Thermosetting polyurethane foams by physical blowing agents: chasing the synthesis reaction with the pressure

Abstract

The aim of the present study was the introduction of a novel method to produce microcellular thermosetting polyurethane foaming by the gas-foaming technique, using high-pressure physical blowing agents. In particular, to tackle the inherent difficulties of imposing a rapid pressure quench O(10-2 s) to a material whose synthesis timing is much larger O(102 s), we utilized a two-stage foaming. In a first stage, a rapid pressure quench O(10-2 s) was imposed to nucleate a large amount of bubbles, from the saturation pressure to a low, non-zero pressure; in the second stage, the growth of the nucleated bubble is then controlled by slowly O(102 s) decreasing the pressure to ambient (zero) pressure. In this way, by separating the nucleation from the growth rate and by chasing the synthesis reaction with the pressure to avoid excessive stresses to the curing polymer, we achieved microcellular (size diameter below 10 um), medium-to-low density (below 200 g/L) thermosetting polyurethane foams.

Full text

Thermosetting polyurethane foams by physical blowing
agents: chasing the synthesis reaction with the pressure
Maria Rosaria Di Caprio1, Cosimo Brondi1, Giuseppe Scherillo1, Ernesto Di
Maio1,a), Thomas Mosciatti2, Sara Cavalca2, Vanni Parenti2, Maurizio Corti3 and
Salvatore Iannace4
1Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale, University of Naples Federico
II, P.le Tecchio 80, 80125 Naples, Italy
2Dow Italia s.r.l, Polyurethanes R&D, Via Carpi 29, 42015 Correggio, Italy
3Cannon Afros S.p.A., Via Ferraris Galileo, 33, 21042 Caronno Pertusella, VA, Italy
4Institute for Macromolecular Studies, National Research Council, Via Edoardo Bassini, 15, 20133 Milan, Italy
a)Corresponding author: edimaio@unina.it
Abstract. The aim of the present study was the introduction of a novel method to produce microcellular thermosetting
polyurethane foaming by the gas-foaming technique, using high-pressure physical blowing agents. In particular, to tackle
the inherent difficulties of imposing a rapid pressure quench O(10-2 s) to a material whose synthesis timing is much larger
O(102 s), we utilized a two-stage foaming. In a first stage, a rapid pressure quench O(10-2 s) was imposed to nucleate a
large amount of bubbles, from the saturation pressure to a low, non-zero pressure; in the second stage, the growth of the
nucleated bubble is then controlled by slowly O(102 s) decreasing the pressure to ambient (zero) pressure. In this way, by
separating the nucleation from the growth rate and by chasing the synthesis reaction with the pressure to avoid excessive
stresses to the curing polymer, we achieved microcellular (size diameter below 10 um), medium-to-low density (below 200
g/L) thermosetting polyurethane foams.
INTRODUCTION
Thermosetting polyurethane foams (PUs) were introduced at the beginning of the last century as a synthesis
product of a polyol and an isocyanate (synthesis reaction). Concurrent to the synthesis reaction, a blowing reaction of
isocyanate with water provides the blowing agent, CO2, responsible for foaming the curing matter. In this process, by
recalling the classification of Physical Blowing Agent (PBA) or Chemical Blowing Agent (CBA) utilized in the
foaming of thermoplastics [1], water is a CBA, while CO2 gradually forms and inflates bubbles at relatively low
pressures, not solubilizing in the polyol and/or the isocyanate phases prior to foaming, and cannot be considered a
PBA, here [2]. PUs have been also produced by using PBAs, like chlorofluorocarbons (CFCs) (in the late 1950s) and
hydro-chlorofluorocarbons (HCFCs) (at the beginning of 1990s), then replaced by hydrocarbons (HCs) (e.g. pentane)
because of their negative environmental impact [3]. PBAs are solubilized at relatively low temperatures and pressures
(typically, ambient temperature and pressure) in the polyol and/or isocyanate phases and then allowed to evolve in the
gaseous state by the temperature rise following the exothermic synthesis reaction. With the temperature increase, PBA
solubility decreases, and the solution undergoes phase separation (bubbles nucleation and growth). HCs use was
limited by their inherent flammability, despite being inexpensive and with zero ozone depletion potential (ODP) [3].
In this context, CO2 is considered more sustainable and safer, with zero ODP and the lowest global warming potential
(GWP = 1) among known blowing agents [3]. As a drawback, CO2 solubility is much lower than HCs and much larger
pressures have be utilized in order to reach concentrations appropriate for low-density foams, pushing technology and
know-how development.

