Printer Friendly

Automated parallel polyurethane dispersion synthesis and characterization.

Abstract The synthesis of waterborne polyurethane dispersions (PUDs) using an automated parallel reactor system was explore. Waterborne PUDs are an important class of polymer dispersion that can be used in many applications such as coatings for wood finishing, glass fiber sizing, adhesives, automotive topcoats, and other applications. Herein, we present the synthesis of aqueous PUDs using a Chemspeed Autoplant A 100 [TM] automated parallel reactor system. This is the first time a PUD has been synthesized using an automated parallel reactor system. The synthesis involves the formation of an isocyanate-terminated prepolymer followed by neutralization, dispersion in water, and chain extension. Details of the methodology are discussed with respect to the process of writing the program for the synthesis to synthesizing the PUD itself with the Chemspeed. It is demonstrated that an aqueous PUD can be synthesized with an automated parallel process and the unit-to-unit results are similar. Process variables such as agitator design, rate of neutralization, and rate of water dispersion are varied as these are the three major factors which lead to the desired end product property. The controlled addition of neutralizer, water, and chain extender is an added advantage with this automated technique and gave consistent results in all the units. The PUDs were characterized for their particle size, viscosity, and percent solids.

Keywords Polyurethane, Dispersion, Aqueous, High throughput


Polyurethane dispersions (PUDs) are widely used waterborne binder systems in the coating industry.(1-3) Research in the area of PUDs has already spanned many decades, and the uses of PUDs are numerous. (3-6) Due to the high-performance characteristics of polyurethanes, extensive research efforts continue to be carried out in order to expand knowledge of their structure-property relationships. (6-10) Waterborne PUDs are an important class of polymer dispersion that can be used in many applications such as coatings for wood finishing, glass fiber sizing, adhesives, automotive topcoats, and other applications (3), (4), (8), (10-14) A waterborne coating contains water as the major "solvent" or carrier liquid and may be applied to solution-like waterborne systems, dispersions, emulsions, latex, and mixed systems. In all cases, the polymer system is appropriately designed so that it can form a stable suspension or dispersion in water. Most modifications give the polymer a more polar nature so that it is compatible with water; thus, the polymers are more water sensitive. The constant need for reductions in both volatile organic compound emissions and production costs has stimulated extensive research activity on the development of water-based polymer systems, particularly for coating and adhesive formulations. (5), (7), (8) The development of water-borne coatings was initially driven by ease of use, cost pressures, fire-hazard reduction, and solvent handling problems, but is now driven by environmental pressures as well. (5), (7), (8) Some important issues in waterborne coatings systems lie in developing truly solvent-free systems, eliminating water-sensitivity inherent in water-borne systems, and reducing foaming and corrosion in application. This is because waterborne systems are very sensitive to the factors involved during the development of the coatings and during the application. (2), (4) Some of the advantages of waterborne PUDs are low VOC, minimal toxicity and environmental issues, and freedom from flammability and explosion hazards. Another important structural and functional characteristic of waterborne PUDs is the presence of polar ionic groups (typically carboxylate or sulphonate) in the polymer chain. These groups are necessary for the formation of dispersions, because they act as internal surfactants. Dimethylolpropionic acid is often used to provide carboxylate functional group and provide good hydrophilic character to the polyurethane polymer. (9), (15-18)

Combinatorial and high-throughput methods are an approach to speed up the preparation and exploration of new polymeric materials: a large diversity of parameters can be screened rapidly resulting in the determination of new structure/property relationships. (19) During the last decade, the use of combinatorial and high-throughput approaches has evolved from its infancy into well-established tools to accelerate the discovery and development of new materials. (20-23) At present, combinatorial and high-throughput methods are finding applications beyond the pharmaceutical industry for discovery of materials in chemistry and materials science. (20), (23-26) These materials include catalysts, luminescent and magnetoresistive compounds, polymers, high-temperature superconductors, and many others. (27) The field of polymer research is an ideal area for parallel and combinatorial experiments due to the fact that many parameters--both composition and process variables--can be varied during synthesis. (20-24), (28-30) In the case of PUD synthesis, a number of parameters can be varied, such as the composition of diisocyanates and polyols used, stirring rate, heating and cooling, prepolymer molecular weight, neutralizing amines, chain extenders used, and other reaction and process conditions. Synthesizing a PUD using an automated parallel method has a number of advantages. The first and foremost advantage is that when using a computer-controlled, automated reactor system, every batch is carried out using the same set of process steps and conditions, avoiding human variability. In addition, using a parallel reactor system, a large number of process and compositional variable can be investigated in one experiment or in just a couple of experiments. These experiments can result in the generation of highly systematic data where the experimental results can be directly related to the input variables. Previous reports in the literature on PUD synthesis have used traditional laboratory methods. In this article, we will describe the automated high-throughput approach and the results in comparison with the traditional method.

