The Synthesis and Ring-Opening Metathesis Polymerization of Energetic Norbornene Materials Authors: Valerie A. Sterling, Jack V. Davis, Jeremy T. Tisdale, David E. Chavez Published: 2026 ABSTRACT Energetic norbornenes are promising candidates toward the development of new energetic polymers due to the synthetic versatility of norbornene ring-opening metathesis polymerizations used in commercial applications. We report the synthesis of two energetic norbornene materials that can be made in two steps with modest yields, containing either trinitroethanol or fluorodinitroethanol substituents. The norbornene monomers were then polymerized, and the polymers were characterized by Fourier transform infrared spectroscopy (FT-IR), differential scanning calorimetry (DSC), contact angle measurements, and proton and carbon nuclear magnetic resonance spectroscopies (1H and 13C­{1H} NMR). Additionally, small-scale safety data consisting of electrostatic discharge (ESD), friction (FS), and impact (IS) sensitivities were measured for the norbornene monomers and their resulting polymers. These analyses revealed that energetic norbornene materials are relatively insensitive and have densities comparable to that of TNT (1.47–1.81 g·cm–3). Introduction Polymers are often combined with energetic compounds to form plastic-bonded explosives (PBXs). These polymers bind the explosive particles and enhance the manufacturability of the resulting PBXs, and as a result, many explosive applications utilize PBX formulations. These formulations are highly sought after to improve safety for applications in pyrotechnics and other energetic compositions. The polymers used in these formulations are typically nonenergetic and are chosen based on the desired mechanical or thermal properties of the final product. The polymer-to-energetic material ratio can also be adjusted to influence the detonation performance, mechanical properties, and safety handling of the formulation. Typically, a formulation will contain 5–10% polymer by weight. A drawback of the use of nonenergetic polymers in these formulations is the reduction of the overall explosive performance since the energy contribution comes only from the explosive. This effect is compounded by the higher densities of high explosives relative to hydrocarbon polymers and will increase the contributions of the binder (5–20%) even further on a volumetric scale. One way to improve the performance of explosive formulations is to use an energetic binder, which incorporates energetic functional groups into the polymer. − This allows formulations to achieve higher performance due to both increases in energy release and increases in density. A well-known example of this is the incorporation of nitrocellulose as an energetic binder combined with other explosive compounds. , By synthesizing energetic polymers, the safety and thermal properties can also be fine-tuned through the chemical structure of the polymer. Two energetic functional groups that have been incorporated into energetic polymers include trinitroethanol and fluorodinitroethanol. These compounds were first synthesized in 1960 (trinitroethanol) and 1968 (fluorodinitroethanol) and are ideal for adding to monomer compounds due to their ease of synthesis, free alcohol group open to synthetic reactivity, and abundance of energetic C-nitro groups. Once the energetic monomers have been made, their respective polymers are synthesized using a variety of polymerization methods including catalyzed radical UV irradiation through acrylates, − click-chemistry polymerizations through azide and alkyne/alkene reactions, − and ring-opening polymerizations of oxetanes and epoxides. − Ring-opening metathesis polymerization (ROMP) of energetic monomers is another possible way to construct energetic polymers. ROMP is a type of olefin metathesis that uses transition metal complexes to form new alkene carbon–carbon bonds. Due to the versatility of this reaction, it is a widely used route in industrial processes for the production of plasticizers, detergents, pharmaceuticals, and synthetic perfumes. Despite the wide application of ROMP reactions, there have been no reports of polymerizations of energetic monomers, and only one report in the literature has used ROMP to prepare energetic polymers by polymerizing a nonenergetic norbornene monomer followed by postpolymerization processing to create the energetic polymer. ROMP-based materials would further benefit the field of energetic research due to their characteristic high