High-pressure CO2 foaming proved very effective with thermoplastics. The pressure drop rate (PDR) is the key to
achieve microcellular or nanocellular foams, as introduced in the '80s by the MIT group [4] and, also at ultra-high
PDR O(103 MPa/s), the number of bubbles nucleated per unit volume exponentially increase with the PDR [5]. In
fact, with this method, utmost performances in terms of cell number densities have been reached with numerous
thermoplastic polymers: microcellular and, more recently, nanocellular foams have been produced, characterized by
improved thermal insulating and mechanical properties as compared to standard cell-sized foams [6]. So far,
thermosetting foams stay orders of magnitude behind, in terms of cell size.
To the best of our knowledge, no papers addressed the use of CO2 as a PBA in any thermosetting polymers, in
which CO2 solubilization is conducted before reactants mixing. Because of the environmental concerns and of the
encouraging performance in producing microcellular and nanocellular foams with thermoplastic polymers, high-
pressure CO2 foaming of PUs has attracted a growing industrial interest [7-13]. To ground the foam processing on a
scientifically sound base, a preliminary study included: i) the measurement and modeling of the physical properties
of the polyol/CO2 [14] and the isocyanate/CO2 solutions [15]; ii) the measurement and modeling of the retarding effect
of the CO2 concentration on the PUs synthesis kinetics by FT-NIR spectroscopy [16]; iii) the development of a new
experimental apparatus for processing and characterization of the foaming [17]. As a completion of this explorative
study, we herein report the development of a novel method that uses CO2/polyol and CO2/isocyanate solutions (formed
at high pressure) to produce microcellular (size diameter below 10 um), medium-to-low density (below 200 g/L)
thermosetting polyurethane foams. The method, essentially, copes with the two very different time scales of the gas
foaming O(10-2 s) and the gelling/curing reaction O(102 s), separating the nucleation and growth stages in two different
processing stages.
MATERIALS AND METHODS
Materials
A polyether polyol and polymeric methylene diphenyl diisocyanate (PMDI) were supplied by DOW Chemical
Italy S.r.l. (Correggio, RE, Italy) within the LIFE13-ENV/IT/001238 project [7] and used “as received”. High purity
grade CO2 (99.95% pure) and N2 (99.99% pure) were supplied by SOL (Naples, Italy).
Methods
To perform the foaming attempt, we utilized a novel pressure vessel designed to processing and studying the
foamability of PUs by CO2 as a PBA. The equipment was extensively described in [17]. The main novel features of
the equipment were: i) the use of a rubbery impeller for the reactant sealing (and successive mixing), ii) a high-
pressure-tight sapphire window mounted beneath an IR-transparent sample holder for remote NIR monitoring. Figure
1 reports 3D renderings of the proposed pressure vessel and images of the equipment in two configurations. With
respect to the original configuration (see Fig. 1b), to apply a partial pressure quench, we added a gas tank to the
evacuation system (see Fig. 1c).
(a)
(b)
(c)
FIGURE 1. Equipment utilized to perform the foaming tests. (a) 3D scheme showing a cut of the pressure vessel with the sample
holder and the mixing shaft. (b) Picture of the equipment in the original configuration reported in [19]: (1) actuator for ball-valve

