Studies on the effect of structural parameters on the properties of polysiloxaneurethane dispersions and coatings J Kozakiewicz, 1 A Koncka-Foland, l J Skarzynski, 1 J W Sobczak 2 and M Zielecka 1 1 Industrial Chemistry Research Institute, 8 Rydygiera St, 01-7 93
Warsaw, Poland 2 Polish Academy of Sciences, Institure of Physical Chemistry
Keywords Polysiloxaneurethanes, aqueous dispersions, factorial experirr-ent, polysiloxaneurethane-acrylic/ styrene dispersions, coatings
Summaries Studies on the effect of structural parameters on the properties of polysiloxaneurethane dispersions and coatings The results of studies on the effect of introducing double bonds and siloxane segments to the p@urethaneurea chain on the properties of polyurethane and polyurethane-acrylic/slTrene dispersions and coatings are presented In order to investigate that effect, a 23 factorial experiment was designed and conducted Several properties of dispersions, coatings and films were determined, including,/ntera//a, particle size, stability, and MFFT (minimum film forming temperatures) of dis persions, hardness, resistance to water and solvents, adhesion of coatings and mechanical proper ties and the glass-transition temperature (T) of films Properties of coatings and films subjected to additional curing were also investigated. Analysing the results of the factorial experiment helped to better understand the mutual effect of several factors on the properties of the investigated materials. ESCA (electron spectroscopy for chemical analysis) and DCA (dynamic contact angle) studies of the surface properties of the coatings were also carried out. '.
.
.
.
.
g
,
I;.tudes sur I'effet des parametres de structure sur les proprietes des dispersions et des revetements de polysiloxaneurethane Les resultats des etudes sur I'effet sur les propriet@sdes dispersions et des revOtementspolyurethane et polyurethane-acryylique/styyr@ne,de I'introdudion de liens doubles et de segments siloxane dans la chaine polyur@thaneuree,sont ici pr@sentes Afin d'i nvestiguer cet effet une experience factorielle 2s a et@congue et men@ Plusieurs propri@t@sde dispersions, de revetementset de feuils ont @redetermin@es, y compris, entre autres, la taille de particule, la stabilite, la MFFT (temperature minimale de la formation de film) de dispersions, la durete, la resistance 3 I'eau et aux solvants, I'adh@sionde revOtements et les propri@t@smecaniques et la temperature de transition vitreuse (T,~)de films. Les propri@tes des revetements et des films qui avaient ete sour-is 8 un sechage additi@nel ont @gale merit ~t~examinees. L'analyse des resultats de I'exp@riencefadorielle a facilite une meilleure comprdhension de I'effet mutuel de plusieurs elements sur les propridtds des matdriaux examinees.Aussi, ont ~te effectu@s, des ~tudes ESCA (La spedroscopie d'~ledron pour I'analyse chimique) et DCA (l'angle de contact dynamique) sur les propri~t~s de surface des revetements
Eine Studie des Effektes yon Strukturparametem auf die Eigenschatten yon Polysiloxanurethandispersionen und -Lacken
or correspondencecontat J Kozakiewicz Industrial Cherristry Research Institute, 8 R}/dygieraSt, 01-793 Warsaw, Poland Tel: +48 22 568 2 3 7 8 Fax: +48 22 568 2564
CopyrightOCCA2005
Email:janusz
[email protected]
Wir erforschten den Effekt yon Strukturparametern auf die Eigenschaften yon Polysiloxanurethan Dispersionen und -Lacken Der Polyurethanharnsbff-Kette ~urden Doppelbindungen und Siloxansegmente zugefOgt und tier EinfluB auf die Eigenschaften des Polyurthanes und auf PolyurethanAcTyl oder Polyurethan-SITrendispersionen und -Lacke studierL die Ergebnisse ~erden hier vorgestellt. Der Effekt vvurde mittels eines speziell ent~ickelten 2S-faktorischen Experimentes studiert Wit stellten mehrere Eigenschaften der Dispersionen, Lacke und Filme lest, wie unter anderem PartikelgrOsse,StabilitSt und MFFT (minimale Filmbildungstemperatur) yon Dispersionen, HSrte, Wasser und L~sungsmittelresistenz und AdhSsionskraft yon Lacken, und die mechanischen Eigenschaffen und die Glastransitionstemperatur (T) yon Filmen Die Eigenschaffenyon Lacken und FiImen nach zusatzhchem Harten ~urde auch erforscht Die Analyse des faktorlschen Expenmentes half, die gemeinsamen EinflGsse veryschiedener Faktoren auf die Materialeigenschaften zu verstehen Die Obeffl3cheneigenschafiten der Lacke wurden auch mit Hilfe yon ESCA (Elektronspek troskopie f~Jrchemische Analyse) und DCA (dynamischer Kontaktwinkel) erforscht. ,
