Method of controlled conversion of thermosetting resins and additive manufacturing thereof by selective laser sintering

Abstract

The invention is directed to a method of controlled conversion of thermosetting resins and additive manufacturing thereof by selective laser sintering. Partial curing of a thermosetting formulation can be used to increase the T.sub.g of the resin and minimize the additional cure needed to cross-link a printed object. After printing, the partially cured material is finally cured via a slow temperature ramp maintained just below the material's evolving T.sub.g.

Claims

1. A method for printing a thermosetting polymer, comprising providing a partially cured resin material, wherein the resin material comprises a thermosetting resin and a curing agent, and wherein the thermosetting resin comprises an epoxy, bismaleimide, cyanate ester, alkyne, alkene, acrylate, anhydride, carboxylic acid, isocyanate, or halide, producing a resin powder from the partially cured resin material, printing and sintering the resin powder on a print bed at a bed temperature near a glass transition temperature of the partially cured resin material to provide a printed part having a desired three-dimensional shape, and curing the printed part according to a post-print cure schedule to provide a cured solid object.

2. The method of claim 1, wherein the resin material comprises a stoichiometrically balanced formulation of the thermosetting resin and the curing agent.

3. The method of claim 1, wherein the epoxy comprises difunctional bisphenol A/epichlorohydrine epoxy.

4. The method of claim 1, wherein the curing agent comprises an amine.

5. The method of claim 4, wherein the amine comprises 4,4′-diaminodiphenylsulphone.

6. The method of claim 1, wherein the curing agent comprises a thiol, alkene, anhydride, azide, carboxylic acid, or hydroxyl.

7. The method of claim 1, wherein the step of providing a partially cured resin material comprises mixing the thermosetting resin with the curing agent to provide a thermosetting resin formulation, and partially curing the thermosetting resin formulation to vitrification but before gelation is reached to provide the partially cured resin material.

8. The method of claim 1, wherein the step of providing a partially cured resin material comprises providing a resin-rich formulation wherein the curing agent is fully converted and the resin is not quite crosslinked, or a highly functional resin, providing a curing-agent-rich formulation wherein the thermosetting resin is fully converted but not crosslinked, or a curative polymer, and compounding the resin-rich formulation or highly functional resin with the curing-agent-rich formulation or curative polymer to provide the partially cured resin material.

9. The method of claim 1, wherein the post-print cure schedule comprises heating the printed part at a temperature below the glass transition temperature until gelation is reached, followed by post-curing of the printed part at a temperature above the glass transition temperature.

10. The method of claim 1, when selective laser sintering is used for the step of printing and sintering the resin powder.

11. The method of claim 1, wherein a layer-by-layer additive manufacturing process is used to the step of printing and sintering the resin powder.

12. The method of claim 11, wherein crosslinking between layers is not achieved during the print, but rather during the post-print cure.

13. The method of claim 1, wherein the resin powder is unfilled.

14. The method of claim 1, wherein the resin powder further comprises a filler.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The detailed description will refer to the following drawings, wherein like elements are referred to by like numbers.

(2) FIG. 1 is a schematic illustration of a selective laser sintering process.

(3) FIG. 2 is a graph showing the evolution of storage modulus, reaction progress, and glass transition temperature (T.sub.g) with time (hours) at 120° C. for a stoichiometric Epon 828/4,4′-DDS formulation.

(4) FIG. 3 shows images of post-measurement DSC pans for 828/4,4′-DDS formulation cured at 120° C. for between cure times of 1.97 h and 5.93 h.

(5) FIG. 4 shows an oven-sintered bar suspended over a 15 mm span for an incremental sub-T.sub.g cure schedule and a high-temperature cure.

(6) FIG. 5 is a graph showing the evolution of the cure state and onset T.sub.g (solid circles) of a printed material during the prescribed cure schedule. Vertical dashed lines indicate when the set-point of the curing oven was decreased by 10° C. The initial rapid cure progression indicates the initial 4.25 h hold at 120° C. to produce the printable material.