The schematic of PUD synthesis is given in Scheme 1. The synthesis involves formation of a polyurethane prepolymer followed by neutralization, dispersion in water, and chain extension to obtain the PUD. An automated method using a Chemspeed Autoplant A100 parallel reactor system was adapted to carry out the synthesis, and to our knowledge this is the first time a PUD has been synthesized by an automated method at a laboratory scale. The results obtained by the automated method using the Chemspeed reactor were compared with the traditional laboratory synthesis method.




3-Hydroxy-2-(hydroxymethyl)-2-methyl-propanoic acid (Dimethylolpropionic acid, DMPA) was supplied by Perstorp. N-Methyl-2-pyrrolidone (NMP, Sigma-Aldrich, 99.5%), dibutyltin dilaurate (DBTDL, Aldrich, 95%), triethylamine (TEA, Aldrich, 99.5%), and ethylenediamine (EDA, Aldrich, 99%) were used without further purification. Linear polyester polyol (Polyester PE 170HNA) and dicyclohexylmethane diisocyanate (1, 1'-methylenebis [4-isocyanatocyclohexane]) were obtained from Bayer MaterialScience. DI water from the laboratory was used for the dispersion of the polyurethane.


Viscosity measurements were done using Brookfield DV-I+ viscometer and the viscosity was measured at room temperature around 23[degrees]C with spindle S63 (small sample adapter) at 100 rpm. The particle size was measured on water-diluted samples using Submicron Particle Sizer, NICOMP [TM] 380 and both Gaussian and Nicomp methods were used. An automated surface energy measurement unit manufactured by Symyx Discovery Tools, Inc and First Ten Angstroms was used to measure the surface energy of PUD coatings. Droplets of water and methylene iodide (MI) were deposited on the PUD coating separately and a CCD camera imaged the droplets and then automated image analysis was used to determine the contact angles. Three droplets of water and MI were used for each measurement. Surface energy was calculated from the contact angle data using the Owens-Wendt equation.

Laboratory PUD synthesis

A general synthesis procedure was followed for synthesis using the laboratory method. A three neck round bottom flask was fitted with an agitator, nitrogen inlet, and water condenser. Heating was with an oil bath on a hotplate. The polyol, DMPA, diisocyanate, and NMP are charged and stirred until the mixture becomes homogenous. Then DBTDL catalyst is added, the mixture is heated to 90[degrees]C, and the temperature is controlled so that it does not go above 95[degrees]C to avoid side reactions. When the theoretical NCO content is reached, the mixture is cooled to 70[degrees]C and TEA is added with increased agitation. The dispersing water is at room temperature and it is added under high agitation, 250-350 rpm, over a 25 min period. Then, EDA and water are combined at ambient temperature, ~25[degrees]C, and added to the reaction mixture drop-wise over 10-15 min with agitation of 150-200 rpm. The agitation is continued for another 2 h to complete the water reaction with residual isocyanate and form the dispersion.

Automated PUD synthesis

Automated PUD synthesis was done using a Chem-speed Autoplant A100 [TM]. Details of the Autoplant are given below and the reactor system is shown in Fig. 1. (23), (31) The system used in this study consists of 12 Process Development (PD) units. (The system can be fitted with up to 20 PD units.) Each PD unit contains two 100 mL stainless steel reactors having mechanical stirring up to 600 rpm and reflux cooling. The temperature in each reactor can be controlled independently over a range of -10 to 250[degrees]C. Solid and liquid reagents can be automatically charged to each reactor. In this experiment, the system is setup to operate in semi-continuous mode. Thus, one of the 100 mL vessels in the PD unit is designated as the reactor and the other as the feed vessel. A third 50 mL feed vessel is also present in each PD unit. The reactor vessel is charged with reagents and the other feed vessels are charged with the catalyst and solvent. The contents of the two feed vessels are then fed slowly to the reactor using two syringe pumps on the PD unit. In this mode, up to 12 reactions can be conducted simultaneously and both compositional variations as well as process variations can be explored. Several different inline real-time data analysis factors, such as temperature and stirring, can be monitored for all of the 12 PD units. Gas supply for inertization using nitrogen can be done up to 20 bar working pressure. The synthesizer is equipped with a four-needle liquid handling system with four syringes (1-10 mL) to accurately dispense different volumes of stock solutions and samples.