Surface Coatings International Part B: Coatings Transadions Vol 89, B1, 1-98, March 2006
,
,
,
g
'
,
31
STudies on the effect of structural p~rameters on the properties of polysiloxaneurethmle dispersions and coatings J Kozaldewicz, A Koncka Foland, J Skarzynski, J W Sobczak and M Zielecka
Introduction In the last decade, aqueous dispersions of polyurethaneurea ionomers (commonly known as polyurethane dispersions) have become one of the most popular coating binders, in particular where wood coatings are concerned.l,2 Though much more expensive than acrylic dispersions, polyurethane disper sions have several advantages: depend ing on the composition of the starting materials, they can form glossy, transpar ent coatings which are very hard and abrasion resistant as well as soft feel coatings and textile coatings; they per form well when subjected to water and organic solvents; and, at the same time, they do not crack at low temperatures. A further great advantage of polyurethane dispersions is also their ability to form films at quite low temperatures (ie they show very low minimum film-forming temperatures (MFFT)). The excellent properties of the films formed by polyurethane dispersions result from the specific phase structure of polyurethane ureas (ie micro domains of the hard ure thane and urea segments are distributed in a soft phase consisting of polyetherdi ol or polyesterdiol segments). Hydrogen bonds between urethane/urea segments increase the mechanical strength of the coating while soft segments add some degree of flexibility. The properties of polyurethane dispersions and coatings made from them can be quite easily modified in the desired direction due to the great ability of the polyurethane chemical structure to be tailored to meet the requirements of the user. Further improvement of the prop erties of the film formed by polyurethane dispersions can be achieved s,4 either by using special curing agents (polyaziridine, melamine formaldehyde resins or blocked polyisocyanates) or by oxidative curing through double bonds present in the chain. Other more sophisticated curing methods such as cross-linking of carbonyl groups introduced to the polyurethane ionomer chain via azomethine bonds have been recently published/ Special cross linkable systems, like the combina tion of an aqueous dispersion of a poly ol and an aqueous dispersion of the cur ing agent containing blocked isocyanate groups e-s and hydrolysable alkoxysilane groups) ,1~ have also been revealed in the literature. Polyurethane dispersions are also attractive for chemists and coating formulators due to the possibilities of the development of hybrid systems. Such hybrid systems, in particular polyurethane-acrylic
32
dispersions having a core shell particle structure and prepared by polymerisa tion of acrylic monomers in polyurethane dispersions, ~ have become very popular in the coatings market because they are usually cheaper than pure polyurethane dispersions but still maintain the substantial features of the latter. The kinetics of the process of preparation of such hybrid dispersions have been carefully studied.12,1s Extensive studies on such hybrid polyurethane acrylic/styrene dispersions carried out in the authors' laboratory ~4,1~ have clearly shown, ~r~t~e~aha, that the presence of unsaturation in the main chain of polyurethaneurea ionomer helps achieve better properties of coatings made of such hybrids, presumably because grafting (and possibly also crosslinking) with acrylic/styrene monomers has been facilitated. On the other hand, the authors' earlier investigations proved ~6 that very interesting properties of coatings can also be expected when siloxane oligomer diols are used for the synthesis of polyurethane dispersions instead of standard polyether or polyes terdiols leading to polysiloxaneurethane dispersions. It seemed then reasonable to look for a possibility of combining these two approaches by synthesising polysiloxaneurethane dispersions con taining double bonds in the main chain of the p olyurethaneurea ionomer. The aim of the study presented in part in this paper was to investigate the combined effect of the content of siloxane segments and double bonds in the main chain of polyurethaneurea ionomer and of the ratio of polyurethane to poly(acrylic/styrene) part of the disper sion solids on the properties of polysilox aneurethane dispersions, films and coat ings. A more detailed presentation of the results of that study is intended to be published separately.