(7) FIG. 6A is a photograph of a test part printed from the 828/4,4′-DDS formulation. FIG. 6B is a photograph of the cured printed part.

DETAILED DESCRIPTION OF THE INVENTION

(8) The present invention is directed to SLS-printable thermosets which maintain their shape during post-print cure. The invention uses simple filler-free thermosetting resin formulations which can be printed and then cured without deforming. A preferred method relies on partially curing the thermosetting formulation to a point just before gelation. This pre-reaction serves two purposes. Firstly, advancing the cure chemistry increases the T.sub.g of the non-crosslinked formulation to above 25° C. which is preferable for non-crystalline-printable powders. Not all thermosetting resin formulations will exhibit room temperature (25° C.) vitrification prior to gelation—these formulations will likely not be satisfactorily printed using this approach. Once printed, gelation of the material can be thermally driven below, but near, its T.sub.g. Curing in the mostly glassy state will however result in drastically reduced post-print cure kinetics. Thus, the second reason for partially curing the formulation prior to printing is to reduce the amount of residual chemistry required to achieve gelation and allow for higher temperature sub-T.sub.g curing in an effort to minimize post-print cure time. The stoichiometrically balanced thermosetting formulation can be achieved by mixing the thermosetting resin with a stoichiometric amount of curing agent or curative. A variety of thermosetting resins can be used, including epoxy, bismaleimide, cyanate ester, alkyne, alkene, acrylate, anhydride, carboxylic acid, isocyanate, or halide. A variety of curing agents, or curatives, can be used to cure the thermosetting resin, including amine, thiol, alkene, anhydride, azide, carboxylic acid, or hydroxyl.

(9) Alternatively, a partially cured resin material can be achieved by compounding a resin-rich formulation wherein the curing agent has been fully converted such that resin is nearly crosslinked (but not quite) with a curing-agent-rich formulation wherein the resin has been fully converted but not quite crosslinked, to provide a stoichiometrically balanced, homogenous partially cured resin material. Therefore, very little additional reaction is required to crosslink the resin material after it is printed. The stoichiometries required to produce the off-stoichiometry resin-rich and curing-agent-rich formulations can be determined using the Flory-Stockmayer equation and solving for the stoichiometry at which 100% reaction of the limiting reagent is required to achieve gelation. Alternatively, one or both of these off-stoichiometric formulations can be replaced with a highly functional resin, such as a multifunctional epoxy resin (e.g., EPON 1031), or a curative polymer, respectively.

Cure-Time Controlled Conversion

(10) A controlled conversion approach can theoretically be applied to any thermosetting resin which vitrifies near room temperature prior to gelation. Epon 828/4,4′-diaminodiphenylsulphone (4,4′-DDS) formulation was used as an example, due to its slow simple cure kinetics and high T.sub.g at full cure. Epon 828 is a difunctional bisphenol A/epichlorohydrine liquid epoxy resin sold by Miller-Stephenson that can be crosslinked or hardened with an appropriate curing agent. 4,4′-DDS is an aromatic amine curing agent sold as Hardener HT 976 by Ciba Specialty Chemicals. A stoichiometrically balanced formulation was obtained by mixing Epon 828 with 4,4′-DDS, assuming respective functional equivalent weights of 188.5 g/mol and 62 g/mol. Small scale (<10 g) blending was performed by combining Epon 828 and 4,4′-DDS at room temperature and subsequently heating the mixture to 170° C. under manual stirring until the mixture became clear because of the melting and dissolution of the 4, 4′-DDS. Large scale blending of these constituents was performed at 150° C. by heating both components separately prior to mixing and maintaining a temperature of 150° C. while mixing. The grade of 4,4′-DDS powder used comprised 95-100% of particles <150 microns in diameter. Larger particle diameters may require increased dissolution time when hand-mixing at these temperatures, which may lead to undesired curing during mixing.