In a typical experimental setup, the 100 mL reactor vessel is manually charged with the polyol and DMPA, then, using the four-needle head, diisocyanate in NMP and DBTDL catalyst diluted with NMP are added, the reactor is sealed, and the mixture is heated slowly to the reaction temperature with continuous stirring. The 100 mL feed vessel is charged with water and the third feed vessel is charged with TEA. Once the required free NCO level is reached, TEA, water, and EDA-water mixture are added to the reactor vessel at a constant rate from the feed vessels with the stirring rates of 400, 600, and 200 rpm, respectively, during addition. After all of the additions are over, the reaction is stirred for two more hours for completion of the synthesis. In order to do all of the above process steps, a program is written as outlined with the flowchart in Fig. 2. Once the program is written, this protocol can be repeated as desired and also readily modified for different process conditions.


Results and discussion

The synthesis and determination of structure--property relationships in multi-component polymer systems can be a tedious process consisting of the exploration of process variables such as temperature, stirring rates, and rates of addition of reagents. Often, the magnitude of the effort required confines the experimenter to a relatively small number of experiments within a limited range of variables. The experimenter tends to be more concerned with how many experiments are required and the time involved than with determining how much information can be gained from the set of experiments, even with the use of statistical design of experiments methods. The preparation of PUDs using a high-throughput approach is of great interest due to the presence of a large number of compositional and process variables which can greatly affect the properties of the final product. The systematic evaluation of compositional variables (different diisocyanates and polyols, NCO percentage, different amines, catalyst, solvent) and process variables (temperature, heating and cooling, types of stirrers and other reaction and process conditions, addition of reagents) is a challenging task, but can be made highly efficient using an automated reactor system. Programming the synthesis process requires a translation step from the standard laboratory procedure to a process suitable for the automated parallel system. A representative step-by-step polymerization procedure might contain tasks like initiating reflux and stirring, dispensing reagents in the appropriate amounts, setting reaction temperature, sampling, rinsing the four-needle liquid handling system to avoid cross contamination, and shutting down all equipment when finished. Hence, each single operation performed during the polymerization process has to be identified and listed in a flow scheme before programming the synthesis; a schematic of the process steps used for PUD synthesis is shown in Fig. 2.

To determine feasibility, initial PUD syntheses were performed by two methods: two were done using a conventional laboratory method and another was carried out using a single unit of the automated reactor system. As shown in Table 1, PUDs A and C were synthesized using the conventional laboratory method, whereas PUD-B was synthesized using the automated reactor. Viscosity, pH, particle size, resin solids, contact angle, and surface energy were measured for all the PUDs synthesized and the results are given in Table 1. PUDs A, B, and C gave consistent particle size both before and after filtration. Viscosity measurements indicated that PUD-C had a very high viscosity, whereas PUD-A and B were consistent with each other. All three PUDs are stable and uniformly dispersed as shown in Fig. 3; the PUDs synthesized using the conventional laboratory method were translucent. Contact angles and surface energy of films were measured and the values are consistent among both methods of synthesis. There are differences between PUDs A, B, and C from the experimental results obtained, with the primary differences being between the two PUDs synthesized using the conventional laboratory method. Using an automated method for PUD synthesis, the differences in properties caused by process variations can potentially be minimized or eliminated.

Table 1: Comparison of PUD synthesized by mechanical stirrer and
automated method

PUD                          A              B               C

Method of synthesis     Mechanical      Chemspeed       Mechanical
                        stirrer                         stirrer

Quantity (g)            500             50              500

% NCO/Time (h)          4.10/3.30       -/2.30          3.46/3.30

pH                      10              8.5             9.5

Viscosity (mPa.s)       40              20              520

Particle size (a) (nm)  Gaussian = 112  Gaussian = 218  Gaussian = 154

                        Nicomp = 114    Nicomp = 217    Nicomp = 119

Resin solids (%)        43              45              36

Water contact angle     78              76              81

MI contact angle        27              22              23

Surface energy          51              53              51

(a) Measurements were done after filtering the solution with 5 [micro]m

The recent literature on automated methods of material synthesis and characterization has demonstrated that it is possible for the good control of the various compositional and process parameters. (19-25), (27), (29) This initial result indicated that a PUD can be made using the automated reactor system; however, the synthesis of libraries of PUDs using an automated reactor system has not yet been reported. In order to understand the complex and sensitive PUD synthesis process, seven libraries of PUDs were made and studied thoroughly with different process variables.