Experimental Dispersion synthesis Polysiloxaneurethane dispersions were synthesised using a standard prepolymer-ionomer method described in the literature/,4 In the first step, isophorone diisocyanate was reacted with diol components (siloxane oligomer diol Tegomer H Si 2111 [Goldschmidt, MW ca 950] polyetherdiol PTMG 2000 [BASE MW ca 2000], unsaturated polyesterdiol experimental product with unsaturation incorporated in the chain, hydroxyl nmnber 276, and dimethylolpropionic acid) in small amounts of N methylpyrrolidone, and
the concentrated NCO terminated pre polymer solution was then neutralised with triethylamine to fom~ prepolymer ionomer. In the next step, prepolymer ionomer was emulsified in water and reacted with polyamine to achieve a partly cross-linked polysiloxaneurethaneurea ionomer aqueous dispersion, When the polysiloxaneurethaneacrylic/styrene dispersion was to be prepared, a mixture of monomers butyl acrylate (25%)/methyl methacry late(34%)/styrene(41%), and an aqueous solution of emulsifiers and other addi tives was admixed into the polysiloxa neurethane dispersion, and the mixture was left overnight. Then, a standard emulsion polymerisation process was carried out using potassium persulphate as the initiator. The polymerisation was carried out to completion and the total monomer content measured after polymerisation did not exceed 20ppm. It was anticipated that the monomers were grafted onto the unsaturated part of the polyurethane chain and almost no resid ual unsaturation remained (this can be supported by the observation of the neg ative combined effect of poly(acrylic/ styrene) part content and unsaturated segments content in the film on film hardness (see explanation to Figure 4c below). F o r m a t i o n o f c o a t i n g s a n d films In order to prepare the coatings, disper sions were cast on to glass or metal (steel) plates (depending on the coating test planned) using a 120 micron appli cator and dried for 72 hours at room temperature (RT) (for adhesion and hardness tests) or at 40~ (for the water resistance test). If films were to be prepared, dispersions were cast onto glass Petri dishes and conditioned for 14 days at 25~ and 55% relative humidity (RH). For ultraviolet (UV)-curing tests, coatings and films were dried at 120~ for 30 minutes, and after that, a UVA beam (1000W lamp) was applied for 30 see onds. For oxidative curing, 3% of Addi tol VXW 4930 dryer (fl'om UCB) (,per dis persion solids) was added. Testing o f d i s p e r s i o n s , c o a t i n g s and films Standard EN tests or Polish Standards (PN) tests were used when appropriate (eg Persoz hardness was determined according to PN-79/C-81530). Particle size and zeta potential analysis were determined using a Malvern Zeta Sizer 4 apparatus. MFFT (minimum film-forming temperature) was measured on a MFFT Thermostair apparatus (Coesfeld). Mechanical stability of dispersions (not
Surface Coatings International Part B: Coatings Transactions Vol 89, B1, 1-98, March 2006
Studies on the effect of structural parameters on the properties of polysiloxaneurefllane dispersions and coatings J Kozaldewicz, A Koncka Foland, J Skarzynski, J W Sobezak a n d M Zielecka
fable 1: Selected results of properties of dispersions, coatings and films determined for products of single experiments that constituted the factorial experiment
Y2 mech stability, minutes Ys MFFT,~ }/4 averageparticle size, nm }/6 zetapotential, mV }/7 Persozhardness }/s2- Tg I, ~ }/ss- Tg II, ~
-7,-7, -7
-7,+7, -7
-7,-7, +7
-7,§ +7
§ -7
+7
-7
+7
90 05 494 538 020 -7240 8706
30 0.5 56.9 60.4 0.41 -73.52 68.11
60 0.5 46.2 56.2 0.27 -70.63 78.00
30 0.5 60.3 54.9 0.60 -69.75 69.62
15 75 1671 455 060 -8894
15 414 1651 380 036 -8839 -
15 75 1688 448 034 -8781 8507
15 389 1727 496 079 5895
discussed in this paper) was detemdned by rotating dispersion samples at 4000rpm and measuring the time (rain utes), in which separation occurred. DSC (dynamic scanning calorimetric) analysis to determine the T~ of films was made according to ASTM D 3418 82.