(11) The cure behavior at 120° C. for this Epon 828/4,4′-DDS formulation is shown in FIG. 2. The T.sub.g of the formulation gradually increases as the reaction progresses, and the material reaches gelation at around 4.25 hours. At this point, the onset T.sub.g of the material is approximately 71° C., and the overall reaction extent is approximately 55% which agrees with the predicted gelation point as calculated using the Flory-Stockmayer equation. See P. J. Flory, J. Am. Chem. Soc. 63, 3083 (1941). However, some undetectable amount of reaction may have occurred during the blending of 4,4′-DDS into the Epon 828, which was done at 150° C., so the actual extent of conversion at gelation may be higher than 55%. FIG. 3 shows samples at increasing reaction extent which have been examined by differential scanning calorimetry (DSC) up to 150° C. Near gelation (4.28 h), the material retains some surface texture when heated to 150° C., and past this point (4.96 h and 5.93 h) sintering no longer occurs. However, during the 10° C./min thermal ramp to 150° C. used for the DSC scans it is possible that the reaction extent of these samples increased to the point of gelation. Thus, while this series of samples qualitatively demonstrates how reaction extent affects printability, it may not be representative of actual sintering behavior during SLS printing where the material experiences a rapid heating to higher temperatures.

(12) Using the correlation of reaction extent, T.sub.g, and gelation behavior shown in FIG. 2 as a guide, 600 g of partially cured Epon 828/4,4′-DDS resin material was produced by partially curing the mixture at 120° C. for ˜4 hours. At this point, the material displayed an onset T.sub.g of ˜70° C. which did not change over 6 months suggesting that this formulation has a long shelf-life at ambient conditions, likely due to the inherent slow kinetics of the Epon 828/4,4′-DDS reaction and storage well below its T.sub.g. Printability was initially assessed by placing the powdered material in a 200° C. oven for 10 seconds, which confirmed that it could flow at elevated temperatures. This short exposure did not detectably progress the reaction, as evidenced by its unchanged onset T.sub.g. Further, printed samples which were allowed to remain at room temperature for 6 additional months also did not show an increase in T.sub.g and completely liquified on exposure to 200° C. Thus, for this formulation, the printing process likely would not progress the reaction through gelation; the formation of a cross-linked network must occur during post-print thermal treatment.

Printing of the Partially Cured Resin Material

(13) The resin powder was printed using a Sintratec Kit SLS printer. Relevant technical specifications for the printer are shown in Table 1.

(14) TABLE-US-00001 TABLE 1 Specifications for Sintratec Kit SLS Printer Build Volume (max) 110 × 110 × 110 mm Layer height 50-150 um Chamber temperature 150° C. (max) Surface Ternperature 180° C. (max) Laser 2.3 W/455 nm, power/wavelength non-variable Laser scan speed 650 mm/s max

(15) The results of printed powders with various printer conditions are shown in Table 2. Initial printing of the unfilled powder formulation was unsuccessful due to poor absorption of the laser energy. To remedy this, approximately 0.7% carbon black was dry mixed with the resin powder to increase its absorption efficiency. This powder is considered to be “unfilled” for the purposes of demonstrating the effectiveness of the method to SLS printable materials. At such low loadings of a non-elongated filler, viscoelastic properties of the material are not affected, and, given a more powerful printing laser, the carbon black energy absorber would not be needed. Using a sub-T.sub.g bed temperature, while a scanning speed of 650 mm/s resulted in poor sintering behavior, lowering the scan speed imparted a thermal gradient to the printed layer which resulted in curling due to the top of the layer cooling faster than the bottom. See J. P. Kruth et al., CIRP Ann. 56, 730 (2007). This “curl” in the printed layer is then frozen in as that layer cools to below its T.sub.g. This layer curling is detrimental to the print, as when the next layer of powder is applied, the roller or blade will contact the curl and sweep the printed layer off of the print bed, ruining the print.

