Nanosized powders (NP) show different properties with respect to the corresponding bulk materials. The enhanced reactivity connected to their very high specific surface area, arises interest also for energetic systems application: their use allows increasing solid propellants (SP) burning rate and reducing agglomerate formation at or near the combustion surface region. Moreover, addition of nanosized metals to solid fuel (SF) seems to be a promising pathway to solve the low regression rate problem of hybrid rocket propulsion. NP are also characterized by different surface interactions, resulting in difficulties in handling because of the tendency to form clusters that hinder the complete exploitation of nanomaterials potential. In the experimental part of this work various strategies, ranging from sonochemistry to pre-dispersion in solvent to surface treatment, are investigated to improve the dispersion degree of NP into a polymeric matrix. Solid fuels for hybrid rocket engines (HRE) and propellants for solid rocket motors (SRM) are considered. In both cases HTPB is one of the main components: it represents the SF that can be loaded with various percentages of nanosized aluminum (nAl) in HRE and the binder entrapping fuel and oxidizer in SRM. In the former case, strategies to improve the dispersion degree of nanosized fuel (nAl) are evaluated as well as manufacture techniques to well dispersed nanosized Fe2O3, used as combustion modifier in SP. The evaluation of dispersion degree is carried out, whenever possible, through the direct inspection of samples with an optical microscope, and in any case with indirect strategies, exploiting the connection between NP dispersion and some physical properties of the mixtures such as thermal conductivity or mechanical and rheological properties. Combustion tests are performed to evaluate the effectiveness of the improved dispersion on the combustion features of SF and SP. SF are tested in a 2D radial burner at 10 bar under a gaseous oxygen mass flow rate up to to 350 kg/(m2s). SP combustions at pressure going from 1 to 70 bar, are conducted in a strand burner, using N2 as pressurizing gas. Both for SF and SP optical methods are used to evaluate the regressing surface. The experimental campaign shows that the best technique to prepare SF loaded with ALEXTM is the pre-dispersion of dry powders in ethyl acetate, that grants an increment of regression rate up to 64% at 350 Kg/(m2s) with respect to the HTPB baseline. Nevertheless this procedure is not compatible with coated powders, for which dry addition remains the only possibility. In this case, the best option is given by the use of an ultrasonic bath for the dry powders before their addition to the other ingredients, followed by 30’ mixing of the whole compound with a propeller at 100 rpm using another ultrasonic device. Combustion tests on samples prepared with this procedure containing ALEXTM, show an enhancement in regression rate up to 40% at 350 kg/(m2s) with respect to unloaded HTPB. Optical and mechanical analyses show no improvement of dispersion quality given by coating, when stearic acid or fluorotelomer alcohol are considered, nevertheless in combustion tests these formulations exhibit a regression rate 50% higher with respect to the HTPB baseline at 350 kg/(m2s). Pre-dispersion of NP in ethyl acetate is considered also for SP. In formulations containing nanosized Fe2O3 and μAl this technique led to a burning rate augmented up to 215% at 40 bar with respect to the analogous SP prepared with the standard procedure. The effectiveness of the technique is connected to the low percentage of combustion catalyst in the formulation (0.5%). In fact its application to SP containing 18% of nAl results rather ineffective. For this kind of formulations another strategy to improve homogeneity in dispersion is evaluated: ALEXTM is coated with HTPB before the addition to the other ingredients (H-ALEX). Combustion tests show an enhancement in burning rate up to 26% with respect to uncoated ALEXTM containing SP and up to 117% with respect to the μAl filled composition. The quality of dispersion affects also rheological aspects of the uncured formulations with consequences in manufacture and castability, and thus in ballistic behavior of the SRM. The theoretical part of the work aims to find a criterion to define the quality of dispersion of NP inside SP choosing the stiffening coefficient (H) as controlling parameter. First only μAl containing formulations are considered to investigate the connection between propellants microstructure and rheological properties. The simulation of realistic heterogeneous material is demanded to a packing code, that gives the distribution of the centres of each particle in a cube centred in the origin. The obtained distribution is used to evaluate the number of fine particles (μAl) around every coarse one (AP) enhancing the oxidizer fraction. The first result is the linear growth of the number of particles with AP fraction, up to 20% in volume. In this result it is possible to see the explanation for the limit of validity of Einstein equation, that states a linear relation between viscosity and the volumetric solid fraction into a fluid. For higher percentages of solid content the deviation from linear behaviour becomes clear and the number of Al particles around each AP grain increases faster and faster with AP content. The research for the influence of microstructure on rheological properties of SP shows a cubic correlation between H and the number of Al particles around each AP grain. This correlation is used to evaluate the number of micrometric particles in nAl/μAl containing formulations. In the hypothesis of homogeneous dispersion of NP only μAl constitutes the fine fraction, while nAl is included in the liquid phase. In case of cohesion, clusters act as micrometric particles. In this way, the correlation obtained for H can be used to predict the theoretical viscosity of a well disperse formulation. The difference between theoretical and actual viscosity is used as a method to evaluate the quality of the dispersion, comparing ALEXTM and H-ALEX containing formulations: in case of 5% nAl, clusters increase theoretical number of μAl particle of 7% for the better disperse formulation and of 21% for ALEXTM containing ones. The number of cluster increases with nAl fraction; this is explained considering that particles are closer and attractive forces are more effective.