(2); (3) pressure vessel. (c) Upgraded configuration with a gas tank (4), in addition, to evacuate the pressure vessel to a non-zero
pressure.
In a typical foaming attempt, the two components of the PU formulation are loaded in the cylindrical sample
holder, kept separate by the rubbery impeller, and subjected, at 35°C, to a CO2 pressure ranging from 2.0 to 10.0 MPa.
After a defined sorption time, needed to achieve the desired average CO2 mass fraction in the total formulation, the
impeller is put in rotation for the curing reaction to start. Then, at a certain degree of curing is achieved, the actuated
ball valve is opened for pressure release, at ambient pressure (configuration in Fig. 1b) or to a non-zero pressure
(configuration in Fig. 1c). In the latter case, by using the valve on top of the gas tank the pressure is eventually brought
to ambient after an additional curing period. The foamed samples are finally extracted from the sample holder for the
characterization.
RESULTS
Figure 2 reports the effect of one-stage and two-stage foaming on the PU cell morphology. The reactants were
kept under CO2 sorption for 3 hours at 35°C and 4 MPa and then (after mixing) cured for 12 minutes. In the case of
one-stage foaming, after the curing stage, the pressure was reduced to ambient pressure by slow pressure release (SPR)
at 0.1 MPa/s. A bi-modal cell size distribution was generated with bigger cells around 70 µm and a density of 400
g/L. In the other case, two-stage foaming was successfully applied to achieve the foam reported in Fig. 2b. After the
curing stage, the pressure was reduced to an intermediate pressure of 0.35 MPa by a fast pressure release (FPR) at 40
MPa/s. The PU sample was kept under the same pressure for 3 minutes and the second pressure reduction was
conducted by SPR. The two-stage foaming allowed to reach a uniform cell distribution with an average cell size of 10
µm and a density of 150 g/L.
(a)
(b)
FIGURE 2. SEM micrographs of PU samples kept under CO2 sorption for 3 hours at 35°C and 4 MPa and then cured for 12
minutes under pressure. (a) One-stage foaming attempt by SPR. (b) Two-stage foaming performed by FPR, the subsequent
holding stage and then the controlled SPR. Scale bars are 100 µm.
The main issue with one-stage foaming was, to our point of view, the very different time scales between reaction
time, O(102s), and pressure quench, O(10-2s), and, more generally, the idea to overturn the long-term development of
PUs foaming. The impossibility to achieve the desired goal by one-stage foaming induced us to derive a new method.
The main idea under the present two-stage method is to separate bubble nucleation from bubble growth and have
the two phenomena following the two timescales, O(10-2s) and O(102s). The two-step foaming stages and their
pressure history are shown in Figs. 3a and b respectively. In particular, in the first holding stage the pressure () is
kept constant (0-1) allowing the cure of the polyurethane mixture, in the first foaming stage (1-2) the bubble nucleation
is promoted by fast pressure (FPR) quenching (as fast as possible by the specific equipment) the reacting mixture, the
pressure () is kept constant during the second holding stage and, at last, the pressure is slowly reduced (SPR) to
ambient pressure (3-4). In this stage, the gas contained within the bubbles will expand.

(a)
(b)
FIGURE 3. Two-step foaming stages and their pressure history: (0-1) first holding stage, (1-2) fast pressure release, (2-3) second
holding stage, (3-4) slow pressure release.
It is possible to design the pressure drop history for the SPR in order to optimize foaming. As a proof of concept,
and as it is also simpler to apply to an equipment, we decided to work at a constant pressure drop rate, , and derived
a model to attain a desired value of final curing degree () when   (ambient pressure), starting from certain
initial values of  and :
  