thermal and chemical stability that are desirable for energetic formulations, in addition to the commercial availability of starting materials that can be readily functionalized with energetic groups prior to polymerization. Due to the wide commercialization of ROMP materials, we believe that norbornene compounds and ROMP-based polymers are an underrepresented area in the field of energetic research and thus should be reported to facilitate more progress. Herein, we report the synthesis and characterization of the first energetic norbornene monomers and their polymerization by using ROMP polymerizations to produce novel energetic polymers. These materials were characterized for their thermal and safety properties. Results and Discussion Synthesis Previous efforts by Jiang et al. to use ROMP to prepare energetic polymers included the addition of energetic azide groups to a polymeric norbornene scaffold resulting from a four-step synthesis starting from cyclopentadiene. While this route achieves a bis-ester-containing energetic norbornene polymer, it requires postpolymerization modification to install the energetic moiety. There have been reports that suggest postpolymerization modification can remain incomplete due to inaccessibility to sites within the polymer macrostructure and slower or different reaction kinetics from the polymer , that can produce polymers with a mixture of functional groups and properties not intended. Additionally, the energetic polymer reported by Jiang et al. did not include theoretical or experimentally determined density. We chose to investigate synthetic routes to energetic norbornene monomers that could then be polymerized using ROMP. We began our efforts with commercially available trans-3,6-endo­methyl­ene-1,2,3,6-tetra­hydro­phth­aloyl chloride. Both trinitroethanol (TNE) ,, and fluorodinitroethanol (FDNE) − nucleophiles were linked to the norbornene molecule under basic conditions to produce compounds 1a and 1b in 84% and 82% yields, respectively (Scheme ). After short column chromatography, characterization of 1a and 1b was confirmed using Fourier transform infrared spectroscopy (FT-IR, see Experimental Methods), differential scanning calorimetry (DSC, see Experimental Methods), and proton and carbon nuclear magnetic resonance spectroscopies (1H and 13C­{1H} NMR, see Supporting Information S1–S4). The sensitivity properties including material response to impact, friction, and electrostatic discharge stimuli of compounds 1a and 1b were also determined (Table ). Synthesis of Norbornene Materials (1a and 1b) and Their Polymerizations for the Production of 2a and 2b 1Calculated and Measured Energetic Properties of 1a, 1b, 2a, and 2b Compared to Glycidyl Azide Polymer (GAP) 1a 1b 2a 2b GAP ρ [g·cm–3]1.81 1.74 1.511.471.3ΔH f [kcal·mol–1] –235.8–121.4–248.6 –135.8 34.0OBCO2% –59.8–50.4–59.8 –50.4 –163.7Largest chain M n [g·mol–1] --5173 ± 30895,551 ± 61865500Smallest chain M n [g·mol–1]--256 ± 33495 ± 14- T d [°C]94.9111.2155.070.7235 T g [°C]--33.9–2.52–35IS [cm] (J)>320 (>78.4 J)133.0 (32.6 J)178.3 (43.7 J)>320 (>78.4 J)>9.04 JFS [N] >360>360>360>360>444.8ESD [J] 0.120.120.060.06-aCalculated values.bLANL type 12, 50% drop height, and 2.5 kg weight. Neyer d-optimal method.c50% load Bruceton up/down method.dABL spark, 3.4% threshold initiation level (TIL).eCalculated for single repeating unit.fAverage molecular weight determined by GPC.gReference for GAP data including calculations for density and heat of formation. ,, Calculated and Measured Energetic Properties of 1a, 1b, 2a, and 2b Compared to Glycidyl Azide Polymer (GAP) Calculated values. LANL type 12, 50% drop height, and 2.5 kg weight. Neyer d-optimal method. 50% load Bruceton up/down method. ABL spark, 3.4% threshold initiation level (TIL). Calculated for single repeating unit. Average molecular weight determined by GPC. Reference for GAP data including calculations for density and heat of formation. ,, Ring-opening metathesis of 1a and 1b was conducted using Grubbs second-generation catalyst (G2) , in a sealed container with a solution of dichloromethane at 60 °C (Scheme ). After vacuum filtration and drying, polymers 2a and 2b were collected as dark brown solids. Filtration of the organic solution allowed for the removal of most unreacted monomer (1a or 1b) and catalyst. These polymers were then characterized using FT-IR (see Experimental Methods and Supporting Information S9 and S10), DSC (see Supporting Information S11–S14), and proton and carbon nuclear magnetic resonance spectroscopies (1H and 13C­{1H} NMR, see Supporting Information S5–S8). The sensitivity properties including material response to impact, friction, and electrostatic discharge stimuli of polymers 2a and 2b were also determined (Table ). ROMP procedures using Grubbs second-generation catalyst report catalyst loading ranges between 0.01 and 10 mol %. − While we originally attempted polymerizations with a catalyst loading of 0.01 mol %, we still observed unreacted monomer in the proton NMR of the collected sample. We attribute the necessary higher catalyst loading of 10 mol % to the low reactivity of both monomers. The low polymerization reactivity of the monomers is further supported by the use of a sealed and heated vessel during the polymerization reaction for the reaction to proceed. Considering that the ROMP reaction of norbornenes is an exothermic process, there is concern about the possibility of inducing thermal runaway, especially when reacting energetic monomers. Thermal runaway in this case could have led to further heat transfer into the energetic monomers, which could have resulted in decomposition events. While these materials contain several energetic functional groups (C-nitro), no strong exothermic events were observed during our polymerization reactions. Typical ROMP procedures include a quench of the reaction by the use of ethyl vinyl ether in order to terminate the polymer and convert the catalyst to an inactive species. Since this procedure did not involve quenching the polymerization, the polymer was not terminated, and it left these materials susceptible to polymer instability, degradation of the product, or loss of control over molecular weight. Considering the only purification was the organic solvent filter without a quenching step, this instability and lack of molecular weight control is observed by the wide polydispersity (Pd) range for polymer 2a (Pd = 1.34–0.98) and polymer 2b (Pd = 1.02–1.18) (see Supporting Information S15 and S17). This difference in Pd between polymers 2a and 2b is likely a result of different reaction rate kinetics for each monomer in addition to the mild purification, which could have either left more trapped and unreacted material in the sample or influenced the reversibility of the ROMP reaction, specifically for polymer 2a. Confirmation of the polymer formation was conducted using gel-permeation size exclusion chromatography (GPC) to determine the average molecular weight of each material (see Supporting Information S15–S18). Material Characterization The physicochemical parameters of materials 1a,1b, 2a, and 2b are listed in Table . Gas phase heats of formation for the monomers were calculated in the Gaussian suite of programs at the B3LYP/6-31+g­(d/p) level of theory, as described in the Supporting Information (pages 16–33). − These were converted to solid state heats of formation through the calculation of heats of sublimation reported by Mathieu with a correction for the fluorodinitro functional group. Differences in ΔH f° per fluorodinitro or trinitro group between 1a and 2a (−57 kcal·mol–1) and 1b and 2b (−56.4 kcal·mol–1) compare favorably with the experimentally measured difference between 2-fluoro-2,2-dinitroethanol and 2,2,2-trinitroethanol (−52.4 kcal·mol–1). For the two polymers, solid state heats of formation were found through the calculation of the heat of polymerization assuming hydride termination for 1, 2, and 3 repeating units. The averages used (−12.8 and −14.4 kcal·mol–1 for 1a and 2a, respectively) compare favorability with the experimentally measured heat of polymerization for norbornene (−14.9 kcal·mol–1). Monomer densities were calculated and predict that monomer 1a has the highest theoretical density at 1.81 g·cm–3. The monomer densities are higher than the resulting polymers due to the 3D cage structure of the norbornene monomer backbone. The cage is opened upon polymerization, and the density increase is lost. When the monomers are polymerized, the resulting polymers exhibit glass transition temperatures (T g). These T g values for 2a and 2b show that both polymers transition from a solid glassy state to liquids at 33.9 °C for 2a and −2.52 °C for 2b. Mechanical sensitivities including impact (IS), friction (FS), and electrostatic discharge (ESD) were measured for monomers 1a and 1b as well as polymers 2a and 2b. Both monomers had the same sensitivities to friction (>360 N) and ESD (0.12 J); however in regard to impact sensitivity, 1b (133 cm) was more sensitive than 1a (>320 cm). This difference