Obtaining a good and stable PUD is influenced by various process parameters such as the geometry of the stirrer and reactor used, rate of addition of the reagents, and effective control over temperature and stirring. Using a high-throughput approach the effect of these variables can be studied in parallel in a couple of experiments. The parameters studied in the present study include the number of PD units, quantity of PUD synthesized, types of the stirrers used, temperature of the prepolymer synthesis, time required for TEA addition after the prepolymer is formed, rate of stirring, and time for the formation of dispersion. Some of the variables were kept constant for all the libraries studied. Table 2 shows the details of all of the experiments conducted with all of the parameter values.
Table 2: Variable studied for the synthesis of PUD by automated method

PUD samples     PUD-D        PUD-E          PUD-F          PUD-G

Number of    12             2              12             2

Quantity     60             60             60             47

Type of      A-1            A-1            A-2 & P        A-1

Stir rate    200            200            200            200

Temperature  90 [+ or -] 3  90 [+ or -] 3  90 [+ or -] 3  85 [+ or -] 3

Time for     120            120            120            120

Stir rate    400            400            400            400
before TEA

Temperature  70 [+ or -] 3  70 [+ or -] 3  70 [+ or -] 3  70 [+ or -] 3
before TEA

Time for     5              10             5              5

Stir rate    600            600            600            500/600

Temperature  70 [+ or -] 3  70 [+ or -] 3  70 [+ or -] 3  70 [+ or -] 3

Temperature  35 [+ or -] 3  35 [+ or -] 3  35 [+ or -] 3  35 [+ or -] 3
set for

Time for     40             80             40             40

Stir rate    200            200            200            200

Temperature  35 [+ or -] 3  35 [+ or -] 3  35 [+ or -] 3  35 [+ or -] 3

Time for     20             20             20             20

Stir time    120            120            120            120
for water-

Good         9/12           1/2            3/6, 2/6       1/2

  PUD samples/variables       PUD-H          PUD-I           PUD-J

Number of PD units         12             12             12

Quantity (g)               47             47             47

Type of stirrer used       A-1            A-1            A-1

Stir rate for prepolymer   200            200            200

Temperature for            85 [+ or -] 3  85 [+ or -] 3  85 [+ or -] 3
prepolymer ([degrees]C)

Time for prepolymer        120            120 (a)        120 (a)
synthesis (min)

Stir rate before TEA       400            400            400
addition (rpm)

Temperature before TEA     70 [+ or -] 3  70 [+ or -] 3  70 [+ or -] 3
addition ([degrees]C)

Time for TEA addition      5              15             15

Stir rate during water     600            600            600
addition (rpm)

Temperature before water   70 [+ or -] 3  70 [+ or -] 3  70 [+ or -] 3
addition ([degrees]C)

Temperature set for water  35 [+ or -] 3  35 [+ or -] 3  35 [+ or -] 3
addition ([degrees]C)

Time for water addition    60             40             40

Stir rate before           200            200            200
water-EDA addition (rpm)

Temperature before         35 [+ or -] 3  35 [+ or -] 3  35 [+ or -] 3
water-EDA addition

Time for water-EDA         20             15             15
addition (min)

Stir time for water-EDA    120            120            120
addition (min)

Good translucent/opaque    5/12           12/12          12/12

(a) After 60 min catalyst (DBTDL-NMP) was added

To start with, three types of stirrers, namely anchor (A-1 and A-2) and propeller (P) types, were used in this study. Table 3 shows the stirrers used and gives their dimensions. The usage of the stirrers for the various libraries of PUDs synthesized is given in Table 2. The 100 mL vessel used for synthesis of the PUD is shown in Fig. 4. In most of the cases, stirrer type A-1 was used; this gave the best performance and a high number of successful PUDs. This can be attributed to the fact that the reactor inner diameter (4.7 cm) and the A-1 type stirrer diameter (4.5 cm) configurations are close, placing the blade of the anchor stirrer close to the reactor wall. The closer the stirrer is to the wall of the reactor is believed to provide for more complete mixing of all the reagents while the PUD is synthesized. This is also very important when the dispersing water is added. With the other A-2 and P-type stirrers, the effective mixing and dispersing of the polyurethane becomes difficult because of the geometry of the stirrers. As given in Table 2 for the library PUD-F, the number of good PUDs formed is lower and this is attributed to the type of the stirrer used. At the end of the synthesis, in some cases the reactor had only paste/cake of polyurethane present or solidification of the urethane polymer around the stirrer was observed. This is indicative of less effective mixing and poor dispersing of the polyurethane in the water dispersion step.