Testing of surface properties of coatings Surface properties of coatings were investigated using two independent methods: electron spectroscopy for chemical analysis (ESCA) using a ESCALAB-210, Fisons Instruments apparatus, and dynamic contact angle (DCA) using a Processor Tensiometer, Kruss apparatus.
Design of factorial experiment A standard factorial experiment of the 2 s type was designed. Three independent variables (X), selected after preliminary investigations and changing from -1 to
§
+ 1 (non real values), were: X1 content of the poly(acrylic/styrene) part in dis persion solids (real range 0 to 60% w/w); X2 content of unsaturated diol in the mixture of diols (real range 0 to 20% w/w); and X3 content of siloxane oligomer diol segments in the mixture of diols (real range 0 to 40% w/w). According to the standard factorial experiment plan, 12 single experiments were conducted (eight single experiments in the 'corners' of the plan, ie + 1 a n d - 1 combinations and four repeated single experiments in the centre of the plan, ie 0,0,0). Over 70 dependent variables (Y) were determined for the products obtained in each experiment, including properties of dispersions (inter alia MFFT, average par ticle size, mechanical stability), of coat ings (m~.er a?Cta Persoz hardness, water and solvent resistance, drying time, adhesion), and of films (krAe~ ~}2t mechanical properties, T# cross-linking density).
Fitted surface;variable:part size 2"*(3-0) design;MSresidual 761.4501 Y4:particlesize(nm)forX3= 0 (middlesiloxanelevel)
§
§
0,0,0 30 0.5 77.5 55.9 0.33 -77.00 75.94
Selected results of single experiments that constituted the factorial experiment (only the effects which are detailed in the paper are shown) are presented in Table 1.
Analysis of factorial experiment results Analysis of the factorial experiment results was made using tile Statistica computer program. Because of limited space allowed for this paper, only a few examples of the results obtained in the factorial experiment can be presented here in the form of graphs generated by the computer and representing the effect of two variables (Xl - poly(acrylic/ styrene part content and X2 - unsaturated diol segment content) on the selected properties of the dispersions, coatings or films (Yi) for two different levels of the third variable (X3 siloxane segment content). Function Yi f(Xl, X2, X3) is also shown on the graphs. So called 'Pareto charts' generated by the comput
Fittedsurface;variable:part size 2**(3-0) design;MSresidual = 761.4501 Y4: particlesize (rim)for X3 = 1 (highsiloxanelevel)
1.50
go 09
1.00
t 9
~! == 36.696 50.493
0
0.50
0.00 -0.50 1.00 , -1.00
,
,
-0.50
0.00
[ 0.50
1.00
,
! = 64.29 = 78.086 ~= 91.883 ! = 105.68 = 119.477 = 133.274 I(= 147.071 = 160.868 10 li = 174.665
-~
00!/!//://///
%
0.00
~
G *~
1.00
1 = 49.286 2 63.477 3 77.667 4 91.857 5 106.048 6 = 120.238 7 = 134.428 8 148.619 9 162.809 10 = 176.999
9
0.50 -1.00
1.50
-1.00 -0.50 0.00 0.50 1.00 1.50 Xl: contentof poly(acrylic/styrene)part
Xl: contentofpoly(acrylic/styrene)part
I
Y4 = 99.70g33 + 57.6125Xl + 3.g625X2 + -1.5375XlX2 + 0.2125XlX3 + 1.5625X2X3 - 0.0875XlX2X3 + X3 I
I
I
Y4 = 99.70933 + 57.6125]1 + 3.g625X2L+ 1.5375]1X2 + + 1.5625]2]3
O.Og75XlX2X3+0.2625
o.2125XlX3I [
Figure la and b: Combined effectof Xl [poly(acryliclstyrene)part content] and X2 [unsaturated segments content] on particlesize of polysiloxaneurethane acrylic/styrenedispersions [nm] (a) for middle levelof X3 (siloxane segments content); (b) for high levelof X3 (siloxane segments content) Surface Coatings International Part B: Coatings Transactions
Vol 89, B1, 1-98, March 2006
33
Studies on the effect of structural parameters on the properties of polysiloxaneurethmle dispersions and coatings J Kozaldewicz A Koncka Foland, J Skarzynski, J W Sobczak and M Zielecka
er are also included because they show the separate effect of each variable, combined mutual effect of any pair of two variables, and also the combined effect of all three variables on the investigated properties. Based on information provided by the 'Pareto charts', interesting observations regarding the complex relationships between structural parameters of the polymers under investigation can be made.