in impact sensitivity can be attributed to the two additional energetic nitro groups and improved oxygen balance within 1b compared to the stabilizing effect of the fluorine atoms within 1a. Lastly, the mechanical sensitivity measurements for the polymers 2a and 2b also had the same friction (>360 N) and ESD (0.06 J) sensitivities. However, 2a (178.3 cm) was more sensitive to impact than 2b (>320 cm). Glycidyl azide polymer (GAP) is a common benchmark for energetic polymer comparisons due to its wide use and commercialization. Additionally, since GAP contains repeating C–O ether linkages, we hypothesized that energetic ROMP-based materials would be more thermally stable due to the repeating CC alkene backbone. , Comparisons of the norbornene energetic materials to GAP are listed in Table . GAP exhibits a higher thermal stability (235 °C) but is more sensitive to impact (>9.04 J) compared to the norbornene monomers and polymers. For safety considerations, higher thermal stability is desirable for polymers used in energetic formulations that could see a wide range of temperature gradients. While GAP is superior in this regard, another important safety consideration is the response of a material to external forces, such as impact, spark, and friction sensitivity. From the data collected, the norbornene monomers and polymers exhibit less sensitivity to impact forces, making these materials safer to handle compared to GAP. Additionally, GAP has a lower density (1.3 g·cm–3) compared to both the starting norbornene materials and their respective polymers. Density is an important property for energetic materials since it is directly correlated to material performance. This relationship is well-established through Kamlet–Jacobs equations. There is a noticeable difference in comparing the glass transition temperatures between GAP (T g = −35 °C) and norbornene polymers 2a (T g = 33.9 °C) and 2b (T g = −2.52 °C). Since many formulations are tuned for their desired applications, there could be a case where a formulation may require a higher T g temperature (2a and 2b) rather than a lower T g temperature (GAP). For example, storage considerations for formulated materials matter if there are wide temperature ranges throughout the year that, depending on which binder is used, could result in binder migration , should the temperature exceed or cycle above −35 °C (GAP). By having slightly higher T g values, the norbornene polymers are less susceptible to thermal migration and have better stability in this example case. Film Properties To determine the bulk material characteristics of these energetic polymers (2a and 2b), we made drop-cast films in order to test the films for surface wettability. To prepare the drop-cast films, we prepared separate solutions containing high explosive (HE) polymer (0.10–0.11 g) in 2 mL of tetrahydrofuran. This solution was mixed for 5 min, until the polymer was observed to completely dissolve, and then dropwise added to a cleaned microscope slide and dried at 40 °C. This process was repeated until the film appeared to be uniform across the slide. The solution containing dissolved polymer was drop-cast as prepared and was not further filtered. After drying overnight at 40 °C, the polymer films were then imaged using a Keyence VR-5000 optical profilometer (Figure ). Polymer films (2a (left, a and b) and 2b (right, c and d)) prepared from THF solutions and drop-cast onto microscope slides to observe film roughness and uniformity. These images illustrate the roughness and uniformity of the films as prepared. From these images, it can be seen that in both samples there are layers less than 0.2 mm thick (Figure a,c) that appear in a uniform layer. Both samples also have island features greater than 0.8 mm in thickness (Figure a,c) that are also observed in the normal imaging modes (Figure b,d). The roughness exhibited in these films is likely due to the lack of filtration treatment of the polymer solution prior to it being drop-cast onto the microscope slides. The additional filtration step is useful for presenting uniform film materials but does not capture the natural state of an “as prepared” film. While this is a drawback to the usefulness of these materials as an “as prepared” film, there is still potential for these polymers to serve as energetic binders within a PBX formulation should they go through a postpolymerization filtration. To further expand upon the film properties, additional slides were prepared using a single cast droplet on microscope