Diisocyanates have highly reactive NCO groups and are susceptible to other side reactions. Hence, to understand how much time is required to achieve a predetermined value of free NCO, separate experiments were done at 90[degrees]C using all the reactors, removing samples periodically, and it was found that in 2 h a value of less than the target value of 3.5% NCO was achieved. The percentage NCO in the polyurethane prepolymer was determined by titration using the method in ASTM D2572-97. For all the libraries, the polyurethane prepolymer step was carried out by heating and stirring for 2 h. In these experiments, the temperature was set to either 90 or 85[degrees]C to determine the effect of temperature. For both temperatures, excellent results were obtained except for libraries PUD-F and PUD-H, for which other parameters appeared to play a major role and hence the number of good PUDs obtained was lower.

Neutralization of the polyurethane prepolymer formed using TEA is an important step for the formation of a good PUD, as this step converts the carboxylic acid groups to carboxylate anions needed on the surface of the particles for the reversal of the phase--i.e., hydrophobic to hydrophilic--during the addition of dispersing water. This process is necessary for the formation of a stable dispersion in water; the final particle size of the dispersion depends on both neutralization and agitation. (2), (9), (15), (16) For all the libraries the controlled addition of TEA was done over a period of 5 or 10 min after the temperature reached 70[degrees]C while stirring at 400 rpm. Changing the rate of addition of TEA in one of the libraries, PUD-E, did not have a significant effect on the number of good PUDs formed.

Due to the viscosity of the prepolymer, it is not possible to form the dispersion by pumping the prepolymer into the water using the syringe pumps and narrow-bore tubing on the PD units. Thus, the dispersion is formed by adding water to the neutralized prepolymer. After the TEA addition, the stirring rate was increased, the temperature was set to ramp from 70 to 35[degrees]C, and the water addition was initiated. The time for the water addition coincided with the time for reducing the temperature to 35[degrees]C, as given in Table 2. This step is a critical step as this determines whether a good PUD is formed or not. In almost all the libraries studied the stirring rate was set to 600 rpm and in one library, PUD-G, both 500 and 600 rpm were used, and the former failed during the addition of dispersing water. As shown in Table 2, the rate of water addition was varied in the libraries of PUDs synthesized. Forty minutes seems to be the best suited rate of addition of dispersing water. Decreasing the rate of addition of dispersing water leads to a lower number of good PUDs formed and this is also dependent on the type of stirrer used.

After the preset time all the dispersing water is added and the PUD is now ready for chain extension. Chain extension was done using an EDA-water mixture and the addition was done in 20 min from the feed vessel using the syringe pump. The PUD was stirred for 2 h at 35[degrees]C for the completion of the chain extension reaction.

Understanding what is happening inside a reactor, such as viscosity, temperature, stirring, and the rate of addition of reagents, may be useful during the PUD synthesis. Visual observation is not possible since the reactors are made of stainless steel. But important information can be obtained using the Chemspeed software, which can make real-time plots of stirring and temperature with time, and amount of reagents transferred from feed vessel to the reactor for all the individual PD units at any instant during the synthesis procedure. The data are also stored in log files for analysis after completion of a reaction run. Log file plots of stirring rate and current draw for a typical run are shown in Fig. 5. The current draw is proportional to the viscosity of the reactor contents. From these plots the viscosity at each of the stages of PUD synthesis can be observed, providing a snapshot of what is occurring in the reactor during the different stages of the synthesis process.


Characterization of the PUDs synthesized for all the libraries was done and the results for the library PUD-D is shown in Table 4. Characterization results for other libraries are not presented here but the number of good PUDs formed in each library is presented in Table 2. Good PUDs were defined as those that formed satisfactory dispersions, with minimal amounts of coagulum remaining in the reactor vessel or on the stirrer. For the library PUD-D, the results are tabulated in the ascending order of good PUDs formed in each of the 12 PD units of the Chemspeed. In PD unit 7, the PUD was almost gelled and hence complete characterization was not done. In the case of PD units 1 and 3, a low amount of usable PUD was formed as a result of solid formation around the stirrer, but the properties were similar for both. PUDs formed in PD units 2, 4, 5, 6, 9, and 11 gave almost similar results and all of them were translucent. Similarly, the PUDs formed in PD units 8, 10, and 12 were milky and appeared as stable dispersions. Figure 6 shows the PUDs from all of the PD units from library PUD-D. One can note that characterization values for PUDs formed in the PD units 2, 4, 5, 6, 9, 11, 8, 10, and 12 for the properties measured are almost similar with the exception of the particle size. Obviously the milky dispersions have larger particle size and the translucent are smaller in particle size. Since PUDs are very sensitive materials, these values are acceptable for various applications. The number of good PUDs formed in each of the libraries is given in Table 2; excellent results were obtained for all the libraries studied upon varying the different parameters. Drawdowns of all of the good PUDs were made on glass microscope slides and, after drying, all of the PUDs were clear and free of grit.