Effect of polymer structure related factors (X1, X2 and X3) on the properties of polysiloxaneurethane-acrylic/ styrene dispersions The results of the factorial experiment have clearly demonstrated that out of all three investigated factors, X 1 (content of poly(acrylic/styrene) part of the hybrid polymer which constituted the disper sion solids) had undoubtedly the greatest influence on the properties of the dis persions. This influence can be observed in particular on the graphs representing the average particle-size dependence on Xl and X2 for two levels of X3 (see Figures 1a and lb) and on the corresponding 'Pareto chart' (see Figure lc). While the average particle size for 'starting' polysiloxaneurethane dispersions (Xl 1) ranges from 50 to 85nm, after polymerisation and fomlation of poly siloxaneurethane acrylic/styrene disper sions, it grows up to 160 to 175nm. This growth can be clearly noticed in Figure 2 where the particle size distribution curves for one of the starting polysiloxa neurethane dispersions [Xl = -1, X2 = +1, X3 = +1] (a) and corresponding polysiloxaneurethane-acrylic/styrene dispersion [Xl = +1, X2 = +1, X3 = + 1] (b) are shown. The starting polysiloxaneurethane dispersion shows two peaks (bi-modal particle-size distribution) that sometimes occurs in this type of dispersion, but the final hybrid poly siloxaneurethane acrylic/styrene disper sion shows only one peak which can be explained by quite large differences in the average particle size between the starting dispersion and the hybrid disper sion (60.3nm and 172.7nm, respective ly). The phenomenon of particle size growth after polymerisation suggests that the 'core-shell' or 'gradient' particle morphology has been obtained. Obviously, also other specific hybrid particle structures as, for example, the 'embedded sphere' morphology which was previously described for similar dispersions is can be expected. Therefore, further morphological studies are need ed to fully confitrn the actual particle structure. It is interesting that the Zeta
34
Pareto chartof standardisedeffects,variable:part size 2"*(3 0) design;MSresidual= 761+4501 Y~:particlesize p =0.5 (1)xl (2)xZ
5.905281 0.3950066
-
2by3 -
0.1601564
1by2 -
-0.157504
(3)x3 -
0.0269063
lby3 - - - [
0+0217813 -0.008069
-1
0
1
2
3
4
5
liffectestimate (absolutevalue) Figure lc: Effect of all three variables [X1, X2 and X3] on particle size of polysiloxaneurethane-acrylidstyrene dispersions - 'Pareto chart'
a
b
Sizedistribution(s)
....
+_,____~|~____~ i
i
Sizedislribulion(s) 15-
i
i
t t
t
t
i
i
t t
~.-_,
10-
5. . . . .