slides, and the wettability of the polymers 2a and 2b was characterized using contact angle measurements with water (Figure ). Surface wettability is a way to measure and compare the adhesion properties of the liquid–solid to liquid–liquid interactions and is useful for characterizing hydrophilic versus hydrophobic films. The hydrophobic or hydrophilic nature of films relates to the surface free energy of the material, which is important for determining material compatibility. Typically, hydrophobic materials, such as waxy plastics, have a contact angle >90°, while hydrophilic materials, such as glass or metals, have a contact angle <90°. The contact angle of water on each polymer film after deposition of the drop was slightly above 90° (contact angle on 2a = 93.23°, contact angle on 2b = 94.76°). The contact angle for each drop was measured at 2 s intervals for 1 min after deposition to also look at the change in the contact angle over a short period of time. The contact angle on polymer 2a reduced to 90.53° after 1 min, and the contact angle on polymer 2b reduced to 87.47° after 1 min. The complete data set for each drop shape analysis measurement is provided in the Supporting Information (S19 and S20). Polymer films (2a (left) and 2b (right)) prepared from THF solutions and drop-cast onto microscope slides to observe film water contact angle measurements. The angle that the water droplet creates on the surface characterizes if the film surface is hydrophilic or hydrophobic. From this contact angle analysis, polymers 2a and 2b can be described as having neutral wettability or neither hydrophilic nor hydrophobic characteristics. This result is interesting because of the variety of functional groups present in each polymer that could influence hydrophilic/hydrophobic interactions (fluorine, nitro groups, carboxylic esters, alkyenes, and alkyl linkages). Conclusion Energetic norbornenes are promising candidates toward the development of new energetic polymers due to the synthetic versatility of norbornene ring-opening metathesis polymerizations used in commercial applications. We have synthesized two energetic norbornene materials that can be made in two steps and modest yields, containing either trinitroethanol or fluorodinitroethanol substituents. The resulting polymers from the norbornene compounds were synthesized using Grubbs second-generation catalyst and exhibited molecular weights above 5000 g·mol–1. Additionally, safety data consisting of electrostatic discharge (ESD), friction (FS), and impact (IS) sensitivities were measured for the norbornene monomers and their resulting polymers, proving that norbornene materials are less sensitive to impact, making them safer to handle than other widely known energetic polymers, such as GAP. However, while ROMP-based polymers typically offer high thermal stability, the low thermal stability of these energetic norbornene polymers needs to be improved upon by further investigation of other energetic functional groups. Lastly, analysis of the norbornene films was conducted to observe film surface roughness and measure surface wettability. These results concluded that both of the films have irregularity and a neutral wettability (contact angles near 90°), which is useful for coating applications. Experimental Methods NMR data were collected on a Bruker Avance III 400 MHz NMR spectrometer at room temperature. Spectra were referenced to residual DMSO-d 6 (proton 2.50 ppm, carbon 39.52 ppm), chloroform, CDCl3 (proton 7.26 ppm, carbon 77.16 ppm) with chemical shifts reported in δ values (ppm), or acetone, deuterioacetone-d 6 (proton 2.05 ppm and carbon 29.84 ppm, 206.26 ppm) and J values in hertz (Hz). The following abbreviations are used to describe peaks: d (doublet), brs (broad singlet), m (multiplet). FT-IR spectra were measured with a Thermo-Nicolet iS5 FT-IR spectrophotometer using OMNIC software, and data are reported in reciprocal centimeters (cm–1). IR spectra measurements were made using an ATR cell. The GPC instrument used for this analysis is a Malvern OMNISEC (Westborough, MA, USA) equipped with quad detection: refractive index (RI), UV–vis photodiode array (PDA), multiangle light scattering, and viscometer. A sample injection volume of 100 μL was used with a mobile phase flow rate of 1 mL/min. For all analyses, two serially coupled PL Gel Mixed C columns (Agilent Technologies, Palo Alto, CA, USA) were implemented using THF as the eluent. DSC measurements were conducted on either (A) an STD 650 instrument (TA