Table 4: Characterization of PUDs synthesized by Chemspeed with all the
PD units for library PUD-D

PD unit      Key        Particle               %      Water      Ml
         observations     size               Resin   contact  contact
           after PUD      (nm)               solids   angle    angle

                        Gaussian  Nicomp

7        Completely        660    239, 808     5         -       27
         very less PUD

1        Become solid      160    116, 402    28        70       28
         with less PUD

3        Become            143    134         31        71       31
         solid with
         less PUD and

2        Good,              85    82          37        72       26

4        Good,              84    92          38        71       29

5        Good,             109    110         34        69       30

6        Good,              87    92          37        71       30

9        Good,              98    100         38        69       28

11       Good,              96    99          38        69       25

8        Good,             120    127         38        70       27

10       Good, opaque      292    292         38        70       28

12       Good, opaque      157    160         38        72       26

PD          Key  observations           Surface energy  Viscosity
unit       after PUD synthesis              (mN/m)       (mPa.s)

7     Completely gelled, very less PUD        79            -

1     Become solid with                       51           498
      less PUD and translucent

3     Become solid with less                  50           216
      PUD and translucent

2     Good, translucent                       51            58

4     Good, translucent                       50            70

5     Good, translucent                       51            91

6     Good, translucent                       50            74

9     Good, translucent                       52           111

11    Good, translucent                       52            78

8     Good, opaque                            51            22

10    Good, opaque                            51            18

12    Good, opaque                            51            15

After exploring a number of the variables in the first few PUD libraries, and selecting the best process conditions, two more libraries, PUD I and PUD J, were made under identical process conditions to make sure that all the PUDs formed were consistent in all 12 PD units. The three important variables such as type of stirrer, rate of addition of water, and effective stirring were fixed as given in Table 2 for PUD I and PUD J. These libraries were done on two separate days and all 24 of the PUDs formed were characterized. All of the PUDs were successfully produced with properties that are acceptable. For the particle size data the average value is 156 [+ or -] 47 nm with a coefficient of variation of 30%. This represents very good run-to-run variability for PUD synthesis. The viscosity data has an average value of 229 [+ or -] 156 mPa.s with a coefficient of variation of 68%. While high, this variation is still well within acceptable limits for PUDs.

Hence, with this automated approach of PUD synthesis all the PUDs formed are not only consistent within one parallel synthesis run, but also consistent over two runs done on two different days. The three important parameters found in this study are type of stirrer used, rate of addition of dispersing water, and effective stirring. Since all of the syntheses were done using the Chemspeed, all of the experiments were done in a short period of time, giving a significant amount of valuable information regarding the impact of several process variables on PUD synthesis. Now that this preliminary study has established baseline processing parameters for the synthesis of this model PUD, it is possible to systematically explore both the effects of compositional variables and process variables on the properties of the PUDs synthesized.


PUD synthesis was performed using an automated synthesis system successfully for the first time. The results are comparable using both a traditional laboratory and the automated Chemspeed methods; with the automated method the results are more consistent. A series of PUD synthesis experiments were carried out using various process parameters to optimize the procedure used in the automated reactor system. The use of A-1 type anchor stirrer (closer to the wall of the reactor) is preferred as this helps for the complete mixing of all the polyurethane during the addition of dispersing water. The rate of water addition determines the final PUD properties and we found that a shorter time is better. This investigation provided some useful indications on how to optimize the properties of aqueous PUDs and gave the most promising product development process in terms of particle size of the dispersions and physical properties of the final product.

Acknowledgments The authors thank Bayer MaterialScience and the Office of Naval Research for supporting this research under Grants N00014-04-1-0597 and N00014-05-1-0822, and Dr. Olaf Kohler, Chemspeed, for his help and support on doing the synthesis of library PUD-B. The authors also thank students Kaley Ward, Vishal Sonalkar, and Robert Hoshaw for measuring the contact angles, surface energy, and particle size.