Le polveri nanometriche (NP) sono caratterizzate da proprietà differenti rispetto ai corrispondenti materiali di dimensioni maggiori. In particolare si può sfruttare la loro notevole reattività, legata all’elevata superficie specifica, anche in ambito propulsivo: il loro utilizzo permette di aumentare la velocità di combustione dei propellenti solidi e di ridurre la formazione degli agglomerati che lasciano la superficie di combustione. Inoltre, l’additivazione di combustibili solidi con metalli nanometrici può costituire una soluzione per risolvere il problema della bassa velocità di regressione che contraddistingue i sistemi propulsivi ibridi. Le NP sono caratterizzate anche da differenti interazioni superficiali che si traducono nella formazione di coaguli che impediscono il completo sfruttamento delle potenzialità dei nano materiali. Nella parte sperimentale di questo lavoro sono state investigate diverse strategie per migliorare il livello di dispersione dei materiali nanometrici in una matrice polimerica, andando dalla sonochimica, alla predispersione in solventi, ai trattamenti superficiali. Si considerano sia propellenti solidi (SP) sia combustibili ibridi per applicazioni spaziali. In entrambi i casi l’HTPB è uno dei componenti principali: rappresenta il combustibile che può essere caricato con nano-alluminio (nAl) nel caso di sistemi ibridi e il legante che intrappola ossidante e combustibile nel caso di SP. In quest’ultimo caso si sono considerate tecniche per aumentare l’omogeneità della dispersione sia di combustibili nanometrici (nAl), sia di altri ingredienti quale Fe2O3 usato come catalizzatore di combustione. La valutazione del grado di dispersione è fatta, quando possibile, attraverso la diretta ispezione dei campioni con un microscopio ottico e, in ogni caso, mediante strategie indirette, sfruttando la connessione fra la dispersione delle NP e alcune proprietà fisiche dei composti come la conducibilità termica o le proprietà reologiche e meccaniche. I test di combustione vengono effettuati per valutare l’efficacia della miglior dispersione sulle caratteristiche di combustione di combustibili ibridi e propellenti solidi. I combustibili solidi sono testati in un bruciatore radiale a 10 bar in flusso di ossigeno gassoso che raggiunge al massimo 350 kg/(m2s). La combustione dei SP avviene in un micro-bruciatore in pressioni che variano tra 1 e 70 bar, utilizzando N2 come gas pressurizzante. In entrambi i casi la valutazione della superficie che regredisce avviene con tecniche ottiche. I risultati sperimentali mostrano che la miglior tecnica per preparare i combustibili solidi caricati con ALEX si basa sulla pre-dispersione delle polveri in acetato di etile, che garantisce un aumento della velocità di regressione fino al 64% a 350 Kg/(m2s) rispetto all’ HTPB non additivato. Ciò nonostante questa procedura non è compatibile con le polveri ricoperte, per le quali l’unica possibilità rimane aggiungere le polveri asciutte. In questo caso la miglior opzione è data dalla sonicatura delle polveri prima dell’introduzione nella miscela che viene successivamente mischiata con un mixer ad elica per 30’ a 100 rpm e contemporaneamente sollecitata attraverso un sonicatore. I test di combustione effettuati su campioni ottenuti con questa procedura, contenenti ALEXTM mostrano un incremento della velocità di regressione fino al 40% a 350 kg/(m2s) rispetto all’HTPB non additivato. Analisi ottiche e meccaniche mostrano che la presenza di coating sulle particelle non aumenta il livello di dispersione quando acido stearico o fluortelomero sono utilizzati, anche se la velocità di regressione di queste formulazioni si rivela maggiore del 50% rispetto