(16) TABLE-US-00002 TABLE 2 Effect of various printing parameters on print quality Bed/ Scan Hatch Laser Material chamber speed distance power examined temp (° C.) (mm/s) (μm) (W) Result Unfilled 65/55  50 250 2.3 No sintering Carbon 65/55 100 250 2.3 Burning of filled (0.7 material wt %) 65/55 200 250 2.3 Edge curling 65/55 300 250 2.3 Edge curling 65/55 450 250 2.3 Edge curling 65/55 650 200 2.3 Poor sintering 69/59 550 200 2.3 Edge curling

Curing and Post-Curing of the Printed Resin Powder

(17) Initial attempts to print the powder using a laser scan speed of 650 mm/sec resulted in porous parts. However, the structural stability of these printed bars was sufficient to demonstrate the effect of the post-print cure schedule on printed parts. An incremental sub-T.sub.g cure schedule allows the material to reach gelation whilst in the glassy state and subsequently be post-cured while maintaining its as-printed shape. After testing multiple post-print cure schedules, the following cure profile resulted in the least amount of deformation during cure: 24 h at 65° C., 24 h at 75° C., 24 h at 85° C., 6 h at 95° C. and 2 h at 120° C., as shown in FIG. 5. Using this cure schedule, an oven-sintered bar of material maintained its shape with minimal sagging while suspended over a 15 mm span, as shown in FIG. 4. Further cure for 2 h at 200° C. also resulted in no observable warpage. For comparison, immediately after sintering, another bar of material was held at 120° C. over the same length span for 2 minutes over which time it exhibited severe sagging. These examples illustrate that firstly, a sub-T.sub.g cure schedule facilitates crosslinking in the glassy state, and secondly, such a cure schedule is vital to induce crosslinking while maintaining the shape of the printed object, especially in unfilled thermoset materials.

(18) Printing of this black powder resulted in porous parts even when using the most optimal laser speed (650 mm/s). However, the structural stability of these printed bars was sufficient to demonstrate the effect of the previously mentioned cure schedule on printed objects. As shown in FIG. 4, heating the printed material to 120° C. results in partial re-melting and excessive sagging, and immediate exposure to 200° C. results in complete re-melting. However, the incremental sub-Tg cure schedule allows the material to reach gelation and subsequently be post-cured while maintaining its as-printed shape.

(19) The print bed temperature can be optimized to further improve print quality. Most reports on the printing of amorphous thermoplastics suggest that a sub-T.sub.g print bed temperature is required to prevent the powder bed from coalescing. See J. P. Kruth et al., CIRP Ann. 56, 730 (2007). However, it was found that using a bed temperature slightly above the powder's T.sub.g is acceptable for the epoxy resins and, importantly, reduces the edge curling effect. For this material, using a bed temperature of 80° C. (5° C. above the midpoint T.sub.g) enabled the use of a slower laser scanning speed (500 mm/s) which reduced porosity without causing the printed layer to curl. Additionally, at this temperature, the powder did not solidify prematurely. However, some powder agglomeration did evolve making extraction of printed parts more difficult. Nevertheless, it was possible to sieve this agglomerated powder through a 150-micron sieve for reuse.

(20) FIG. 6A shows a complex test geometry printed from the 828/4,4′-DDS formulation. Even with “optimized” parameters, this printed part still displays significant porosity, and the as-printed, or “green”, part was chalky with low resistance to abrasion. Still, a slow-stepped cure which follows the increasing T.sub.g of the part was sufficient to cure this material without any heat induced dimensional changes of part, as shown in FIG. 6B.

(21) The present invention has been described as method of controlled conversion of thermosetting resins and additive manufacturing thereof by selective laser sintering. It will be understood that the above description is merely illustrative of the applications of the principles of the present invention, the scope of which is to be determined by the claims viewed in light of the specification. Other variants and modifications of the invention will be apparent to those of skill in the art.