(1.) Barni, A, Levi, M, "Aqueous Polyurethane Dispersions: A Comparative Study of Polymerization Processes." J. Appl. Polym. Sci., 88 (3) 716-723 (2003)

(2.) Keyvani, M, "Improved Polyurethane Dispersion Stability via Continuous Process." Adv. Polym. Technol., 22 (3) 218-224 (2003)

(3.) Satguru, R, McMahon, J, Padget, JC, Coogan, RG, "Aqueous Polyurethanes-Polymer Colloids with Unusual Colloidal, Morphological, and Application Characteristics." J. Coat. Technol., 66 (830) 47-55 (1994)

(4.) Dieterich, D, "Aqueous Emulsions, Dispersions and Solutions of Polyurethanes; Synthesis and Properties." Prog. Org. Coat 9 (3) 281-340 (1981)

(5.) Hourston, DJ. Williams, GD, Satguru, R, Padget, JC, Pears, D, "The Influence of the Degree of Neutralization, the Ionic Moiety, and the Counterion on Water-Dispersible Polyurethanes." J. Appl. Polym. Sci., 74 (3) 556-566 (1999)

(6.) Asif, A, Shi, W, Shen, X, Nie, K, "Physical and Thermal Properties of UV Curable Waterborne Polyurethane Dispersions Incorporating Hyperbranched Aliphatic Polyester of Varying Generation Number." Polymer, 46 (24) 11066-11078 (2005)

(7.) Durrieu, V, Gandini, A, Belgacem, MN, Blayo, A, Eisele, G,Putaux, J-L, "Preparation of Aqueous Anionic Poly-(Urethane-Urea) Dispersions: Influence of the Nature and Proportion of the Urethane Groups on the Dispersion and Polymer Properties." J. Appl. Polym. Sci., 94 (2) 700-710 (2004)

(8.) Delpech, MC, Coutinho. FMB, "Waterborne Anionic Polyurethanes and Poly(Urethane-Urea)s: Influence of the Chain Extender on Mechanical and Adhesive Properties." Polym. Test., 19 (8) 939-952 (2000)

(9.) Madbouly, SA, Otaigbe, JU, Nanda, AK, Wicks, DA. "Rheological Behavior of Aqueous Polyurethane Dispersions: Effects of Solid Content, Degree of Neutralization, Chain Extension, and Temperature." Macromolecules, 38 (9) 4014-4023 (2005)

(10.) Nanda, AK, Wicks. DA, Madbouly, SA, Otaigbe, JU. "Nanostrueturcd Polyurethane/POSS Hybrid Aqueous Dispersions Prepared by Homogeneous Solution Polymerization." Macromolecules, 39 (20) 7037-7043 (2006)

(11.) Sundar, S, Aruna, P. Venkateshwarlu, U, Radhakrishan, G, "Aqueous Dispersions of Polyurethane Cationomers: A New Approach for Hydrophobic Modification and Crosslinking." Colloid Polym. Sci., 283 (2) 209-218 (2004)

(12.) Sundar, S, Vijayalakshmi, N, Gupta, S, Rajaram, R, Radhakrishan, G, "Aqueous Dispersion of Polyurethane-Polyvinyl Pyridine Cationomers and their Applications as Binder in Base Coat for Leather Finishing." Prog. Org. Coat., 56 (2-3) 178-184 (2006)

(13.) Anzlovar, A, Zigon, M, "Semi-Interpenetrating Polymer Networks Based on Polyurethane, Polymethacrylate Functions Prepolymers: Morphology and Mechanical Properties in Dependence of the Concentration of Function Groups." J. Polym. ScL B Polym. Phys.. 40 (1) 115-123 (2001)

(14.) Nishio, A, Mochizuki. A. Sygivama. J-I. Takeuchi, K, Asai, M, Yonetake, K, Ueda, M, "Synthesis of Head-to-Tail Polyurethanes from Nonsymmetric Diisocyanate, Ethylene Glycol with Organotin Catalysts." J. Polym. Sci. A Polym. Chem., 39 (3) 416-429 (2001)

(15.) Kim, BK, "Aqueous polyurethane dispersions." Colloid Polym. Sci, 274 (7) 599-611 (1996)

(16.) Visser, SA, Cooper, SL, "Comparison of the Physical Properties of Carboxylated and Sulfonated Model Plyurethance Ionmers," Macromolecules, 24 (9) 2576-2583 (1991)

(17.) Kim, BK, Lee, JC, "Waterborne polyurethanes, their properties." J. Polym, Sci. A Polym. Chem., 34 (6) 1095-1104 (1996)

