria to be used to correlate to application data. Combinations of these basic tests influence characteristics, such as tack or transfer.
By using this graph, a formulator can communicate more clearly to his supplier ways to possibly correct or surpass his current lithographic requirements. A detailed discussion of the "Stargraph" is in the June issue of Ink World magazine.
Communication is essential to growth. By talking and using the same terminology, we can reach new heights in resin technology for the new millennium.
The authors are indebted to the technical support supplied by the Akzo Nobel technical departments in Baxley, GA; Bergen op Zoom, Netherlands; Concordia, Argentina; and Matteson, IL.
Doug Weisel, technical specialist for Akzo Nobel, has 15 years experience in R&D, production, and sales of rosin-based resins and derivatives.
Phil Sieg, applications chemist, has been with Akzo Nobel for 10 years dealing with resin development and application. He brings an additional 10 years of ink experience from Flint Ink and The Ink Company.
(1.) R.B. Seymour and C.E. Carraher, Jr., "Polymer Chemistry: An Introduction," 2nd Edition, Marcel Dekker, Inc., New York and Basel, (1988) p. 94.
(2.) D.R. Weisel and F.P. Sieg, "Fingerprint of a Resin," Ink World Magazine, June, 1999.
(3.) R.B. Seymour and C.E. Carraher, Jr., "Polymer Chemistry: An Introduction," 2nd Edition, Merkel Dekker Inc., New York and Basel,(1988) p. 85.
(4.) H.A. Barnes, J.F. Hutton and K. Walters, "An Introduction to Rheology," theology Series Vol. 3, Elsevier, Amsterdam, (1997) p. 49.
(5.) "American Heritage Dictionary, Second College Edition," Houghton Muffin Co., Boston, (1991), p. 1208.
(6.) F.W. Billmeyer, Jr., "Textbook of Polymer Science," 3rd Edition, John Wiley & Sons, New York, (1984), p.301.
Correlation Factors to Molecular Weight Test Method Location Only Entire Database R & B Softening Point 0.1778 0.4207 1% Intrinsic Viscosity 0.9313 0.9244 1:2 ARLO Viscosity --- 0.4435 Results from the TXIB Solution Duke Viscosity @ 2.5 [sec.sup.-1] 0.3826 0.5408 Duke Viscosity @ 2500 [sec.sup.-1] 0.1059 0.3509 Duke Shortness Ratio 0.9405 0.7546 Laray Viscosity @ 2.5 [sec.sup.-1] 0.2209 --- Laray Viscosity @ 2500 [sec.sup.-1] 0.0599 --- Lemann's Shortness Ratio 0.6438 --- Carrimed Rotation Viscosity 0.3272 0.4524 Carrimed Oscillatory Results G' @ 20[degrees]C 0.4249 0.3391 G' @ 40[degrees]C 0.6185 0.1075 G' @ 20[degrees]C 0.2620 0.6035 G' @ 40[degrees]C 0.5016 0.3758 Complex Viscosity @ 20[degrees]C 0.2933 0.1378 Complex Viscosity @ 40[degrees]C 0.5213 0.3995 tan delta @ 20[degrees]C 0.9018 0.7351 tan delta @ 40[degrees]C 0.9493 0.8454 tan delta @ 20/40 0.9495 0.8638 Tan Delta Resin A 20[degrees]C 40[degrees]C 40/20 1:2 ARLO Viscosity 550 Molecular Weight [greater than]50,000 Resin B 20[degrees]C 40[degrees] 40/20 1:2 ARLO Viscosity 800 Molecular Weight 50,000-100,000 Resin C 20[degrees] 40[degrees] 40/20 1:2 ARLO Viscosity 500 Molecular Weight [greater than]200,000 40% Resin Solutions ARLO 2.43 4.84 2.01 2.35 4.29 1.82 2.51 3.00 1.19 TXIB 3.71 11.5 3.11 5.38 13.5 2.50 2.75 3.14 1.14 2:1 of 6/9:ARLO 2.38 6.35 2.67 insoluble 2.50 3.20 1.28 PWK 28/31AR 2.31 6.81 2.95 3.57 8.74 2.46 2.54 2.98 1.17 PWK 6/9AR 2.81 10.4 3.69 4.47 17.1 3.82 2.53 2.95 1.17 Solvent Comparison Solvent Boiling Range [degrees]F Aromatics % Kauri-Butanol Anailine Pt. N-40 Inko 246-302 5.0 37 66.7 PKWF 6/9 260-290 20 27 76 PKWF 6/9AR 260-290 ca. 50 ca. 50 ca. 35 PKWF 28/31AR 280-310 ca. 50 ca. 50 ca. 35 40% Resin in ARLO Resin A Resin B Resin C Duke Vicosity @ 2.5 [sec.sup.