Instruments) coupled to a Discovery II quadruple mass spectrometer (TA Instruments) or (B) a DSC Q-2000 instrument (TA Instruments). Contact angle measurements were conducted using a Kruss DSA100S drop shape analyzer. When the drop of water was deposited onto the film, an initial measurement was recorded. After the initial measurement, additional measurements of the contact angle were made at 2 s intervals for an additional minute. The water used for the contact angle measurements was LC-MS grade water procured from Honeywell Burdick & Jackson (Cat. LC365-1). Caution! The compounds presented are highly energetic with sensitivity to various stimuli. While we encountered no issues while working with these materials, proper protective measures (Kevlar gloves, face shield, and grounded equipment) should still be used at all times. 1a: trans-3,6-Endo­methyl­ene-1,2,3,6-tetra­hydro­phth­aloyl chloride (1.095 g, 5.00 mmol) was dissolved in methylene chloride (50 mL) and placed in an addition funnel. In a 250 mL round bottomed flask, methylene chloride (40 mL), 2-fluoro-2,2-dinitroethanol (1.54 g, 10.0 mmol), and 2,4,6-collidine (1.21 g, 10.0 mmol) were added. A solution of the acid chloride was then added dropwise from the addition funnel to the reaction flask over 1 h. The reaction was allowed to stir for an additional hour. The reaction solution was then concentrated, and tert-butyl methyl ether was added. The 2,4,6-collidinium chloride was precipitated and filtered. The filtrate was concentrated, dissolved in a minimum amount of methylene chloride, and filtered through a short plug of silica gel to provide the desired product in 84% yield. Mp: 94.9 °C (dec). 1H NMR (deuterioacetone-d 6, ppm): 1.48 (dd), 1.58 (d), 2.71 (m), 3.15 (d), 3.26 (bs), 3.46 (t), 5.49 (m), 6.06 (dd), 6.34 (dd). 13C NMR (deuterioacetone-d 6, ppm): 45.60, 46.43, 46.92, 47.25, 47.33, 60.33, 60.49, 60.52, 60.69, 118.31, 121.20, 134.51, 137.71, 170.15, 172.50. IR (cm–1): 3010, 2987, 2954, 2895, 1760, 1745, 1618, 1608, 1445, 1389, 1313, 1265, 1248, 1206, 1176, 1162, 1111, 1083, 1054, 852, 799, 771, 725. 1b: trans-3,6-Endo­methyl­ene-1,2,3,6-tetra­hydro­phth­aloyl chloride (1.095 g, 5.00 mmol) was dissolved in methylene chloride (50 mL) and placed in an addition funnel. In a 250 mL round bottomed flask, methylene chloride (40 mL), 2,2,2-trinitroethanol (1.81 g, 10.0 mmol), and 2,4,6-collidine (1.21 g, 10.0 mmol) were added. A solution of the acid chloride was then added dropwise from the addition funnel to the reaction flask over 1 h. The reaction was allowed to stir for an additional hour. The reaction solution was then concentrated, and tert-butyl methyl ether was added. The 2,4,6-collidinium chloride was precipitated and filtered. The filtrate was concentrated, dissolved in a minimum amount of methylene chloride, and filtered through a short plug of silica gel to provide the desired product in 82% yield. Mp: 111.2 °C (dec). 1H NMR (deuterioacetone-d 6, ppm): 1.50 (dd), 1.59 (d), 2.73 (m), 2.82 (s), 3.16 (d), 3.27 (bs), 3.49 (t), 5.85 (m), 6.08 (dd), 6.35 (dd). 13C NMR (deuterioacetone-d 6, ppm): 46.59, 47.59, 48.00, 48.11, 48.43, 125.43, 135.59, 138.71, 170.64, 171.95. IR (cm–1): 3010, 2971, 2895, 1744, 1594, 1387, 1334, 1306, 1267, 1246, 1153, 1115, 1058, 913, 855, 801, 781. Polymerization Procedure (Adapted from Ref ) Energetic norbornene material (0.200 g) was added to a round bottomed flask, followed by dichloromethane (10 mL) and Grubbs II catalyst (0.02 g). The mixture was prepared under ambient conditions without the use of an inert atmosphere. The mixture was stirred, followed by sealing and heating to 60 °C for 12 h. After 12 h, the reaction was cooled back to room temperature, diluted with dichloromethane (20 mL), and vacuum filtered to obtain the respective polymer as the precipitate, with residual monomer and remaining catalyst in the organic solvent; however, trace amounts of these materials may remain trapped in the polymer and can lead to undesired reactivity. 2a polymer: trans-3,6-Endo­methyl­ene-1,2,3,6-tetra­hydro­phth­aloyl 2-fluoro-2,2-dinitroethanol. Isolated 0.184 g (89% yield). 13C NMR (400 MHz, CDCl3) δ (ppm): 174.12, 129.30, 124.40, 121.16, 60.92, 60.38, 34.30, 26.55, 20.2. IR (cm–1): 929, 1081, 1166, 1600, 1733, 2934. 2b polymer: trans-3,6-Endo­methyl­ene-1,2,3,6-tetra­hydro­phth­aloyl 2,2,2-trinitroethanol. Isolated 0.175 g (86% yield). 13C NMR (400 MHz, DMSO-d 6) δ (ppm): 206.48, 174.23, 128.89, 59.48, 30.67, 26.21, 25.75. IR (cm–1): 854, 1150, 1592, 1740, 2959. Supplementary Material