t
t i
i
t
t i
i
t
t
i
t
i
i
t
"--~- . . . . . t
1
'-~---
t
i
5 10
50 100
500 1000
Diameter (nm)
5 10
50 100
500 1000
Diameter (nrn)
Figure 2: Particle-size distribution of: (a) starting polysiloxaneurethane dispersion [Xl = 1, X2 +1, X3 +1], average particle size: 60.3nm, and (b) corresponding polysiloxaneurethane-acrylidstyrene dispersion [Xl = +1, X2 = +1, X3 = +1], average particle size: 172.7nrn potential (r of the polysiloxaneurethane-acrylic/styrene dispersion (see Figure 3b) is less negative than that of its polysiloxaneurethane precursor (see Figure 3a) which indicates that dis persion stability diminishes after poly merisation. (This has been confirmed by a mechanical stability test not shown here for explanation of the test see q~eslX~g o~ d~spe~s~ons, coatX~gs avid f~lv~s section)
Effect of polymer structure related factors (Xl, X2 and X3) on the properties of coatings made from polysiloxaneurethaneacrylic/styrene dispersions The substantial properties of aqueous dispersions-based coatings which decide in practice their application potential are hardness, adhesion and water/solvent
Surface Coatings International Part B: Coatings Transactions
Vol 89, B1, 1-98, March 2006
Studies on the effect of structural parameters on the polysiloxaneurefllane dispersions and coatings
properties of
J Kozaldewicz,A Kondm Foland, J Skarzynski,J W Sobczak and M Zieleeka ing double bonds), Is leading conse quently to the diminishing coating hard
b
ness.
8 -
8-
6 -
6 I
4 -
4-
2-
Effect of polymer structure related factors (X1, X2 and X3) on the properties of films made from polysiloxaneurethane acrylic/styrene dispersions It was not possible to make a computer analysis of the mechanical properties of the films o b t a i n e d from dispersions investigated in this study because most of the dispersions containing the poly(acrylic/styrene) part did not form continuous films. Lack of continuous film formation by these dispersions could be attributed to stresses resulting fl'om too high brittleness (relatively high T~) of the acITlic part of the polysiloxa n e u r e t h a n e acrylic/styrene polymer. However, if films w e r e formed, the mechanical strength they showed was very high (15 to 25MPa without any cur ing agent and 17 to 35MPa with curing agent polyaziridine). Elongation at break varied from 5 to 25% for samples prepared without curing agent, to 20 to 60% for samples with curing agent. Distinctly better m e c h a n i c a l properties were observed for films obtained from dispersions synthesised from diol mixtures containing unsaturated diol. This, once again, proves the positive influence of the presence of mlsaturation in the polymer chain on the perfom~ance of p o l y u r e t h a n e dispersions in coating applications.
2-
.d 100
100 Zeta potential(mV)
I
I
I
100
0
100
Zeta potential(rag)
Figure 3: Zeta potential (~) of: (a) starting polysiloxaneurethane dispersion [Xl = -1, X2 = +1, X3 = +1], ~= -54.9mV, and (b) corresponding polysiloxaneurethane-acrylidstyrene dispersion [Xl = +1, X2 = +1, X3 = +1], ~ = 49.6mV resistance. However, though the effect of Xl, X2 and X3 on these three substantial parameters (and also on many other coating properties) was investigated in this study, only hardness will be discussed here because of limited space. As demonstrated in Figure 4, coating hardness is positively affected by all three X variables investigated in this study. The combined mutual effect of X2 (unsaturated segments content) and X3 (siloxane segments content) and X2 alone have the greatest influence on coating hardness. The c o m b i n e d effect of all three variables is unexpectedly high which suggests that the effect of these factors on coating hardness is quite complex. This has been confirmed by
the authors' earlier study 16 where it was found that coating hardness changed in quite a complex way when the content of the siloxane segments in the polymer was increased. It is also interesting that the combined effect of X1 (poly(acrylic/styrene) part content) and X2 (unsaturated segments content) on coating hardness is negative. It is believed that this may be due to grafting of the acrylic/styrene m o n o m e r s on the double bonds of the polymer, which results in the blocking of double bonds present in the polymer and thus the prevention of oxidative curing (a well known p h e n o m e n o n occurring for coatings m a d e from dispersions of polymers, including polyurethanes, contain-
Fittedsurface;variable:hardness 2**(3-0) design;MSresidual= 0.0117113 YT:hardness (Persoz)for X3 = 1 (highsfloxanelevel)
Fittedsurface;variable:hardness 2**(3-0) design;MSresidual= 0.0117113 YT:hardness (Persoz)for X3 = 0 (middlesiloxanelevel)
L..