-1] 4111 10207 3048 Duke Vicosity @ 2500 [sec.sup.-1] 623 795 550 Duke Shortnes Ratio 16.51 32.11 13.85 Cammed Oscillatory G' @ 20[degrees]C 16710 27950 8510 G" @ 20[degrees]C 40690 65710 21350 G' @ 40[degrees]C 856.9 1313 957 G" @ 40[degrees]C 4160.5 5625 2886 n @ 20[degrees]C 4657 7560 2433 n @ 40[degrees]C 449.7 611.5 319.9 40% Resin in N-40 Inkol Resin A Resin B Resin C Duke Viscosity @ 2.5 [sec.sup.-1] 3468 6615 3339 Duke Viscosity @ 2500 [sec.sup.-1] 421 658 481 Duke Shortness Ratio 20.61 25.13 17.37 Cammed Oscillatory G' @ 20[degrees]C 14840 139100 9943 G" @ 20[degrees]C 31240 275400 23500 G' @ 40[degrees]C 500.4 973 649.5 G" @ 40[degrees]C 2543 7160 1913.5 n @ 20[degrees]C 3661 32670 2702 n @ 40[degrees]C 274.4 499.2 213.9 40% Resin, 40% PKWF 6/9, 20% ARLO Resin A Resin B Resin C Duke Viscosity @ 2.5 [sec.sup.-1] 1308 INSOLUBLE 1036 Duke Viscosity @ 2500 [sec.sup.-1] 222 192 Duke Shortness Ratio 14.72 13.47 Carrimed Oscillatory G' @ 20[degrees]C 4874 2793 G' @ 20[degrees]C 11600 6967 G' @ 40[degrees]C 214.6 273.1 G' @ 40[degrees]C 1356 872.6 n @ 20[degrees]C 1332 794.7 n @ 40[degrees]C 145.3 96.8 40% Resin in TXIB Resin A Resin B Resin C Duke Viscosity @ 2.5 [sec.sup.-1] 613 298 539 Duke Viscosity @ 2500 [sec.sup.-1] 194 149 119 Duke Shortness Ratio 7.89 5.01 11.28 Carimed Oscillatory G' @ 20[degrees]C 1815 650.6 1292 G' @ 20[degrees]C 6738 3501 3552 G' @ 40[degrees]C 86.6 43.8 257 G' @ 40[degrees]C 998.3 589.5 808.4 n @ 20[degrees]C 738.7 376.9 400.2 n @ 40[degrees]C 106.1 62.59 89.81 40% Resin in PKWF 6/9AR Resin A Resin B Resin C Duke Viscosity @ 2.5 [sec.sup.-1] 557 396 444 Duke Viscosity @ 2500 [sec.sup.-1] 114 214 88 Duke Shortness Ratio 12.20 4.62 12.55 Carrimed Oscillatory G' @ 20[degrees]C 1674 1290 1223 G' @ 20[degrees]C 4700 5764 3087 G' @ 40[degrees]C 67.0 24.46 180 G' @ 40[degrees]C 694.1 417.7 532.4 n @ 20[degrees]C 528.2 625.3 351.5 n @ 40[degrees]C 73.8 44.3 59.51 40% Resin, 40% PKWF 28/31AR Resin A Resin B Resin C Duke Viscosity @ 2.5 [sec.sup.-1] 1233 1098 1001 Duke Viscosity @ 2500 [sec.sup.-1] 226 427 213 Duke Shortness Ratio 13.66 6.42 11.73 Carrimed Oscillatory G' @ 20[degrees]C 14840 4862 2447 G" @ 20[degrees]C 31240 17370 6233 G' @ 40[degrees]C 500.4 111.8 316.4 G" @ 40[degrees]C 2543 976.2 941.4 n @ 20[degrees]C 1294 1910 708.8 n @ 40[degrees]C 142.2 104 105.1 Acid Value Softening Point, R&B 1% Intrinsic Value Methanol Number Alcohol Number 1:2 ARLO Viscosity Magiesol 47 Dilution Carrimed Rotational Viscosity Laray Viscosity Carrimed Oscillatory Measurements Duke Viscosity GPC
|Printer friendly Cite/link Email Feedback|
|Author:||Weisel, Doug; Sieg, Phil|
|Date:||Jan 1, 2000|
|Previous Article:||CPMA's Annual Holiday Meeting is Major Success.|
|Next Article:||China Coat 99.|