 (1)
Where H is the Henry constant for CO2 in the averaged polymeric matter. Equation 1 describes the effect of the
reaction rate parameters (, ), the Henry constant, , the initial conditions (, ) and the desired  on .
CONCLUSIONS
The first attempts to foam the reacting PU formulation were based on the idea that, within the PU curing reaction
time O(102s), it was possible to guess the curing degree and, hence, a polymer viscosity suitable to tolerate the pressure
quench, which has to be fast enough to generate a high number of bubbles O(10-2s). However, the morphology results
(average cell size of 70 µm and a density of 400 g/L) proved the impossibility to achieve such a goal.
The novel two-stage foaming method allowed to obtain a PU foam with uniform cell size distribution around 10
µm and a density of 150 g/L. In the first stage, a rapid pressure quench O(10-2 s) was imposed to nucleate a large
amount of bubbles, from the saturation pressure to a low, non-zero pressure; in the second stage, the growth of the
nucleated bubble is controlled by slowly O(102 s) decreasing the pressure to ambient (zero) pressure. In this way, by
separating the nucleation from the growth stage, we may describe this method as "chasing the synthesis reaction with
the pressure" (pondus mixturam progredientem persequitur), intending the use of external gas pressure to control
expansion of the gas contained in the bubbles, while the curing reaction is inexorably running
ACKNOWLEDGMENTS
The European union’s financial support (LIFE13-ENV/IT/001238 project [7]) is gratefully acknowledged.
REFERENCES
1. J. Štěpek and H. Daoust, in Additives for Plastics, Polymers-Properties and Applications (New York, NY,
Springer, 1983)
2. D. Randall and S. Lee, in The polyurethanes book (New Jersey, John Wiley & Sons, Inc., 2003), p. 113.
3. D. Randall and S. Lee, in The Polyurethanes book (New Jersey, John Wiley & Sons, Inc., 2003), pp. 127.
4. C. B. Park, . D. F. Baldwin and N. P. Suh, Polym. Eng. Sci., 35, 432 (1995).

5. D. Tammaro, A. Astarita, E. Di Maio and S. Iannace, Ind. Eng. Chem. Res., 55, 5696, (2016).
6. S. Costeux, J. Appl. Polym. Sci., 131, (2014).
7. PU Disruptive technology to dramatically improve Energy Efficiency of Household Appliances, available at
http://ec.europa.eu/environment/life/project/Projects/index.cfm?fuseaction=search.dspPage&n_proj_id=5077.
8. B. Barron and J. Dunlap , Air frothed polyurethane foams, U.S. Patent US3821130A (1972).
9. C. Fiorentini and A. C. M. Griffiths , Froth process for continuous manufacture of polyurethane foam slab-stocks,
U.S. Patent US5665287A (1993).
10. E. Di Maio and E. Kiran, J. Supercrit. Fluids, 134, 157, (2018).
11. R. Eiben, W. Ebeling, H.M. Sulzbach and R. Raffel , Method and device for the production of foam using carbon
dioxide dissolved under pressure, U.S. Patent US5883143A (1998).
12. S. Lindner, W. Friederichs, R. Strey, T. Sottmann, E. Khazova, L. Kramer, V. Dahl and A. Chalbi , Process for
producing a polyurethane foam by means of supercritical or near-critical blowing agent, U.S. Patent
US9139683B2 (2009).
13. M. Fricke, M. SCHÜTTE, T. M. STAUDT, C. Holtze, S. Koch and F. Bartels, Micro-emulsions, WIPO (PCT)
Patent WO2013026813A1 (2011).
14. M. R. Di Caprio, G. Dal Poggetto, M. G. Pastore Carbone, E. Di Maio, S. Cavalca , V. Parenti and S. Iannace,
Fluid Phase Equilib., 425, 342, (2016).
15. M. R. Di Caprio, B. Immirzi, E. Di Maio, S. Cavalca, V. Parenti, S. Iannace and G. Mensitieri, Fluid Phase
Equilib., 456, 116, (2018).
16. M. R. Di Caprio, C. Brondi, E. Di Maio, T. Mosciatti, S. Cavalca, V. Parenti, S. Iannace, G. Mensitieri and P.
Musto, Polyurethane synthesis under high-pressure CO2, a FT-NIR study, Eur. Polym. J. (Submitted).
17. M. R. Di Caprio, D. Tammaro, E. Di Maio, S. Cavalca, V. Parenti, A. Fangareggi and S. Iannace, Polym. Test.,
62, 137, (2017).