a quella della baseline di HTPB a 350 kg/(m2s). La pre-dispersione di NP in acetate di etile è stata considerata anche per SP. Nelle formulazioni contenenti Fe2O3 nanometrico e μAl questa tecnica ha portato ad un aumento della velocità di combustione del 215% a 40 bar rispetto alla stessa formulazione ottenuta con la procedura standard. L’efficacia della tecnica è da collegarsi alla bassa percentuale di catalizzatore di combustione presente nella formulazione (0.5%). La stessa procedura infatti, applicata a SP contenenti il 18% di nAl si è rivelata inefficace. Per queste formulazioni si è seguita un’altra strategia per incrementare l’omogeneità della dispersion: le polveri sono state ricoperte di HTPB prima di aggiungerle agli altri ingredienti (H-ALEX). I test di combustione mostrano un aumento della velocità di combustione del 26% rispetto alla stessa formulazione contente ALEXTM non ricoperto e fino al 117% rispetto a SP contenti μAl. La qualità della dispersione si riflette anche sulle proprietà reologiche delle formulazioni non reticolate con conseguenze in fase di manifattura e collaggio che vanno a pregiudicare il comportamento balistico del motore a propellente solido. La parte teorica del lavoro mira a trovare un criterio per definire la qualità della dispersione delle NP nei SP scegliendo come parametro di controllo il rapporto (H) fra la viscosità della sospensione e quella del fluido. Inizialmente si considerano solo formulazioni contenenti μAl per analizzare il collegamento fra la microstruttura del propellente e le sue proprietà reologiche. La simulazione realistica di un materiale eterogeneo è demandata ad un codice di impacchettamento che restituisce la posizione dei centri di ogni particella in un cubo centrato nell’origine. La distribuzione che si ottiene viene utilizzata per valutare il numero di particelle fini (μAl) che ci sono attorno a ciascuna particella grossa (AP) all’aumentare della frazione di ossidante. Il primo risultato che si ottiene è la crescita lineare del numero di particelle fini aumentando la frazione di AP fino al 20% in volume. In questo risultato si può vedere una spiegazione per il limite di validità dell’equazione di Einstein che stabilisce una relazione lineare tra la viscosità e la frazione volumetrica di particolato solido in un fluido. Per percentuali maggiori lo scostamento dall’andamento lineare diventa evidente e il numero di particelle di Al attorno a ciascun cristallo di AP cresce via via più rapidamente all’aumentare del contenuto di AP. Valutando l’influenza della microstruttura sulle proprietà reologiche del SP si nota una correlazione cubica tra H e il numero di particelle di Al attorno a ciascun grani di AP. Questa correlazione viene utilizzata per valutare il numero di particelle micrometriche in formulazioni contenenti miscele di nAl e μAl, ipotizzando che le NP siano omogeneamente disperse e quindi incluse nella fase liquida. In questo modo solo il μAl va a costituire la frazione fine. In caso di coesione i coaguli di NP si comportano come particelle micrometriche. La correlazione tra H e il numero di particelle viene usata per predire la viscosità di formulazioni omogeneamente disperse; la differenza tra viscosità prevista e quella sperimentalmente misurata è utilizzata per valutare la qualità della dispersione di formulazioni contenti ALEXTM e H-ALEX. Per SP con il 5% di nAl i cluster aumentano il numero teorico di particelle del 7% per le formulazioni meglio disperse e del 21% per quelle contenenti ALEXTM. Il numero di coaguli aumenta con la frazione di nAl, cosa che è spiegabile considerando che più le particelle sono vicine tra loro, più efficaci saranno le forza attrattive.
Nano-metal fuels for hybrid and solid propulsion
REINA, ALICE
Abstract
Nanosized powders (NP) show different properties with respect to the corresponding bulk materials. The enhanced reactivity connected to their very high specific surface area, arises interest also for energetic systems application: their use allows increasing solid propellants (SP) burning rate and reducing agglomerate formation at or near the combustion surface region. Moreover, addition of nanosized metals to solid fuel (SF) seems to be a promising pathway to solve the low regression rate problem of hybrid rocket propulsion. NP are also characterized by different surface interactions, resulting in difficulties in handling because of the tendency to form clusters that hinder the complete exploitation of nanomaterials potential. In the experimental part of this work various strategies, ranging from sonochemistry to pre-dispersion in solvent to surface treatment, are investigated to improve the dispersion degree of NP into a polymeric matrix. Solid fuels for hybrid rocket engines (HRE) and propellants for solid rocket motors (SRM) are considered. In both cases HTPB is one of the main components: it represents the SF that can be loaded with various percentages of nanosized aluminum (nAl) in HRE and the binder entrapping fuel and oxidizer in SRM. In the former case, strategies to improve the dispersion degree of nanosized fuel (nAl) are evaluated as well as manufacture techniques to well dispersed nanosized Fe2O3, used as combustion modifier in SP. The evaluation of dispersion degree is carried out, whenever possible, through the direct inspection of samples with an optical microscope, and in any case with indirect strategies, exploiting the connection between NP dispersion and some physical properties of the mixtures such as thermal conductivity or mechanical and rheological properties. Combustion tests are performed to evaluate the effectiveness of the improved dispersion on the combustion features of SF and SP. SF are tested in a 2D radial burner at 10 bar under a gaseous oxygen mass flow rate up to to 350 kg/(m2s). SP combustions at pressure going from 1 to 70 bar, are conducted in a strand burner, using N2 as pressurizing gas. Both for SF and SP optical methods are used to evaluate the regressing surface. The experimental campaign shows that the best technique to prepare SF loaded with ALEXTM is the pre-dispersion of dry powders in ethyl acetate, that grants an increment of regression rate up to 64% at 350 Kg/(m2s) with respect to the HTPB baseline. Nevertheless this procedure is not compatible with coated powders, for which dry addition remains the only possibility. In this case, the best option is given by the use of an ultrasonic bath for the dry powders before their addition to the other ingredients, followed by 30’ mixing of the whole compound with a propeller at 100 rpm using another ultrasonic device. Combustion tests on samples prepared with this procedure containing ALEXTM, show an enhancement in regression rate up to 40% at 350 kg/(m2s) with respect to unloaded HTPB. Optical and mechanical analyses show no improvement of dispersion quality given by coating, when stearic acid or fluorotelomer alcohol are considered, nevertheless in combustion tests these formulations exhibit a regression rate 50% higher with respect to the HTPB baseline at 350 kg/(m2s). Pre-dispersion of NP in ethyl acetate is considered also for SP. In formulations containing nanosized Fe2O3 and μAl this technique led to a burning rate augmented up to 215% at 40 bar with respect to the analogous SP prepared with the standard procedure. The effectiveness of the technique is connected to the low percentage of combustion catalyst in the formulation (0.5%). In fact its application to SP containing 18% of nAl results rather ineffective. For this kind of formulations another strategy to improve homogeneity in dispersion is evaluated: ALEXTM is coated with HTPB before the addition to the other ingredients (H-ALEX). Combustion tests show an enhancement in burning rate up to 26% with respect to uncoated ALEXTM containing SP and up to 117% with respect to the μAl filled composition. The quality of dispersion affects also rheological aspects of the uncured formulations with consequences in manufacture and castability, and thus in ballistic behavior of the SRM. The theoretical part of the work aims to find a criterion to define the quality of dispersion of NP inside SP choosing the stiffening coefficient (H) as controlling parameter. First only μAl containing formulations are considered to investigate the connection between propellants microstructure and rheological properties. The simulation of realistic heterogeneous material is demanded to a packing code, that gives the distribution of the centres of each particle in a cube centred in the origin. The obtained distribution is used to evaluate the number of fine particles (μAl) around every coarse one (AP) enhancing the oxidizer fraction. The first result is the linear growth of the number of particles with AP fraction, up to 20% in volume. In this result it is possible to see the explanation for the limit of validity of Einstein equation, that states a linear relation between viscosity and the volumetric solid fraction into a fluid. For higher percentages of solid content the deviation from linear behaviour becomes clear and the number of Al particles around each AP grain increases faster and faster with AP content. The research for the influence of microstructure on rheological properties of SP shows a cubic correlation between H and the number of Al particles around each AP grain. This correlation is used to evaluate the number of micrometric particles in nAl/μAl containing formulations. In the hypothesis of homogeneous dispersion of NP only μAl constitutes the fine fraction, while nAl is included in the liquid phase. In case of cohesion, clusters act as micrometric particles. In this way, the correlation obtained for H can be used to predict the theoretical viscosity of a well disperse formulation. The difference between theoretical and actual viscosity is used as a method to evaluate the quality of the dispersion, comparing ALEXTM and H-ALEX containing formulations: in case of 5% nAl, clusters increase theoretical number of μAl particle of 7% for the better disperse formulation and of 21% for ALEXTM containing ones. The number of cluster increases with nAl fraction; this is explained considering that particles are closer and attractive forces are more effective.File | Dimensione | Formato | |
---|---|---|---|
2013_03_PhD_Reina.pdf
non accessibile
Descrizione: testo della tesi
Dimensione
14.02 MB
Formato
Adobe PDF
|
14.02 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/74821