(18.) Nanda, AK, Wicks, DA, Madbouly, SA, Otaigbe, JU, "Effect of Ionic Content, Solid Content, Degree of Neutralization, and Chain Extension on Aqueous Polyurethane Dispersions Prepared by Prepolymer Method." J. Appl. Polym. Sci., 98 (6) 2514-2520 (2005)

(19.) Iden, R, Schrof, W, Hadeler, J, Lehamann, S. ""Combinatorial Materials Research in the Polymer Industry: Speed Versus Flexibility." Macromol. Rapid Commun.. 24 (1) 63-72 (2003)

(20.) Webster, DC, Bennett, J, Kuebler, S, Kossuth, MB, Jonasdottir, S, "High Throughput Workflow for the Development of Coatings." J. Coat. Technol., 1 (6) 34-39 (2004)

(21.) Hoogenboom, R, Meier, MAR, Schubert, US. "Combinatorial Methods, Automated Synthesis and High-Throughput Screening in Polymer Research: Past and Present." Macromol. Rapid Commun., 24 (1) 15-32 (2003)

(22.) Potyrailo, RA, Olson, DR, Medford, G, Brennan, MJ, Development of Combinatorial Chemistry Methods for Coatings: High-Throughput Optimization of Curing Param eters of Coatings Libraries." Anal. Chem., 74 (21) 5676-5680 (2002)

(23.) Nasrullah, MJ, Webster, DC, "Polymerization of Styrene and t-butyl Acrylate by Atom Transfer Radical Polymerization--high Throughput Approach." Polym. Prepr., 47 (1) 217-218 (2006)

(24.) Webster, DC, "Radical Change in Research and Development: The Shift from Conventional Methods to High Throughout Methods." J. Coat. Technol., 2 (15) 24-29 (2005)

(25.) Ekin, A, Webster, DC, "Library Synthesis, Characterization of 3-Aminopropyl-Terminated Poly (Dimethylsiloxane)s and Poly(e-Caprolactone)-b-Poly(Dimelhylsiloxane)s."J. Polym. Sci. A Polym. Chem., 44 (16) 4880-4894 (2006)

(26.) Song, R, Chiang, MYM, Crosby, AJ, Karim, A. Amis, EJ, Eidelman, N, "Combinatorial Peel Tests for the Characterization of Adhesion Behavior of Polymeric Films." Polymer, 46 (5) 1643-1652 (2005)

(27.) Jandeleit, B, Schaefer, DJ, Powers, TS, Turner, HW, Weinberg, WH, "Combinatorial Materials Science and Catalysis." Angew. Chem. Int. Ed., 38 (17) 2494-2532 (1999)

(28.) Potyrailo, RA Chisholm, BJ, Olson, DR, Brennan. MJ, Molaison, CA, "Development of Combinatorial Chemistry Methods for Coatings: High-Throughput Screening of Abrasion Resistance of Coatings Libraries." Anal. Chem,. 74 (19) 5105-5111 (2002)

(29.) Zhang, H, Hoogenboom, R, Meier, MAR. Schubert, US. "Combinatorial and High-Throughput Approaches in Polymer Science." Meas. Sci. Technol., 16 (1) 203-211 (2005)

(30.) Anderson, DG, Tweedie, CA, Hossain, N, Navarro. SM, Brey, DM, Van Vliet, KJ, Langer, R, Burdick, JA. "A Combinatorial Library of Photocrosslinkable and Degradable Materials." Adv. Mater., 18 (19) 2614-2618 (2006)


M. J. Nasrullah, J. A. Bahr, C. Gallagher-Lein Center for Nanoscale Science and Engineering, North Dakota State University. Fargo, ND, USA

D.C. Webster

Department of Coatings and Polymeric Materials,

North Dakota State University, 1735 NDSU Research Park

Drive, Fargo, ND 58102, USA


R. R. Roesler, P. Schmitt

Bayer MaterialScience LLC, Pittsburgh, PA, USA
COPYRIGHT 2009 American Coatings Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nasrullah, Mohammed J.; Bahr, James A.; Gallagher-Lein, Christy; Webster, Dean C.; Roesler, Richard
Publication:JCT Research
Date:Mar 1, 2009
Previous Article:Shear-induced 1-D alignment of alumina nanoparticles in coatings.
Next Article:Synthesis and evaluation of functional polymeric additives with urethane linkage as dispersing agents for preparation of pigment concentrates with...

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters