CHEMICAL ENGINES AND METHODS FOR THEIR USE, ESPECIALLY IN THE

INJECTION OF HIGHLY VISCOUS FLUIDS

RELATED APPLICATIONS This application claims priority to U.S. Provisional Patent Applications Nos. 61/713,236 and 61/713,250; both of which were filed October 12, 2012 and U.S. Provisional Patent

Application No. 61/817,312, filed April 29, 2013.

INTRODUCTION

This invention concerns technologies in which a gas is generated via a chemical reaction. The forces created the released gas can be harnessed to power useful processes. The chemical reactions are not combustion and avoid many of the problems associated with combustion.

Instead, the chemical reactions usually involve the generation of C0 ₂ from bicarbonate (HCO ₃ ). In general, this technology is termed chemical engine technology, or simply ChemEngine® as the technology developed by Battelle Memorial Institute is known. The invention is especially useful for the delivery of protein therapeutics.

Protein therapeutics is an emerging class of drug therapy that can treat a broad range of diseases. Because of their large size and limited stability, proteins must be delivered by parenteral delivery methods such as injection or infusion. For patients suffering from chronic diseases that require regular treatment, the trend is towards self-administration by subcutaneous injection, for example in the administration of insulin by diabetics. Typical subcutaneous injection involves delivery of 1 mL of formulation, but sometimes up to 3 mL, in less than 20 sec. Subcutaneous injection may be carried out with a number of devices, including syringes, auto-injectors, and pen injectors.

Transitioning therapeutic protein formulations from intravenous delivery to injection devices like syringes requires addressing challenges associated with delivering high

concentrations of high molecular weight molecules in a manner that is easy, reliable, and causes minimal pain to the patient. In this regard, while intravenous bags typically have a volume of 1 liter, the standard volume for a syringe ranges from 0.3 milliliters up to 25 milliliters. Thus, depending on the drug, to deliver the same amount of therapeutic proteins, the concentration may have to increase by a factor of 40 or more. Also, injection therapy is moving towards smaller needle diameters and faster delivery times for purposes of patient comfort and compliance.

Delivery of protein therapeutics is also challenging because of the high viscosity associated with such therapeutic formulations, and the high forces needed to push such formulations through a parenteral device. Formulations with absolute viscosities above 20 centipoise (cP), and especially above 40-60 centipoise (cP) are very difficult to deliver by conventional spring driven auto-injectors for multiple reasons. Structurally, the footprint of a spring for the amount of pressure delivered is relatively large and fixed to specific shapes, which reduces flexibility of design for delivery devices. Next, auto-injectors are usually made of plastic parts. However, a large amount of energy must be stored in the spring to reliably deliver high- viscosity fluids. This may cause damage to the plastic parts due to creep, which is the tendency of the plastic part to permanently deform under stress. An auto-injector typically operates by using the spring to push a needle-containing internal component towards an outer edge of the housing of the syringe. There is risk of breaking the syringe when the internal component impacts the housing, due to the high applied force needed to inject a high- viscosity fluid. Also, the sound associated with the impact can cause patient anxiety, reducing future compliance. The generated pressure versus time profile of such a spring driven auto-injector cannot be readily modified, which prevents users from fine tuning pressure to meet their delivery needs.

The force that is required to deliver a given formulation depends on several factors including the needle diameter (d), the needle length (L), formulation viscosity (μ), and volumetric flow rate (Q). In the simplest approximation - one that does not consider frictional forces between the plunger and the barrel— the force is related to the pressure drop (ΔΡ) multiplied by the cross-sectional area of the plunger (A). The pressure drop (ΔΡ) of a fluid in laminar flow through a needle may be described by the Hagen-Poiseuille equation:

_{F = P} . _{A =} . _A

π-d ⁴

In a syringe, the force is provided by the user. Reasonable finger force is considered to be less than 15-20 N for healthy patient populations and somewhat less for patients with limited dexterity, such as the elderly or those suffering from Rheumatoid arthritis or multiple sclerosis. In typical auto-injectors, the force is provided by a spring. The force provided by a spring decreases linearly with displacement, and the spring must be chosen so that sufficient force is available to sustain the injection. Viscosities above 20 cP, become difficult to deliver in a reasonable time by the typical spring driven auto-injector:

• Breakage of plastic parts that hold compressed spring; large energy stored leads to creep · Syringe breakage (high initial force)

• Incomplete doss delivered due to stalling (insufficient final force)

• Inflexibility of device designs, including large footprint of spring

Other sources of energy have been considered for auto-injectors. One source is the use of an effervescent reaction that creates pressure on-demand. A study funded by the Office of Naval Reserve (SoRI-EAS-85-746), entitled "Development of an On-Demand, Generic, Drug-Delivery System, 1985 described the use of bicarbonates mixed with acids to generate C02 that could drive slow delivery of a drug fluid. These devices were targeted at slow, long-term delivery over 24 h. Bottger and Bobst disclosed the use of syringe that uses a chemical-reaction to deliver fluid (US2011/0092906). Good et al. in "An effervescent reaction micropump for portable microfluidic systems," Lab Chip, 2006, 659-666 described formulations intended for micropumps using various concentrations of tartaric acid and sodium bicarbonate and different sizes of sodium bicarbonate particles. However, their invention provides delivery in a manner that the injection force increases exponentially over time. The prior art chemical engines do not provide adequate delivery, especially for conditions where the impact of the expanding volume in the piston is non-negligible, such as when the reagent volume is minimized to minimize the engine footprint and overshoot; without accounting for the expanding volume, chemical engines can stall during delivery in the same way that springs stall.

The present invention provides solutions to the foregoing problems by employing improvements to chemical engine technology. In especially preferred aspects, processes and devices are described in which a chemical engine can be used to comfortably and quickly self- administer a high-viscosity fluid with a relatively small injector. These processes and devices could be used to deliver high-concentration protein, or other high viscosity pharmaceutical formulations. SUMMARY OF THE INVENTION The invention provides chemical engines and methods of using the chemical engines to drive a fluid. The invention also includes methods of making chemical engines.

In a first aspect, the invention provides a chemical engine, comprising: a closed container comprising an acid, bicarbonate, water, and a plunger; a mechanism adapted to combine the acid, the water, and the bicarbonate; and further characterized by a power density of at least 50,000 W/m , as measured at a constant nominal backpressure of 40 N, or a power density ratio of at least 1.4 as compared to a control comprising a 3: 1 molar ratio of sodium bicarbonate and citric acid and having a concentration of 403 mg citric acid in 1 g H ₂ 0.

The characterization of the chemical engine by a power density is necessary because, in view of the variety of factors described herein, it is not possible to define the full breadth of the invention by other means. The stated levels of power density were not obtained in prior devices and the claimed levels of power density were not previously identified as either desirable or attainable. This feature brings together numerous technical advantages such as providing ease of holding a powered syringe, that delivers a viscous solution with greater comfort and less risk of breakage than conventional, spring-powered autoinjectors or previously described gas-powered injectors. The claimed characteristic has the additional advantages of ease of measurement and high precision of the measured values.

In some preferred embodiments, at least 50 wt of the bicarbonate is a solid. It has been surprisingly discovered that potassium bicarbonate provides a faster reaction and generates more C02 than sodium bicarbonate under otherwise identical conditions. Thus, in preferred embodiments, a chemical engine comprises at least 50 wt potassium bicarbonate. Preferably, the acid is citric acid, and in some preferred embodiments the citric acid is dissolved in water; the configuration with solid potassium carbonate and citric acid in solution can provide enhanced power density. In some preferred embodiments, the closed container comprises 1.5 mL or less of a liquid. In some preferred embodiments, the closed container has a total internal volume, prior to combining the acid and the carbonate, or 2 mL or less. In some embodiments, the acid and bicarbonate are present as solids and the water is separated from the acid and the bicarbonate.

Formulations for chemical engines may be improved by the addition of a

convection agent. Improved pressure profiles may also be provided where the bicarbonate comprises a solid mixture of at least two types of particle morphologies. The power density is typically used to describe a latent characteristic of a chemical engine; although, less usually, it can be used to describe a system undergoing the chemical reaction. In preferred embodiments, displacement of a plunger or flexible wall in the chemical engine begins with 2 sec, more preferably within 1 sec of the moment when acid, carbonate and solvent (water) are combined; this moment is the moment when the chemical engine is initiated.

In another aspect, the invention provides a chemical engine, comprising: a closed container comprising an acid solution comprising an acid dissolved in water, and bicarbonate, wherein the acid solution is separated from the solid bicarbonate, and a plunger; a conduit comprising apertures disposed within the closed container and adapted such that, following initiation, at least a portion of the acid solution is forced through at least a portion of the apertures. Preferably, the bicarbonate is in particulate form and wherein the conduit comprises a tube having one end that is disposed in the solid bicarbonate such that, when the solution is forced through the apertures it contacts the solid bicarbonate particulate. In some preferred embodiments, at least a portion of the bicarbonate is in solid form disposed inside the conduit. In some embodiments, a spring is adapted to force the acid solution through the conduit.

In some preferred embodiments of any aspect of the invention, a chemical engine has an internal volume of 2 ml or less, in some embodiments, 1.5 ml or less, in some embodiments 1.0 ml or less, and in some embodiments in the range of 0.3 ml to 2 ml, 0.3 ml to 1.5 ml, 0.5 ml to 1.5 ml, or 0.7 ml to 1.4 ml.

In another aspect, the invention provides a chemical engine, comprising:

a closed container comprising an acid solution comprising an acid dissolved in water, and potassium bicarbonate, wherein the acid solution is separated from the potassium bicarbonate, and a plunger; and a mechanism adapted to combine the acid solution and the potassium bicarbonate. In some embodiments, the potassium bicarbonate is mixed with sodium

bicarbonate. The molar ratio of potassium: sodium in the bicarbonate is 100:0, or at least 9, or at least 4, or at least 1; and in some embodiments is at least 0.1, in some embodiments in the range of 0.1 to 9; in some embodiments in the range of 0.5 to 2.

In a further aspect, the invention provides a chemical engine, comprising:

a closed container comprising an acid solution comprising an acid dissolved in water, solid bicarbonate particles, and solid particulate convection agents, wherein the acid solution is separated from the solid bicarbonate, and a plunger; and a mechanism adapted to combine the acid solution and the solid bicarbonate. Solid particulate convection agents are present:

in the range of less than 50 mg per ml of combined solution and at a level selected such that, all other variables being held constant, the generation of C0 ₂ is faster during the first 5 seconds in which the acid solution and solid bicarbonate are combined than the generation of C0 ₂ in the presence of 50 mg per ml of the particulate convection agents; or

in a concentration of 5 mg to 25 mg per ml of combined solution. In some embodiments, 5 mg to 15 mg or 5 mg to 10 mg per ml of combined solution. .

The term "combined solution" means the volume of liquid after the acid solution and solid bicarbonate are mixed. The term "solid bicarbonate" means that there is at least some solid bicarbonate present, although there could also be some liquid phase (typically aqueous phase) present with the bicarbonate. In some preferred embodiments, the bicarbonate is at least 10% present as a solid, in some embodiments at least 50%, at least 90% or substantially 100% present as a solid in the chemical engine prior to combination with the acid solution.

The term "solid convection agents" refers to solid particulates that have a lower solubility than the solid bicarbonate, preferably at least two times slower dissolving than the solid bicarbonate, more preferably at least 10 times slower dissolving than the solid bicarbonate under the conditions present in the chemical engine (or, in the case of the unreacted chemical engine, defined at standard temperature and pressure); in some embodiments at least 100 times slower. The "solid convection agents" preferably have a density, as measured by mercury porisimetry at ambient pressure, that is at least 5%, more preferably at least 10% different from water or the solution in which the convection agents are dispersed. The "solid convection agents" preferably have a density that is at least 1.05 g/ml; more preferably at least 1.1 g/ml; in some embodiments at least 1.2 g/ml, and in some embodiments in the range of 1.1 to 1.5 g/ml. Alternatively, the "solid convection agents" may have a density that is less than water, for example 0.95 g/ml or less, 0.9 g/ml or less, and in some embodiments 0.8 to 0.97 mg/ml. In preferred embodiments, the convection agents are used in a system that is not supersaturated; this is typically in the case in short time scale systems such as an autojector operating over 1 minute or less, preferably 30 seconds or less, more preferably 20 seconds or less, and still more preferably 10 seconds or 5 seconds or less. Thus, the diatomaceous earth formulations of this invention differ from the system of LeFevre in US Patent No. 4,785,972 which uses large quantities of diatomaceous earth to act as a nucleating agent in a supersaturated solution over a large time scale. The present invention includes methods of operating a chemical engine over a short period utilizing a convection agent in which greater than 50% of the bicarbonate (more preferably at least 70% or at least 90%) is converted to gaseous carbon dioxide within a short time scale of a minute or less. Preferably, the diatomaceous earth or other convection agent is present at a level that is at least 50 mass% less than would be used to optimize C0 ₂ discharge from a supersaturated system that is designed to release gaseous C0 ₂ over more than 30 minutes of the C0 ₂ being formed in solution.

In another aspect, the invention provides a chemical engine, comprising: a closed container comprising an acid solution comprising an acid dissolved in water, and solid bicarbonate particles, wherein the acid solution is separated from the solid bicarbonate, and a plunger; a mechanism adapted to combine the acid solution and the solid bicarbonate;

wherein the solid bicarbonate particles comprise a mixture of particle morphologies. In some embodiments, the solid bicarbonate particles are derived from at least two different sources, a first source and a second source, and wherein the first source differs from the second source by at least 20% in one or more of the following characteristics: mass average particle size, surface area per mass, and/or solubility in water at 20 C as measured by the time to complete dissolution into a 1 molar solution in equally stirred solutions using the solvent in the chemical engine (typically water).

In a further aspect, the invention provides a method of ejecting a liquid medicament from a syringe, comprising: providing a closed container comprising an acid solution comprising an acid dissolved in water, and bicarbonate, wherein the acid solution is separated from the bicarbonate, and a plunger; wherein the acid solution and the bicarbonate in the container define a latent power density; wherein the plunger separates the closed container from a medicament compartment; combining the acid solution and the bicarbonate within the closed container;

wherein the acid solution and the bicarbonate react to generate C0 ₂ to power the plunger, which, in turn, pushes the liquid medicament from the syringe; wherein pressure within the container reaches a maximum within 10 seconds after initiation, and wherein, after 5 minutes, the latent power density is 20% or less of the initial latent power density, and wherein, after 10 minutes, the pressure within the closed container is no more than 50% of the maximum pressure. In some preferred embodiments, the closed container further comprises a C0 ₂ removal agent that removes C0 ₂ at a rate that is at least 10 times slower than the maximum rate at which C0 ₂ is generated in the reaction.

Disclosed in some embodiments is a device for delivering a fluid by chemical reaction, comprising: a reagent chamber having a plunger at an upper end and a one- way valve at a lower end, the one-way valve permitting exit from the reagent chamber; a reaction chamber having the one-way valve at an upper end and a piston at a lower end; and a fluid chamber having the piston at an upper end, wherein the piston moves in response to pressure generated in the reaction chamber such that the volume of the reaction chamber increases and the volume of the fluid chamber decreases.

In any of the inventive aspects, the devices or methods can be characterized by one or more of the following characteristics. The reaction chamber preferably has a volume of at most

1.5 cm 3 , in some embodiments at most 1.0 cm 3. Preferably, the fluid chamber contains a high- viscosity fluid having an absolute viscosity of from about 5 centipoise to about 1000 centipoise, or a viscosity of at least 20, preferably at least 40 centipoise, in some embodiments in the range of 20 to 100 cP. The reagent chamber may contain a solvent and/or a bicarbonate or acid dissolved in the solvent. The solvent preferably comprises water. In some preferred

embodiments, the reaction chamber may contain a dry acid powder and a release agent. In some embodiments, the acid powder is citrate and the release agent is sodium chloride. Alternatively, the reaction chamber can contain at least one or at least two chemical reagents that react with each other to generate a gas. The reaction chamber may further comprise a release agent.

In some embodiments, an upper chamber may contain a solvent. The lower chamber may contain at least two chemical reagents that react with each other to generate a gas. The lower chamber may, for example, contain a bicarbonate powder and an acid powder.

The devices may include a piston comprising a push surface at the lower end of the reaction chamber, a stopper at the upper end of the fluid chamber, and a rod connecting the push surface and the stopper. A piston is one type of plunger; however, a plunger often does not contain a rod that connects a push surface and a stopper.

A plunger may include a thumbrest, as well as a pressure lock that cooperates with the upper chamber to lock the plunger in place after being depressed. The pressure lock can be proximate the thumbrest and cooperate with an upper surface of the upper chamber. A plunger that includes a thumbrest can be termed an initiation plunger since it frequently is employed to cause mixing to occur in which an acid and a carbonate are combined in solution.

In some preferred embodiments, a chemical engine may comprise a lower chamber defined by the one-way valve, a continuous sidewall, and a piston, the one-way valve and the sidewall being fixed relative to each other such that the volume of the lower chamber changes only through movement of the piston.

In preferred embodiments, the upper chamber, the lower chamber, and the fluid chamber are cylindrical and are coaxial. The upper chamber, the lower chamber, and the fluid chamber can be separate pieces that are joined together to make the device. A one-way valve can feed a balloon in the lower chamber, the balloon pushing the piston. Sometimes, either the upper chamber or the lower chamber contains an encapsulated reagent.

Also described in various embodiments is a device for delivering a fluid by chemical reaction, comprising: an upper chamber having a seal at a lower end; a lower chamber having a port at an upper end, a ring of teeth at the upper end having the teeth oriented towards the seal of the upper chamber, and a piston at a lower end; and a fluid chamber having the piston at an upper end; wherein the upper chamber moves axially relative to the lower chamber; and wherein the piston moves in response to pressure generated in the lower chamber such that the volume of the reaction chamber increases and the volume of the fluid chamber decreases.

The piston may include a head and a balloon that communicates with the port. The ring of teeth may surround the port. The upper chamber may travel within a barrel of the device.

Sometimes, the upper chamber is the lower end of a plunger. The plunger may include a pressure lock that cooperates with a top end of the device to lock the upper chamber in place after being depressed. Alternatively, the top end of the device can include a pressure lock that cooperates with a top surface of the upper chamber to lock the upper chamber in place when moved sufficiently towards the lower chamber.

A fluid chamber may contain a high- viscosity fluid having a viscosity of at least 5 or at least 20 or at least 40 centipoise. An upper chamber may contain a solvent. A lower chamber may contain at least two chemical reagents that react with each other to generate a gas.

Sometimes, the upper chamber, the lower chamber, and the fluid chamber are separate pieces that are joined together to make the device. In yet other embodiments, either the upper chamber or the lower chamber contains an encapsulated reagent. Also described herein is a device for delivering a fluid by chemical reaction, comprising: an upper chamber; a lower chamber having a piston at a lower end; a fluid chamber having the piston at an upper end; and a plunger comprising a shaft that runs through the upper chamber, a stopper at a lower end of the shaft, and a thumbrest at an upper end of the shaft, the stopper cooperating with a seat to separate the upper chamber and the lower chamber; wherein pulling the plunger causes the stopper to separate from the seat and create fluid communication between the upper chamber and the lower chamber; and wherein the piston moves in response to pressure generated in the lower chamber such that the volume of the reaction chamber increases and the volume of the fluid chamber decreases.

The present disclosure also relates to devices for delivering a fluid by chemical reaction, comprising: a reaction chamber divided by a barrier into a first compartment and a second compartment, the first compartment containing at least two dry chemical reagents that can react with each other to generate a gas, and the second compartment containing a solvent; and a fluid chamber having an outlet; wherein fluid in the fluid chamber exits through the outlet in response to pressure generated in the reaction chamber.

The pressure generated in the reaction chamber may act on a piston or plunger at one end of the fluid chamber to cause fluid to exit through the outlet of the fluid chamber.

In some embodiments, the reaction chamber includes a flexible wall, proximate to the fluid chamber; and wherein the fluid chamber is formed from a flexible sidewall, such that pressure generated in the reaction chamber causes the flexible wall to expand and compress the flexible sidewall of the fluid chamber, thus pushing fluid to exit through the outlet.

The reaction chamber and the fluid chamber may be surrounded by a housing.

Sometimes, the reaction chamber and the fluid chamber are side-by-side in the housing. In other embodiments, a needle extends from a bottom of the housing and is fluidly connected to the outlet of the fluid chamber; and the reaction chamber is located on top of the fluid chamber.

A reaction chamber may be defined by the one-way valve, a sidewall, and a plunger, the one-way valve and the sidewall being fixed relative to each other such that the volume of the reaction chamber changes only through movement of the plunger.

Also disclosed in various embodiments is a device for dispensing a fluid by chemical reaction, comprising: a reaction chamber having first and second ends; a plunger at a first end of the reaction chamber, the plunger being operative to move within the device in response to a pressure generated in the reaction chamber; and a one-way valve at the second end of the reaction chamber permitting entry into the reaction chamber.

The device may comprise a reagent chamber on an opposite side of the one-way valve. The reagent chamber may contain a solvent and a bicarbonate powder dissolved in the solvent. The solvent can comprise water. The device may further comprise a plunger at an end of the reagent chamber opposite the one-way valve. The plunger may cooperate with the reagent chamber to lock the initiation plunger in place after being depressed.

Also disclosed in various embodiments is a device for delivering a fluid by chemical reaction, comprising: a barrel which is divided into a reagent chamber, a reaction chamber, and a fluid chamber by a one-way valve and a piston (or other type of plunger); and an initiation plunger at one end of the reagent chamber; wherein the one-way valve is located between the reagent chamber and the reaction chamber; and wherein the piston separates the reaction chamber and the fluid chamber, the piston (or other type of plunger)being moveable to change the volume ratio between the reaction chamber and the fluid chamber.

The present disclosure also relates to a device for delivering a fluid by chemical reaction, comprising: a barrel containing a reaction chamber and a fluid chamber which are separated by a moveable piston; and a thermal source for heating the reaction chamber. The reaction chamber may contain at least one chemical reagent that generates a gas upon exposure to heat. The at least one chemical reagent can be 2,2'- azobisisobutyronitrile. The generated gas can be nitrogen gas.

The present disclosure also describes a device for delivering a fluid by chemical reaction, comprising: a barrel containing a reaction chamber and a fluid chamber which are separated by a moveable piston; and a light source that illuminates the reaction chamber. The reaction chamber may contain at least one chemical reagent that generates a gas upon exposure to light. The at least one chemical reagent can comprise silver chloride.

The initiation of a gas generating reaction in a chemical engine can be performed by dissolving at least two different chemical reagents in a solvent. The at least two chemical reagents can include a chemical compound having a first dissolution rate and the same chemical compound having a second different dissolution rate. The dissolution rates can be varied by changing the surface area of the chemical compound, or by encapsulating the chemical compound with a coating to obtain the different dissolution rate. The pressure versus time profile from a gas generating may include a burst in which the rate of gas generation increases at a rate faster than the initial generation of gas.

The reaction chamber may contain a dry acid reagent, with a solvent containing a predissolved bicarbonate (or a predissolved acid) being added to the reaction chamber from a reagent chamber on an opposite side of the one-way valve to initiate the reaction. The reaction chamber can further comprise a release agent, such as sodium chloride. The solvent may comprise water. In some preferred embodiments, the dry acid reagent is a citric acid powder or an acetic acid powder. The gas produced is preferably carbon dioxide.

Also described herein are a device for delivering a fluid by chemical reaction, comprising: a barrel containing a reagent chamber, a reaction chamber, and a fluid chamber; wherein the reagent chamber is located within a push button member at a top end of the barrel; a plunger separating the reagent chamber from the reaction chamber; a spring biased to push the initiation plunger into the reagent chamber when the push button member is depressed; and a piston separating the reaction chamber from the fluid chamber, wherein the piston moves in response to pressure generated in the reaction chamber. The push button member can comprise a sidewall closed at an outer end by a contact surface, a lip extending outwards from an inner end of the sidewall, and a sealing member proximate a central portion on an exterior surface of the sidewall. The barrel may include an interior stop surface that engages the lip of the push button member.

The initiation plunger may comprise a central body having lugs extending radially therefrom, and a sealing member on an inner end which engages a sidewall of the reaction chamber. The interior surface of the push button member can include channels for the lugs.

A reaction chamber may be divided into a mixing chamber and an arm by an interior radial surface, the interior radial surface having an orifice, and the piston being located at the end of the arm.

As a general feature a reaction chamber sometimes includes a gas permeable filter covering an orifice that allows gas to escape after a plunger has been moved to force fluid out a fluid chamber. This feature provides a release for excess gas.

The barrel can be formed from a first piece and a second piece, the first piece including the reagent chamber and the reaction chamber, and the second piece including the fluid chamber. The invention includes methods of making injectors comprising the assembly of the first and second piece into an injector or injector component.

Also disclosed in different embodiments is an injection device for delivering a

pharmaceutical fluid to a patient by means of pressure produced by an internal chemical reaction, comprising: a reagent chamber having an activator at an upper end and a one- way valve at a lower end, the one-way valve permitting exit of a reagent from the reagent chamber into a reaction chamber upon activation; the reaction chamber operatively connected to the reagent chamber, having means for receiving the one-way valve at an upper end and a piston at a lower end; and a fluid chamber operatively connected to the reaction chamber, having means for receiving the piston at an upper end, wherein the piston moves in response to pressure generated in the reaction chamber such that the volume of the reaction chamber increases and the volume of the fluid chamber decreases.

It is intended that, in various embodiments, the invention includes all combinations and permutations of the various features that are described herein. For example, the formulations described herein can be employed in any of the devices as would be understood by a skilled person reading these descriptions. Likewise, for every device described herein, there is a corresponding method of using the device to deliver a viscous fluid, typically a medicant. The invention also includes methods of making the devices comprising assembling the components.

The invention further includes the separate chemical engine component and kits including the chemical engine and other components that are assembled into an injector. The invention may be further characterized by the measurements described herein; for example, the power density characteristic or any other measured characteristics described in the figures, examples or elsewhere. For example, characterized by an upper or lower limit or range established by the measured values described herein.

Various aspects of the invention are described using the term "comprising;" however, in narrower embodiments, the invention may alternatively be described using the terms "consisting essentially of or, more narrowly, "consisting of."

In any of the chemical engines, there can be an initiation plunger that is typically directly or indirectly activated to initiate the gas generation in the chemical engine; for example, the cause an acid and bicarbonate to combine in solution and react to generate C0 ₂ . Preferably, the chemical engine comprises a feature (such as lugs) that lock the initiator plunger in place so that the reaction chamber remains closed to the atmosphere and does not lose pressure except to move a plunger to force a fluid out of the fluid compartment.

In various aspects, the invention can be defined as a formulation, injector, method of making a formulation or injector (typically comprising an injector body, an expansion compartment, a plunger (for example, a piston), and a viscous fluid component that is preferably a medicant. Typically, of course, a needle is connected to the medicament compartment. In some embodiments, the expansion compartment can be releasably attached such that the expansion compartment piece (also called the reaction chamber) can be detached from medicament compartment. In some aspects, the invention can be defined as a method of pushing a solution through a syringe, or a method of administering a medicament, or a system comprising apparatus plus formulation(s) and/or released gas (typically C0 ₂ ). Medicaments can be conventional medicines or, in preferred embodiments, biologic(s) such as proteins. Any of the inventive aspects can be characterized by one feature or any combination of features that are described anywhere in this description.

Preferred aspects of the invention provide a chemical engine that can provide

• Delivery of viscous fluids (e.g. greater than 20 cP) with a flow rate of 0.06 mL/sec or higher

• Energy on demand to eliminate the need to store energy

• Minimal start-up forces to prevent syringe breakages

· Relatively constant pressure through-out the injection event to prevent stalling

• Tunable pressure or pressure profile, depending on the viscosity of the fluid or user requirements

Our invention provides a gas -generating chemical-reaction to create pressure on-demand that may be used to deliver pharmaceutical formulations by parenteral delivery. The pressure can be produced by combining two reactive materials that generate a gas. An advantage of our gas- generating reaction over prior art is that this can be done in a manner that realizes rapid delivery (less than 20 sec) of fluids with viscosity greater than 20 cP and minimize the packaging space required, while maintaining a substantially flat pressure versus time profile as shown in the Examples. A further advantage of our invention is that pressure versus time profile can be modified for different fluids, non-Newtonian fluids, patient needs, or devices. BRIEF DESCRIPTION OF THE DRAWINGS

The following is a brief description of the drawings, which are presented for the purposes of illustrating the exemplary embodiments disclosed herein and not for the purposes of limiting the same.

FIG. 1 is a diagram of a chemical reaction that produces a gas for moving a piston within a chamber.

FIG. 2 is a diagram of a first embodiment of a device for delivering a fluid by chemical reaction. The chemical reaction here is generated when two dry chemical reagents are dissolved in a solvent and react. This figure shows the device in a storage state, where the dry reagents are separated from the solvent.

FIG. 3 is a diagram showing the device of FIG. 2 after the dry reagents are combined with the solvent.

FIG. 4 is diagram showing the device of FIG. 2 with the piston being pushed by gas pressure to deliver the fluid.

FIG. 5 is a diagram showing another exemplary embodiment of a device for delivering a fluid by chemical reaction of two reagents in a solvent. This device is made in four separate pieces that are joined together to form a combined device similar to that shown in FIG. 2.

FIG. 6 is a diagram of a first embodiment of a device for delivering a fluid by chemical reaction. The chemical reaction here is generated when a chemical reagent is exposed to heat. The device includes a thermal source.

Figure 7 is a side cross-sectional view of a first exemplary embodiment of an injection device. This embodiment uses a one-way valve to create two separate chambers.

Figure 8 is a cross-sectional perspective view of the engine in an exemplary embodiment of Figure 7.

Figure 9 is a side cross-sectional view of a second exemplary embodiment of an injection device. This embodiment uses a seal to create two separate chambers, and a ring of teeth to break the seal.

Figure 10 is a cross-sectional perspective view of the engine in the second exemplary

embodiment of Figure 9. Figure 11 is a side cross-sectional view of a third exemplary embodiment of an injection device. In this embodiment, pulling the handle upwards (i.e. away from the barrel of the device) breaks the seal between two separate chambers. This figure shows the device prior to pulling the handle upwards.

Figure 12 is a cross-sectional perspective view of the engine in the third exemplary embodiment of Figure 11 prior to pulling the handle upwards.

Figure 13 is a cross-sectional perspective view of the engine in the third exemplary embodiment of Figure 11 after pulling the handle upwards.

Figure 14 is a side cross-sectional view of an exemplary embodiment showing the engine using an encapsulated reagent. This figure shows the device in a storage state.

Figure 15 is a side cross-sectional view of the exemplary embodiment showing the engine using an encapsulated reagent. This figure shows the device in a use state.

Figure 16 is a perspective see-through view of a first exemplary embodiment of a patch pump that uses a chemical reaction to inject fluid. Here, the engine and the fluid chamber are side-by- side, and both have rigid sidewalls.

Figure 17 is a perspective see-through view of a second exemplary embodiment of a patch pump that uses a chemical reaction to inject fluid. Here, the engine is on top of the fluid chamber, and both have a flexible wall. The engine expands and presses the fluid chamber. This figure shows the patch pump when the fluid chamber is empty and prior to use.

Figure 18 is a perspective see-through view of the patch pump of figure 17, where the fluid chamber is filled.

FIG. 19 is a side cross-sectional view of another exemplary embodiment of a syringe that uses a gas -generating chemical reaction. Here, a stopper is biased by a compression spring to travel through the reagent chamber and ensure its contents are emptied into the reaction chamber. FIG. 20 is a bottom view showing the interior of the push button member in the syringe of FIG. 19.

FIG. 21 is a top view of the stopper used in the syringe of FIG. 19.

FIG. 22 is a graph showing the pressure versus time profile for delivery of silicone oil when different amounts of water are injected into a reaction chamber. The y- axis is Gauge Pressure (Pa), and the x-axis is Time (sec). The plot shows results for conditions where three different amounts of water were used - 0.1 mL, 0.25 mL, and 0.5 mL. FIG. 23 is a graph showing the volume versus time profile for delivery of 73cP silicone oil when a release agent (NaCl) is added to the reaction chamber. The y-axis is Volume (ml), and the x- axis is Time (sec).

FIG. 24 is a volume vs. time graph for delivery of a 73 cP silicone fluid in which the use of modifying or mixing bicarbonate morphology is shown: reaction chamber contains 100% as received, 100% freeze-dried, 75% as-received / 25% freeze dried, or 50% as-received / 50% freeze dried.

FIG. 25 is a pressure vs. time graph for delivery of a 73 cP silicone fluid in which the use of modifying or mixing bicarbonate morphology is shown: reaction chamber contains 100% as received, 100% freeze-dried, 75% as-received / 25% freeze dried, or 50% as-received / 50% freeze dried.

FIG. 26 is a normalized pressure vs. time graph delivery of a 73 cP silicone fluid during an initial time period. The use of modifying or mixing bicarbonate morphology is shown: reaction chamber contains 100% as received, 100% freeze-dried, 75% as-received / 25% freeze dried, or 50% as-received / 50% freeze dried.

FIG. 27 is a normalized pressure vs. time graph delivery of a 73 cP silicone fluid during a second time period. The reaction chamber contained bicarbonates with different morphology or mixed morphology: 100% as received, 100% freeze-dried, 75% as-received / 25% freeze dried, and 50% as-received / 50% freeze dried.

FIG. 28 is a volume vs. time graph for delivery of a 73 cP silicone fluid in which the use of reagents with different dissolution rate or structure is shown. The engine contained either 100% as-received baking soda and citric acid powder, 100% alka seltzer adjusted for similar stoichiometric ratio, 75% as-received powders / 25% Alka Seltzer, 50%, 25% as-received powders / 75% alka seltzer.

FIG. 29 is a volume vs. time graph for delivery of a 73 cP silicone fluid in which the use of reagents with different dissolution rate or structure is shown. The engine contained either 100% as-received baking soda and citric acid powder, 100% alka seltzer adjusted for similar stoichiometric ratio, 75% as-received powders / 25% Alka Seltzer, 50%, 25% as-received powders / 75% alka seltzer.

FIG. 30 is a pressure vs. time graph for the delivery of a 1 cP water fluid in which the use of reagents with different dissolution rate or structure is shown. The engine contained either 100% as-received baking soda and citric acid powder, 100% alka seltzer adjusted for similar stoichiometric ratio, 75% as-received powders / 25% Alka Seltzer, 50%, 25% as-received powders / 75% alka seltzer.

FIG. 31 is a normalized pressure vs. time graph for delivery of a 73 cP silicone fluid in which the use of reagents with different dissolution rate or structure is shown. The engine contained either 100% as-received baking soda and citric acid powder, 100% alka seltzer adjusted for similar stoichiometric ratio, 75% as-received powders / 25% Alka Seltzer, 50%, 25% as-received powders / 75% alka seltzer. The pressure is normalized by normalizing the curves in Figure 29 to their maximum pressure .

FIG. 32 is a normalized pressure vs. time graph expanding the first 3 seconds of FIG. 31.

FIG. 33 is a volume vs. time graph for delivery of a 73 cP silicone fluid in which the reaction chamber contained either sodium bicarbonate (BS), potassium bicarbonate, or a 50/50 mixture. FIG. 34 is a pressure vs. time graph for the third set of tests; delivery of a 73 cP silicone fluid in which the reaction chamber contained either sodium bicarbonate (BS), potassium bicarbonate, or a 50/50 mixture.

FIG. 35 is a reaction rate graph for the third set of tests; delivery of a 73 cP silicone fluid in which the reaction chamber contained either sodium bicarbonate (BS), potassium bicarbonate, or a 50/50 mixture.

FIG. 36 is a volume vs. time graph for a fourth set of tests for silicone oil.

FIG. 37 illustrates a conduit with apertures that may be used to deliver a solution into a reaction chamber.

FIG. 38 shows the measured pressure versus time profile for a chemical engine using mixed sodium and potassium bicarbonate delivering silicone oil through a 19 mm long and 27 gauge thin wall needle.

FIG. 39. Shows volume versus delivery time for a control (No NaCl) and a system using solid NaCl as a nucleating agent (NaCl).

FIG. 40 shows the force versus time profile from a two different chemical engines delivering 1 mL of 50 cP fluid through a 27 gauge thin wall needle (1.9 cm long).

FIG. 41 shows the force versus time profile from a two different chemical engines (Formulation 3 and 4) delivering 3 mL of 50 cP fluid through a 27 gauge thin wall needle (1.9 cm long). FIG. 42 shows the force versus time profile from chemical engine (Formulation 5) delivering 3 mL of 50 cP fluid through a 27 gauge thin wall needle (1.9 cm long).

FIG. 43 shows the pressure profiles at constant for the compositions described on the right-hand side of the figure. The observed pressure increased in the order control < nucleating surface « Expancel particles < oxalic acid < diatomaceous earth < calcium oxalate.

FIG. 44 shows the similarity of the effect of vibration and added diatomaceous earth in enhancing gas generation.

FIG. 45 illustrates displacement of a viscous fluid powered by a chemical engine containing potassium bicarbonate, citric acid and 5, 10 or 50 mg of diatomaceous earth acting as a convection agent.

FIG. 46 schematically illustrates apparatus for measuring the power density of a chemical engine.

GLOSSARY

Chemical engine - a chemical engine generates a gas through a chemical reaction and the generated gas is used to power another process. Typically, the reaction is not combustion, and in many preferred embodiments the chemical engine is powered by the generation of C0 ₂ from the reaction of a carbonate (typically sodium or, preferably, potassium carbonate) with an acid, preferably citric acid.

In the context of a chemical engine, a closed container prevents escape of gas to the atmosphere so that the force of the generated gas can be applied against a plunger. In the assembled device (typically an injector) the plunger is moved by the generated gas and forces fluid out of the fluid compartment. In many embodiments, the container is closed at one end by a one way valve, surrounded by chamber walls in the directions perpendicular to the central axis, and closed at the other end by a moveable plunger.

Viscosity can be defined in two ways: "kinematic viscosity" or "absolute viscosity." Kinematic viscosity is a measure of the resistive flow of a fluid under an applied force. The SI unit of kinematic viscosity is mm /sec, which is 1 centistoke (cSt). Absolute viscosity, sometimes called dynamic or simple viscosity, is the product of kinematic viscosity and fluid density. The SI unit of absolute viscosity is the millipascal- second (mPa-sec) or centipoise (cP), where 1 cP=l mPa-sec. Unless specified otherwise, the term viscosity always refers to absolute viscosity. Absolute viscosity can be measured by capillary rheometer, cone and plate rheometer, or any other known method.

Fluids may be either Newtonian or non-Newtonian. Non-Newtonian fluids should be characterized at different shear rates, including one that is similar to the shear rate of the injection. In this case, the viscosity of the fluid can be approximated using the Hagen-Poiseuille equation, where a known force is used for injection through a known needle diameter and length, at known fluid rate. The invention is suitable for either Newtonian or non-Newtonian fluids.

A plunger (also called an "expansion plunger") is any component that moves or deforms in response to C0 ₂ generated in the chemical engine and which can transmit force, either directly or indirectly, to a liquid in a compartment that is either adjacent to or indirectly connected to the chemical engine. For example, the plunger could push against a piston that, in turn, pushes against a liquid in a syringe. There are numerous types of plungers described in this application and in the prior art, and the inventive formulations and designs are generally applicable to a multitude of plunger types.

An initiation plunger is a moveable part that is used to initiate a reaction, usually by directly or indirectly causing the combination of acid, carbonate and water. Preferably, the initiation plunger locks into place to prevent any loss of pressure and thus direct all the generated pressure toward the fluid to be ejected from the fluid chamber.

The term "parenteral" refers to a delivery means that is not through the gastrointestinal tract, such as injection or infusion.

The pressure versus time profile may include a burst, where burst is described as a second increase in pressure during the delivery profile.

The processes of the present disclosure can be used with both manual syringes or auto- injectors and is not limited to cylindrical geometries. The term "syringe" is used interchangeably to refer to manual syringes and auto-injectors of any size or shape. The term "injection device" is used to refer to any device that can be used to inject the fluid into a patient, including for example syringes and patch pumps. In preferred embodiments, any of the chemical engines described herein may be part of an injector and the inention includes these injectors.

DETAILED DESCRIPTION FIG. 1 illustrates the generation of pressure by a chemical reaction for use in delivering a pharmaceutical formulation by injection or infusion. Referring to the left hand side of the figures, one or more chemical reagents 100 are enclosed within a reaction chamber 110. One side of the chamber can move relative to the other sides of the chamber, and acts as a piston 120. The chamber 110 has a first volume prior to the chemical reaction.

A chemical reaction is then initiated within the chamber, as indicated by the "RXN" arrow. A gaseous byproduct 130 is generated at some rate, n(t), where n represents moles of gas produced and t represents time. The pressure is proportional to the amount of gaseous byproduct 130 generated by the chemical reaction, as seen in Equation (1):

P{t) - [ n{t) ^« T] I V (1)

In Equation (1), T represents temperature and V represents the volume of the chamber 110.

The volume of the chamber 110 remains fixed until the additional force generated by the gas pressure on the piston 120 exceeds that needed to push the fluid through a syringe needle. The necessary force depends on the mechanical components present in the system, e.g. frictional forces and mechanical advantages provided by the connector design, the syringe needle diameter, and the viscosity of the fluid.

Once the minimum pressure required to move the piston 120 is exceeded, the volume of the reaction chamber 110 begins to increases. The movement of the piston 120 causes delivery of fluid within the syringe to begin. The pressure in the chamber 110 depends on both the rate of reaction and the rate of volume expansion, as represented by Equation (1). Preferably, sufficient gas is generated to account for the volume expansion, while not generating too much excess pressure. This can be accomplished by controlling the rates of reaction and gas release in the chamber 110.

The pressure build-up from the chemical-reaction produced gaseous byproduct 130 can be used to push fluid directly adjacent to the piston 120 through the syringe. Pressure build-up may also push fluid in an indirect fashion, e.g., by establishing a mechanical contact between the piston 120 and the fluid, for example by a rod or shaft connecting the piston 120 to a stopper of a prefilled syringe that contains fluid.

The one or more chemical reagents 100 are selected so that upon reaction, a gaseous byproduct 130 is generated. Suitable chemical reagents 100 include reagents that react to generate a gaseous byproduct 130. For example, citric acid (C H^O^ or acetic acid (C ₂ H ₄ O ₂ ) will react with sodium bicarbonate (NaHC0 ₃ ) to generate carbon dioxide, C0 ₂ , which can be initiated when the two reagents are dissolved in a common solvent, such as water. Alternatively, a single reagent may generate a gas when triggered by an initiator, such as light, heat, or dissolution. For, example, the single reagent 2,2'-azobisisobutyronitrile (AIBN) can be decomposed to generate nitrogen gas (N ₂ ) at temperatures of 50°C - 65°C. The chemical reagent(s) are selected so that the chemical reaction can be easily controlled.

One aspect of the present disclosure is the combination of various components to result in (i) enough force to deliver a viscous fluid in a short time period, e.g. under 20 s, and (ii) in a small package that is compatible with the intended use, i.e. driving a syringe. In time, size, and in force must all come together to achieve the desired injection. The package size for the chemical engine is defined by the volume of the reagents, including solvents; this is measured at standard conditions (25 C, 1 atm) after all components have been mixed and C0 ₂ released.

The molar concentration of C0 ₂ versus H ₂ C0 ₃ is given by pH. The pKa of H ₂ C0 ₃ is 4.45. For pH well below this value, the percentage of C02 to H2C03 is almost 100%. For pH close to the pKa (e.g. 4.5 to 6.5), the value will decrease from 90% to 30%. For pH greater than 7, the system will consist of mostly H ₂ C0 ₃ and no C0 ₂ . Suitable acids should thus provide buffering of the system to maintain the pH below 4.5 throughout the duration of the injection event. The chemical reaction need not go to 100% conversion during the injection event (e.g., within 5 seconds, or within 10 seconds or within 20 seconds) in the time frame of the injection event, but generally, to minimize generation of excess pressure, conversion should approach at least 30-50%; where conversion is defined as the molar percentage of acid that has been reacted, and in some embodiments in the range of 30 to 80%, in some embodiments in the range of 30 to 50%. In some embodiments, conversion of the C0 ₂ reactant (such as sodium or potassium bicarbonate) is at least 30%, preferably at least 50%, or at least 70% and in some embodiments less than 95%.

Acids that are liquid at room temperature such as glacial acetic acid (pKal 4.76) and butyric acid (pKa 4.82) are suitable. Preferred acids are organic acids that are solid at room temperature; these acids have little odor and do not react with the device. In addition, they can be packaged as powders with different morphology or structure to provide a means of controlling dissolution rate. Preferred acids include citric acid (pH: 2.1 to 7.2 (pKa: 3.1 ; 4.8; 6.4), oxalic acid (pH: 0.3 to 5.3 [pKa: 1.3; 4.4]), tartartic acid (pH: 2 to 4; [pKal= 2.95; pKa2= 4.25]), and phthalic acid (pH: 1.9 - 6.4 [pKa: 2.9; 5.4]). In experiments with added HC1. It was surprisingly discovered that reducing the pH to 3 did not speed the release of C0 ₂ .

In some preferred embodiments, citric acid is used in systems where the injection event occurs at reaction conversions between 15 to 50%. It may be desirable to take advantage of the rapid rate of C0 ₂ build-up, and thus pressure, that occurs at low conversion. Due to the buffering behavior of this acid, the percentage of C0 ₂ to H ₂ C0 ₃ created during the reaction will decrease at high conversion, as the pH increases above 5.5. This system will have less pressure build-up at the back-end of the reaction, after completion of the injection. In other circumstances, it may be desirable to take advantage of the complete reaction cycle and modify the reaction so that it is close to completion at the end of the injection event. In some embodiments, tartaric acid and oxalic acids are preferred choices due to their lower pKa values.

In some preferred embodiments, the bicarbonate is added as a saturated solution to an expansion compartment containing solid acid or solid acid mixed with other components, such as salts, bicarbonate, or other additives. In other embodiments, water is added to the piston containing solid acid, mixed with bicarbonates and other components. In some other preferred embodiments, a solution of aqueous bicarbonate is added to a solid composition comprising solid bicarbonate to provide additional C0 ₂ production at the end of the injection. In still other embodiments, the bicarbonate can be present as a wetted, or only partly dissolved solid. Any form of bicarbonate can also be reacted with a dissolved acid. For example, in some preferred embodiments, bicarbonate is combined with a solution of citric acid.

When the chemistry is confined to small reaction volumes, i.e., a relatively large amount of reagent is confined to a small volume of liquid (saturated solutions) in a pressurized system, the process to produce C0 ₂ (g) becomes considerably more complex. Depending on the circumstances, important rate limiting steps now become:

· dissolution rate of solid reagents

• availability and diffusion rate of bicarbonate ions

• desorption rate of C0 ₂ from bicarbonate surface

• release of C0 ₂ (g) from solution

Depending on the needs of the system, the parameters may be tuned independently or in concert to minimize overshoot and maintain a flat pressure profile curve, where the impact of volume expansion of the chemical reaction chamber does not cause the pressure to drop and delivery to stall.

To achieve fast delivery of viscous fluids, the availability of bicarbonate ion can be an important factor. In solution, bicarbonate salts are in equilibrium with bicarbonate ions, which are the active species in the reaction. Bicarbonate ions can be free species or strongly associated. The concentration of bicarbonate can be controlled by altering solvent polarity - such as adding ethanol to decrease the reaction rate or adding N-methylformamide or N-methylacetamide to increase the reaction rate; taking advantage of common ion effects; and utilizing relatively high content of bicarbonate above the saturation point. The bicarbonate preferably has a solubility in water higher than 9 g in 100 mL, more preferably 25 g in 100 mL. In some preferred

embodiments, saturated solutions of potassium bicarbonate are added to the piston containing acids. In other embodiments, water is added to a piston containing solid acids and potassium bicarbonate.

The pressure profile during delivery can be modified by modifying the rate of dissolution. For example, the addition of a saturated solution of potassium bicarbonate to a piston containing solid citric acid and solid bicarbonate provides first a rapid burst of C0 ₂ as the dissolved bicarbonate reacts with acid and second a secondary sustained level of C0 ₂ as solid bicarbonate is dissolved and becomes available. Dissolution rates can be modified by changing the particle size or surface area of the powder, employing several different species of bicarbonate or acid, encapsulation with a second component, or changes in the solvent quality. By combining powders that have different dissolution rates, the pressure versus time profile can be modified, enabling constant pressure with time or a burst in the pressure with time. Introduction of a catalyst can be used to the same effect.

The pressure in the piston (reaction chamber) is determined by the concentration of C0 ₂ that is released from the solution. Release can be facilitated by introducing methods of agitation or by introducing sites that decrease solubility of C0 ₂ or enhance its nucleation, growth, and diffusion. Methods of agitation may include introduction of rigid spheres suspended in the piston. Suitable spheres include hollow polymeric microspheres such as Expancel, polystyrene microspheres, or polypropylene microspheres. Upon introduction of water or saturated bicarbonate to the piston, the external flow induces forces and torques on the spheres, resulting in their rotations with velocity w as well as they start to move induced by buoyancy force. The flow field generated by rotation of sphere improves gas diffusion toward surface and facilitating C0 ₂ desorption from liquid. The surface of the freely rotating spheres may also be modified by an active layer, such as by coating with bicarbonate. Such spheres are initially heavy and unaffected by buoyancy force. However as coating dissolves or reacts with acid, buoyancy will begin to cause spheres to move toward surface of liquid. During aforementioned motion the unbalance forces on the particle promotes spinning, reducing gas transport limitation and increasing C0 ₂ desorption from liquid.

In some embodiments, a salt, additive or other nucleating agent is added to facilitate release of the dissolved gas into the empty volume. Examples of said nucleating agents include crystalline sodium chloride, calcium tartarate, calcium oxalate, and sugar. Release can be facilitated by adding a component that decreases the solubility of the gas. It can also be facilitated by adding a nucleating agent that facilitates the nucleation, growth, and release of gas bubbles via heterogeneous nucleation.

In preferred embodiments, the flow rate versus time is maintained substantially constant, such that the shear rate on the fluid is similar. For Newtonian fluids, the shear rate is proportional to flow rate and inversely proportional to r , where r is the radius of the needle. The change in shear rate can be defined once flow is initiated after the first 2 or 3 seconds as [(Flow Rate Max - Flow Rate Min) / Flow Rate Min)] -100 for devices where the needle diameter does not change. In preferred embodiments, the change in shear rate is less than 50%, more preferably less than 25%. The fluids may be Newtonian or non-Newtonian. For flow rates typical in subcutaneous delivery through needles with diameters in the 27 gauge to 31 gauge range, shear rates are on the order of lx 10 ⁴ to lx 10 ⁴ s ^"1 and non-Newtonian effects could become important, particularly for proteins.

In some examples described further herein, an injection device using a gas- generating chemical reaction was used to displace fluid having a viscosity greater than 70 centipoise (cP) through a 27 gauge thin- wall (TW) needle in less than 10 seconds. A 27 gauge thin- wall needle has a nominal outer diameter of 0.016 + 0.0005 inches, a nominal inner diameter of 0.010 + 0.001 inches, and a wall thickness of 0.003 inches. Such results are expected to also be obtained with needles having larger nominal inner diameters.

The selection of the chemical reagent(s) can be based on different factors. One factor is the dissolution rate of the reagent, i.e. the rate at which the dry powder form of the reagent dissolves in a solvent, such as water. The dissolution rate can be modified by changing the particle size or surface area of the powder, encapsulating the powder with a coating that dissolves first, or changes in the solvent quality. Another factor is the desired pressure versus time profile. The pressure versus time profile can be controlled by modifying the kinetics of the reaction. In the simplest case, the kinetics of a given reaction will depend on factors such as the concentration of the reagents, depending on the "order" of the chemical reaction, and the temperature. For many reagents 100, including those in which two dry reagents must be mixed, the kinetics will depend on the rate of dissolution. For example, by combining powders that have two different dissolution rates, the pressure versus time profile can be modified, enabling constant pressure over time or a profile having a burst in pressure at a specified time.

Introduction of a catalyst can be used to the same effect. Alternatively, a delivered volume versus time profile can have a constant slope. The term "constant" refers to the given profile having a linear upward slope over a time period of at least 2 seconds, with an acceptable deviation in the value of the slope of + 15%.

This ability to tune the chemical reaction allows the devices of the present disclosure to accommodate different fluids (with varying volumes and/or viscosities), patient needs, or delivery device designs. Additionally, while the chemical reaction proceeds independently of the geometry of the reaction chamber, the shape of the reaction chamber can affect how accumulated pressure acts on the piston.

The target pressure level for providing drug delivery may be determined by the mechanics of the syringe, the viscosity of the fluid, the diameter of the needle, and the desired delivery time. The target pressure is achieved by selecting the appropriate amount and stoichiometric ratio of reagent, which determines n (moles of gas), along with the appropriate volume of the reaction chamber. The solubility of the gas in any liquid present in the reaction chamber, which will not contribute to the pressure, should also be considered.

If desired, a release agent may be present in the reaction chamber to increase the rate of fluid delivery. When a solvent, such as water, is used to facilitate diffusion and reaction between molecules, the generated gas will have some solubility or stability in the solvent. The release agent facilitates release of any dissolved gas into the head space of the chamber. The release agent decreases the solubility of the gas in the solvent. Exemplary release agents include a nucleating agent that facilitates the nucleation, growth, and release of gas bubbles via heterogeneous nucleation. An exemplary release agent is sodium chloride (NaCl). The presence of the release agent can increase the overall rate of many chemical reactions by increasing the dissolution rate, which is often the rate limiting factor for pressure generation for dry (powder) reagents. The release agent may also be considered to be a catalyst.

In some preferred embodiments, the volume of the reaction chamber is 1 to 1.4 cm or less, in some preferred embodiments 1 cm or less, in some embodiments in the range of 0.5 to 1 or 1.4 cm . The other components of the device can be dimensioned to match the volume of the reaction chamber. A reaction chamber no more than 1 to 1.4 cm allows enables chemical- reaction delivery of a high- viscosity fluid with a limited injection space or footprint.

FIG. 2 illustrates one exemplary embodiment of a device (here, a syringe) that can be used to deliver a high-viscosity fluid using a chemical reaction between reagents to generate a gas. The syringe 400 is depicted here in a storage state or a non-depressed state in which the chemical reaction has not yet been initiated. The needle is not included in this illustration. The syringe 400 includes a barrel 410 that is formed from a sidewall 412, and the interior space is divided into three separate chambers. Beginning at the top end 402 of the barrel, the syringe includes a reagent chamber 420, a reaction chamber 430, and a fluid chamber 440. The plunger 470 is inserted into an upper end 422 of the reagent chamber. A one-way valve 450 is present at a lower end 424 of the reagent chamber, forming a radial surface. The one-way valve 450 is also present at the upper end 432 of the reaction chamber. The one-way valve 450 is directed to permit material to exit the reagent chamber 420 and to enter the reaction chamber 430. The lower end 434 of the reaction chamber is formed by a piston 460. Finally, the piston 460 is present at the upper end 442 of the fluid chamber. The orifice 416 of the barrel is at the lower end 444 of the fluid chamber, and at the bottom end 404 of the syringe. It should be noted that the one-way valve 450 is fixed in place and cannot move within the barrel 410. In contrast, the piston 460 can move within the barrel in response to pressure. Put another way, the reaction chamber 430 is defined by the one-way valve 450, the barrel sidewall 412, and the piston 460.

The reaction chamber 430 can also be described as having a first end and a second end. The moveable piston 460 is at the first end 434 of the reaction chamber, while the one- way valve 450 is present at the second end 432 of the reaction chamber. In this illustration, the reaction chamber 430 is directly on one side of the piston 460, and the fluid chamber 440 is directly on the opposite side of the piston.

The reagent chamber 420 contains at least one chemical reagent, a solvent, and/or a release agent. The reaction chamber 430 contains at least one chemical reagent, a solvent, and/or a release agent. The fluid chamber 440 contains the fluid to be delivered. As depicted here, the reagent chamber 420 contains a solvent 480, the reaction chamber 430 contains two different chemical reagents 482, 484 in a dry powder form, and the fluid chamber 440 contains a high- viscosity fluid 486. Again, it should be noted that this figure is not drawn to scale. The chemical reagents, as illustrated here, do not fill up the entire volume of the reaction chamber. Instead, a head space 436 is present within the reaction chamber.

In specific embodiments, the reagent chamber contains a bicarbonate which has been pre-dissolved in a solvent, and the reaction chamber contains a dry acid powder. Itwas found that passive mixing of reagents in the solvent was a problem that would reduce the speed of reaction. Bicarbonate was pre-dissolved, otherwise it was too slow to dissolve and participate in the gas generating reaction. In more specific embodiments, potassium bicarbonate was used. It was found that sodium bicarbonate did not react as quickly. Citrate was used as the dry acid powder because it was fast- dissolving and fast-reacting. Sodium chloride (NaCl) was included as a dry release agent with the citrate. The sodium chloride provided nucleation sites to allow the gas to evolve from solution more quickly.

Each chamber has a volume, which in the depicted illustration is proportional to the height of the chamber. The reagent chamber 420 has a height 425, the reaction chamber 430 has a height 435, and the fluid chamber 440 has a height 445. In this non-depressed state, the volume of the reaction chamber is sufficient to contain the solvent and the two chemical reagents.

In particular embodiments, the volume of the reaction chamber is 1 cm3 or less. The other components of the device can be dimensioned to match the volume of the reaction chamber. A reaction chamber no more than 1 cm3 allows enables chemical-reaction delivery of a high-viscosity fluid with a limited injection space or footprint.

In FIG. 3, the plunger 470 has been depressed, i.e. the syringe is in a depressed state. This action causes the one-way valve 450 to be opened, and the solvent 480 enters into the reaction chamber 430 and dissolves the two chemical reagents (illustrated now as bubbles in the solvent). After the plunger 470 is depressed and no further pressure is being exerted on the one-way valve, the one-way valve 450 closes (this figure shows the valve in an open state). In particular embodiments, the barrel sidewall 412 at the lower end 424 of the reagent chamber may contain grooves 414 or is otherwise shaped to capture the plunger 470. Put another way, the plunger 470 cooperates with the lower end 424 of the reagent chamber 420 to lock the plunger in place after being depressed.

In FIG. 4, the dissolution of the two chemical reagents in the solvent has resulted in the generation of a gas 488 as a byproduct of the chemical reaction. As the amount of gas increases, the pressure exerted on the piston 460 increases until, after reaching a threshold value, the piston 460 moves downward towards the bottom end 404 of the syringe (as indicated by the arrow). This causes the volume of the reaction chamber 430 to increase, and the volume of the fluid chamber 440 to decrease. This results in the high- viscosity fluid 486 in the fluid chamber being dispensed through the orifice. Put another way, the combined volume of the reaction chamber 430 and the fluid chamber remains constant, but the volume ratio of reaction chamber to fluid chamber 440 will increase as gas is generated in the reaction chamber. Note that the one-way valve 450 does not permit the gas 488 to escape from the reaction chamber into the reagent chamber.

The syringe can provide consistent force when the following elements are properly controlled: (i) the particle size of the dry powder reagent; (ii) the solubility of the reagents; (iii) the mass of the reagents and the quantity of release agent; and (iv) the shape configuration of the chambers for consistent filling and packaging.

FIG. 5 illustrates another variation of a device 700 that uses a chemical reaction between reagents to generate gas. This illustration is in a storage state. Whereas the barrel of FIG. 2 is shown as being made from an integral sidewall, the barrel in the device of FIG. 5 is made of several shorter pieces. This construction can simplify manufacturing and filling of the various chambers of the overall device. Another large difference in this variation is that the piston 760 is made up of three different parts: a push surface 762, a rod 764, and a stopper 766, as explained further herein.

Beginning at the top of FIG. 5, the reagent chamber 720 is made from a first piece 726 that has a first sidewall 728 to define the sides of the reagent chamber. The plunger 770 is inserted in the upper end 722 of the piece to seal that end. The first piece 720 can then be turned upside down to fill the reagent chamber 720 with the solvent 780. A second piece 756 containing the one-way valve 750 can then be joined to the lower end 724 of the first piece to seal the reagent chamber 720. A second sidewall 758 surrounds the oneway valve. The lower end 724 of the first piece and the upper end 752 of the second piece can be joined using known means, such as screw threads (e.g. a Luer lock). As illustrated here, the lower end of the first piece would have internal threads, while the upper end of the second piece would have the external threads.

The third piece 736 is used to form the reaction chamber 730, and is also formed from a third sidewall 738. The push surface 762 of the piston is located within the third sidewall 738. After placing the chemical reagents, solvent, and/or release agent upon the push surface, the lower end 754 of the second piece and the upper end 732 of the third piece are joined together. Two reagents 782, 784 are depicted here. The rod 764 of the piston extends down from the push surface 762.

Finally, the fourth piece 746 is used to form the fluid chamber 740. This fourth piece is formed from a fourth sidewall 748 and a conical wall 749 that tapers to form the orifice 716 from which fluid will be expelled. The orifice is located at the lower end 744 of the fluid chamber. The fluid chamber can be filled with the fluid to be delivered, and the stopper 766 can then be placed in the fluid chamber. As seen here, the stopper 766 may include a vent hole 767 so that air can escape from the fluid chamber as the stopper is being pushed down to the surface of the fluid 786 to prevent air from being trapped in the fluid chamber. A cap 768 attached to the lower end of the piston rod 764 can be used to cover the vent hole 767. Alternatively, the lower end of the piston rod can be inserted into the vent hole. The lower end 734 of the third piece and the upper end 742 of the fourth piece are then joined together.

As previously noted, the piston 760 in this variation is formed from the push surface 762, the rod 764, and the stopper 766 being connected together. An empty volume 790 is thus present between the reaction chamber 730 and the fluid chamber 740. The size of this empty volume can be varied as desired. For example, it may be useful to make the overall device longer so that it can be more easily grasped by the user. Otherwise, this variation operates in the same manner as described above with regards to FIGS. 2-4. The push surface portion of the piston acts in the reaction chamber, and the stopper portion of the piston acts in the fluid chamber. It should also be noted that the push surface, rod, stopper, and optional cap can be one integral piece, or can be separate pieces. FIG. 6 illustrates an exemplary embodiment of a device (again, a syringe) that can be used to deliver a high-viscosity fluid using a chemical reaction initiated by heat to generate a gas. Again, the syringe 800 is depicted here in a storage state.

The barrel 810 is formed from a sidewall 812 and the interior space is divided into two separate chambers, a reaction chamber 830 and a fluid chamber 840. The reaction chamber 830 is present at an upper end 802 of the syringe. The upper end 832 of the reaction chamber is formed by a radial wall 838. Located within the reaction chamber is a thermal source 850 that can be used for heating. The thermal source 850 may be located on the radial wall 838 or, as depicted here, on the barrel sidewall 812.

The lower end 834 of the reaction chamber is formed by a piston 860. The reaction chamber 830 is defined by the radial wall 838, the barrel sidewall 812, and the piston 860. The piston 860 is also present at the upper end 842 of the fluid chamber. The orifice 816 of the barrel is at the lower end 844 of the fluid chamber, i.e. at the lower end 804 of the syringe. Again, only the piston 860 portion of the reaction chamber can move within the barrel 810 in response to pressure. The radial wall 838 is fixed in place, and is solid so that gas cannot pass through.

The reaction chamber contains a chemical reagent 882. For example, the chemical reagent can be 2,2'-azobisisobutyronitrile. A head space 836 may be present in the reaction chamber. The fluid chamber 840 contains a fluid 886.

An activation trigger 852 is present on the syringe, which can be for example on top near the finger flange 815 or on the external surface 816 of the barrel sidewall. When activated, the thermal source 850 generates heat. The thermal source can be, for example, an infrared light emitting diode (LED). The chemical reagent 882 is sensitive to heat, and generates a gas (here, N2). The pressure generated by the gas causes the piston 860 to move, expelling the high- viscosity fluid 886 in the fluid chamber 840.

It should be noted again that the piston may alternatively be the push surface, rod, and stopper version described in FIG. 5. This version may be appropriate here as well.

In an alternative embodiment, of a device that can be used to deliver a high- viscosity fluid using a chemical reaction initiated by light to generate a gas. This embodiment is almost identical to the version described in FIG. 6, except that the thermal source is now replaced by a light source 850 which can illuminate the reaction chamber 830. The chemical reagent 884 here is sensitive to light, and generates a gas upon exposure to light. For example, the chemical reagent may be silver chloride (AgCl). The pressure generated by the gas causes the piston to move, expelling the high-viscosity fluid in the fluid chamber. The piston version of FIG. 5 can also be used here if desired.

Any suitable chemical reagent or reagents can be used to generate a gas. For example, bicarbonate will react with acid to form carbon dioxide. Sodium, potassium, and ammonium bicarbonate are examples of suitable bicarbonates. Suitable acids could include acetic acid, citric acid, potassium bitartrate, disodium pyrophosphate, or calcium dihydrogen phosphate. Any gas can be generated by the chemical reaction, such as carbon dioxide, nitrogen gas, oxygen gas, chlorine gas, etc. Desirably, the generated gas is inert and non-flammable. Metal carbonates, such as copper carbonate or calcium carbonate, can be decomposed thermally to produce C0 ₂ and the corresponding metal oxide. As another example, 2,2'-azobisisobutyronitrile (AIBN) can be heated to generate nitrogen gas. As yet another example, the reaction of certain enzymes (e.g. yeast) with sugar produces C0 ₂ . Some substances readily sublime, going from solid to gas. Such substances include but are not limited to naphthalene and iodine. Hydrogen peroxide can be decomposed with catalysts such as enzymes (e.g. catalase) or manganese dioxide to produce oxygen gas.

It is contemplated that the high-viscosity fluid to be dispensed using the devices of the present disclosure can be a solution, dispersion, suspension, emulsion, etc. The high-viscosity formulation may contain a protein, such as a monoclonal antibody or some other protein which is therapeutically useful. The protein may have a concentration of from about 150 mg/ml to about 500 mg/ml. The high-viscosity fluid may have an absolute viscosity of from about 5 centipoise to about 1000 centipoise. In other embodiments, the high-viscosity fluid has an absolute viscosity of at least 40 centipoise, or at least 60 centipoise. The high-viscosity fluid may further contain a solvent or non-solvent, such as water, perfluoroalkane solvent, safflower oil, or benzyl benzoate.

Figure 7 and Figure 8 are different views of the first exemplary embodiment of an injection device (here, a syringe) that can be used to deliver a high- viscosity fluid using a chemical reaction between reagents to generate a gas. The syringe 300 is depicted here in a storage state or a non-depressed state in which the chemical reaction has not yet been initiated. Figure 7 is a side cross-sectional view, and Figure 8 is a perspective view of the engine of the syringe. The syringe 300 includes a barrel 310 whose interior space is divided into three separate chambers.

Beginning at the top end 302 of the barrel, the syringe includes an upper chamber 320, a lower chamber 330, and a fluid chamber 340. These three chambers are coaxial, and are depicted here as having a cylindrical shape. The lower chamber may also be considered a reaction chamber.

The plunger 370 is inserted into an upper end 322 of the upper chamber, and the stopper 372 of the plunger travels through only the upper chamber. A one-way valve 350 is present at a lower end 324 of the upper chamber, forming a radial surface. The one-way valve 350 is also present at the upper end 332 of the lower chamber. The one-way valve 350 is directed to permit material to exit the upper chamber 320 and to enter the lower chamber 330. A piston 360 is present at the lower end 334 of the lower chamber. The piston 360 is also present at the upper end 342 of the fluid chamber. As illustrated here, the piston is formed of at least two pieces, a push surface 362 that is at the lower end of the lower chamber and a head 366 at the upper end of the fluid chamber. The needle 305 is at the lower end 344 of the fluid chamber, and at the bottom end 304 of the syringe. It should be noted that the one-way valve 350 is fixed in place and cannot move within the barrel 310, or in other words is stationary relative to the barrel. In contrast, the piston 360 can move within the barrel in response to pressure. Put another way, the lower chamber 330 is defined by the one-way valve 350, the continuous sidewall 312 of the barrel, and the piston 360.

The lower chamber 330 can also be described as having a first end and a second end. The moveable piston 360 is at the first end 334 of the lower chamber, while the one-way valve 350 is present at the second end 332 of the lower chamber. In this illustration, the lower chamber 330 is directly on one side of the piston 360, and the fluid chamber 340 is directly on the opposite side of the piston.

As previously noted, the piston 360 is formed from at least the push surface 362 and the head 366. These two pieces can be connected together physically, for example with a rod (not shown) that has the push surface and the head on opposite ends. Alternatively, it is also contemplated that an incompressible gas could be located between the push surface and the head. An empty volume 307 would thus be present between the lower chamber 330 and the fluid chamber 340. The size of this empty volume could be varied as desired. For example, it may be useful to make the overall device longer so that it can be more easily grasped by the user.

Alternately, as illustrated in another embodiment in Figure 9 and Figure 10 further herein, the piston may use a balloon that acts as the push surface and acts upon the head 366. As yet another variation, the piston may be a single piece, with the push surface being on one side of the single piece and the head being on the other side of the single piece.

The upper chamber 320 contains at least one chemical reagent or a solvent. The lower chamber 330 contains at least one chemical reagent or a solvent. The fluid chamber 340 contains the fluid to be delivered. It is generally contemplated that dry reagents will be placed in the lower chamber, and a wet reagent (i.e. solvent) will be placed in the upper chamber. As depicted here, the upper chamber 320 would contain a solvent, the lower chamber 330 would contain two different chemical reagents in a dry powder form, and the fluid chamber 340 would contain a high- viscosity fluid. The reagent(s) in either chamber may be encapsulated for easier handling during manufacturing. Each chamber has a volume, which in the depicted illustration is proportional to the height of the chamber. In this non-depressed state, the volume of the lower chamber is sufficient to contain the solvent and the two chemical reagents.

When the plunger in the syringe of Figure 7 and Figure 8 is depressed, the additional pressure causes the one-way valve 350 to open, and the solvent in the upper chamber 320 enters into the lower chamber 330 and dissolves the two chemical reagents. After the plunger 370 is sufficiently depressed and no further pressure is being exerted on the one-way valve, the one- way valve 350 closes. As illustrated here, the plunger includes a thumbrest 376 and a pressure lock 378 on the shaft 374 which is proximate to the thumbrest. The pressure lock cooperates with an upper surface 326 of the upper chamber to lock the plunger in place. The two chemical reagents may react with each other in the solvent to generate gas in the lower chamber. As the amount of gas increases, the pressure exerted on the push surface 362 of the piston 360 increases until, after reaching a threshold value, the piston 360 moves downward towards the bottom end 304 of the syringe. This causes the volume of the lower chamber 330 to increase, and the volume of the fluid chamber 340 to decrease. This results in the high- viscosity fluid in the fluid chamber being dispensed through the orifice (by the head 366). Put another way, the combined volume of the lower chamber 330 and the fluid chamber remains constant, but the volume ratio of lower chamber to fluid chamber 340 will increase as gas is generated in the reaction chamber. Note that the one-way valve 350 does not permit the gas to escape from the lower chamber into the upper chamber. Also, the pressure lock 378 on the plunger permits the stopper 372 to act as a secondary backup to the one-way valve 350, and also prevents the plunger from being pushed up and out of the upper chamber.

In some embodiments, the upper chamber contains a bicarbonate which has been pre- dissolved in a solvent, and the lower chamber contains a dry acid powder. Passive mixing of reagents in the solvent, i.e. combining both dry powders into the reaction chamber and adding water, reduces the speed of reaction. One reagent may be predissolved. For example, bicarbonate may be pre-dissolved. In more specific embodiments, potassium bicarbonate is recommended. Sodium bicarbonate does not react as quickly; therefore C0 ₂ production and injection rates are slower. Citric acid is a preferred dry acid powder because it dissolves well and is fast-reacting, as well as safe. Sodium chloride (NaCl) is included as a dry release agent with the citric acid. The sodium chloride provided nucleation sites and changed the ionic strength to allow the gas to evolve from solution more quickly.

It should be noted that the upper chamber 320, the lower chamber 330, and the fluid chamber 340 are depicted here as being made from separate pieces that are joined together to form the syringe 300. The pieces can be joined together using methods known in the art. For example, the upper chamber is depicted here as being formed from a sidewall 325 having a closed upper end 322 with a port 327 for the plunger. The stopper 372 of the plunger is connected to the shaft 374. The one-way valve 350 is a separate piece which is inserted into the open lower end 324 of the upper chamber. The lower chamber is depicted here as being formed from a sidewall 335 having an open upper end 332 and an open lower end 334. The upper end of the lower chamber and the lower end of the upper chamber cooperate to lock together and fix the one-way valve in place. Here, the locking mechanism is a snap fit arrangement, with the upper end of the lower chamber having the cantilever snap 380 that includes an angled surface and a stop surface. The lower end of the upper chamber has the latch 382 that engages the cantilever snap. Similarly, the lower chamber and the fluid chamber are fitted together with a ring-shaped seal.

Figure 9 and Figure 10 are different views of an exemplary embodiment of an injection device of the present disclosure. The syringe 500 is depicted here in a storage state or a non- depressed state in which the chemical reaction has not yet been initiated. Figure 9 is a side cross- sectional view, and Figure 10 is a perspective view of the engine of the syringe. Again, the syringe includes a barrel 510 whose interior space is divided into three separate chambers. Beginning at the top end 502 of the barrel, the syringe includes an upper chamber 520, a lower chamber 530, and a fluid chamber 540. These three chambers are coaxial, and are depicted here as having a cylindrical shape. The lower chamber 530 may also be considered a reaction chamber.

In this embodiment, the upper chamber 520 is a separate piece located within the barrel 510. The barrel is illustrated here as an outer sidewall 512 that surrounds the upper chamber. The upper chamber 520 is illustrated here with an inner sidewall 525 and a top wall 527. A shaft 574 and a thumbrest I button 576 extend from the top wall 527 of the upper chamber in the direction away from the barrel. Thus, the upper chamber 520 could also be considered as forming the lower end of a plunger 570. The lower end 524 of the upper chamber is closed off with a seal 528, i.e. a membrane or barrier such that the upper chamber has an enclosed volume. It should be noted that the inner sidewall 525 of the upper chamber travels freely within the outer sidewall 512 of the barrel. The upper chamber moves axially relative to the lower chamber.

The lower chamber 530 has a port 537 at its upper end 532. A ring 580 of teeth is also present at the upper end 532. Here, the teeth surround the port. Each tooth 582 is illustrated here as having a triangular shape, with a vertex oriented towards the seal 528 of the upper chamber, and each tooth is angled inwards towards the axis of the syringe. The term "tooth" is used here generally to refer to any shape that can puncture the seal of the upper chamber.

A piston 560 is present at the lower end 534 of the lower chamber 530. The piston 560 is also present at the upper end 542 of the fluid chamber 540. Here, the piston 560 includes the head 566 and a balloon 568 within the lower chamber that communicates with the port 537 in the upper end. Put another way, the balloon acts as a push surface for moving the head. The head 566 may be described as being below or downstream of the balloon 568, or alternatively the balloon 568 can be described as being located between the head 566 and the port 537. The needle 505 is at the lower end 544 of the fluid chamber, and at the bottom end 504 of the syringe. The balloon is made from a suitably non-reactive material.

The top end 502 of the barrel (i.e. the sidewall) includes a pressure lock 518 that cooperates with the top surface 526 of the upper chamber to lock the upper chamber 520 in place when moved sufficiently towards the lower chamber 530. The upper chamber 520 is illustrated here extending out of the outer sidewall 512. The top end 526 of the outer sidewall is shaped to act as the cantilever snap, and the top surface 526 of the upper chamber acts as the latch.

Alternatively, the top end of the device may be formed as depicted in Figure 8, with the pressure lock on the shaft proximate to the thumbrest and cooperating with the top end of the device.

As previously described, it is generally contemplated that dry reagents will be placed in the lower chamber 530, and a wet reagent (i.e. solvent) will be placed in the upper chamber 520. Again, the reagent(s) in either chamber may be encapsulated for easier handling during manufacturing. More specifically, it is contemplated that the reagents in the lower chamber would be located within the balloon 568.

During operation of the syringe of Figure 9 and Figure 10, pushing the button 576 downwards causes the upper chamber 520 to move into the barrel towards the ring 580 of teeth. The pressure of the upper chamber against the ring of teeth causes the seal 528 to break, releasing the contents of the upper chamber into the lower chamber 530. Here, it is contemplated that the gas-generating reaction occurs within the balloon 568. The increased gas pressure causes the balloon to inflate (i.e. lengthen). This pushes the head 566 towards the bottom end 504 of the syringe (note the upper chamber will not be pushed out of the barrel due to the pressure lock). This again causes the volume of the lower chamber 530 to increase, and the volume of the fluid chamber 540 to decrease, i.e. the volume ratio of lower chamber to fluid chamber to increase.

There is an empty volume 507 present between the balloon 568 and the head 566. An incompressible gas could be located in this empty volume. The size of this empty volume can be varied as desired, for example to make the overall device longer.

Again, the upper chamber 520, the lower chamber 530, and the fluid chamber 540 can be made from separate pieces that are joined together to form the syringe. It should be noted that Figure 10 is made from five pieces (590, 592, 594, 596, and 598), with the additional pieces being due to the addition of the balloon in the lower chamber and to the upper chamber being separate from the outer sidewall. However, this embodiment could still be made from fewer pieces as in Figure 8. For example, the balloon could be located close to the ring of teeth.

Figure 11, Figure 12, and Figure 13 are different views of a third exemplary embodiment of an injection device of the present disclosure. In this embodiment, the mixing of the chemical reagents is initiated by pulling the plunger handle away from the barrel, rather than towards the barrel as in the embodiments of Figures 7-10. Figure 11 is a side cross-sectional view of the syringe in a storage state. Figure 12 is a perspective view of the engine of the syringe in a storage state. Figure 13 is a perspective view of the engine of the syringe in its operating state, i.e. when the handle is pulled upwards away from the barrel of the syringe.

The syringe 700' includes a barrel 710' whose interior space is divided into three separate chambers. Beginning at the top end 702' of the barrel, the syringe includes an upper chamber 720', a lower chamber 730', and a fluid chamber 740'. These three chambers are coaxial, and are depicted here as having a cylindrical shape. The lower chamber may also be considered a reaction chamber.

In this embodiment, the plunger 770' is inserted into an upper end 722' of the upper chamber. In the storage state, the shaft 774' runs through the upper chamber from the lower end 724' to the upper end 722' and through the upper surface 726' of the upper chamber. A seal 728' is present at the top end where the shaft exits the upper chamber. The thumbrest 776' at the upper end of the shaft is outside of the upper chamber. The stopper 772' at the lower end of the shaft cooperates with a seat 716' within the barrel such that the upper chamber has an enclosed volume. For example, the top surface of the stopper may have a larger diameter than the bottom surface of the stopper. The seat 716' may be considered as being at the lower end 724' of the upper chamber, and also as being at the upper end 732' of the lower chamber.

A piston 760' is present at the lower end 734' of the lower chamber. The piston 760' is also present at the upper end 742' of the fluid chamber 740' . As illustrated here, the piston 760' is formed of at least two pieces, a push surface 762' and a head 766' . An empty volume 707' can be present. Other aspects of this piston are similar to that described in Figure 8. Again, the piston can move within the barrel in response to pressure. The lower chamber 730' can also be described as being defined by the seat 716, the continuous sidewall 712' of the barrel, and the piston 760'. The needle 705' is at the lower end 744' of the fluid chamber, and at the bottom end 704' of the syringe.

During operation of the syringe of Figures 11-13, it is generally contemplated that dry reagents will be placed in the lower chamber 730', and a wet reagent (i.e. solvent) will be placed in the upper chamber 720', as previously described. Referring now to Figure 11, pulling the plunger 770' upwards (i.e. away from the barrel) causes the stopper 772' to separate from the seat 716'. This creates fluid communication between the upper chamber 720' and the lower chamber 730' . The reagent in the upper chamber travels around the stopper into the lower chamber (reference number 717'). The gas- generating reaction then occurs in the lower chamber 730'. The gas pressure pushes the piston 760' towards the bottom end 704' of the syringe. In other words, the volume of the lower chamber increases, and the volume of the fluid chamber decreases, i.e. the volume ratio of lower chamber to fluid chamber increases. One additional advantage to this embodiment is that once the reagents begin generating gas, the pressure created will continue to push the plunger 710' further out of the upper chamber, helping to push more reagent out of the upper chamber 720' into the lower chamber 730', furthering the generation of gas.

Referring to Figure 12, the barrel 710' is depicted as being made up of three different pieces 790', 792', 794' . A seal 738' is also located between the pieces that make up the lower chamber and the fluid chamber.

Figure 14 and Figure 15 are cross-sectional views of one aspect of another exemplary embodiment of the injection device of the present disclosure. In this embodiment, the liquid reagent (i.e. the solvent) is encapsulated in a capsule is broken when a button is pressed. Figure

14 shows this engine before the button is pressed. Figure 15 shows the engine after the button is pressed.

Referring first to Figure 14, the top end 1002 of the syringe 1000 is shown. A reaction chamber 1030 contains a capsule 1038 and dry reagent(s) 1039. Here, the capsule rests on a ledge 1031 above the dry reagent(s). A push surface 1062 of a piston 1060 is present at the lower end of the reaction chamber. The head 1066 of the piston is also visible, and is at the upper end 1042 of the fluid chamber 1040. A button/plunger 1070 is located above the capsule. A seal 1026 may be present between the button 1070 and the capsule 1038. The barrel contains a safety snap 1019 to prevent the button from falling out of the end of the barrel.

If desired, the portion of the reaction chamber containing the capsule could be considered an upper chamber, and the portion of the reaction chamber containing the dry reagent(s) could be considered a lower chamber.

Referring now to Figure 15, when the button 1070 is pushed, the capsule 1038 is broken, causing the solvent and the dry reagent(s) to mix. This generates a gas that pushes the piston 1060 downward and ejects fluid from the fluid chamber 1040. Pushing the button subsequently engages a pressure lock 1018 that prevents the button from being pushed upwards by the gas pressure.

The embodiments of the figures described above have been illustrated as auto-injectors. Auto-injectors are typically held in the user's hand, have a cylindrical form factor, and have a relatively quick injection time of one second to 30 seconds. It should be noted that the concepts embodied in the above-described figures could also be applied to other types of injection devices, such as patch pumps. Generally, a patch pump has a flatter form factor compared to a syringe, and also has the delivery time is typically greater than 30 seconds. Advantages to using a chemical gas-generating reaction in a patch pump include the small volume required, flexibility in the form/shape, and the ability to control the delivery rate.

Figure 16 is an illustration of a typical bolus injector 1200. The bolus injector includes a reaction chamber 1230 and a fluid chamber 1240 located within a housing 1280. As shown here, the reaction chamber and the fluid chamber are located side-by- side, though this can vary as desired. The reaction chamber 1230 is formed from a sidewall 1235. The fluid chamber 1240 is also formed from a sidewall 1245. The reaction chamber and the fluid chamber are fluidly connected by a passage 1208 at a first end 1202 of the device. The fluid chamber 1240 includes an outlet 1246 that is connected to a needle 1205 located at opposite second end 1204 of the housing. The needle 1205 extends from the bottom 1206 of the housing.

The reaction chamber is divided into a first compartment and a second compartment by a barrier (not visible). In this regard, the first compartment is analogous to the lower chamber, and the second chamber is analogous to the upper chamber previously described.

The reaction chamber can be considered as an engine that causes fluid in the fluid chamber to be ejected. In this regard, it is contemplated that a gas-generating chemical reaction can be initiated by breaking the seal between the first compartment and the second compartment. The barrier could be broken, for example, by bending or snapping the patch pump housing, or by pushing at a designated location on the housing. This causes the reagents to mix. Because the desired delivery time is longer, the speed at which the chemicals are mixed is not as great a concern. The pressure builds up and can act on a piston (not visible) in the fluid chamber, causing fluid to exit through the outlet. It is contemplated that the volume of the reaction chamber and the fluid chamber do not change significantly in this embodiment. Figure 17 and Figure 18 are perspective see-through views of another exemplary embodiment of a patch pump. In this embodiment, the reaction chamber/engine 1230 is located on top of the fluid chamber 1240. The needle 1205 extends from the bottom 1206 of the housing 1280. In this embodiment, the reaction chamber 1230 includes a flexible wall 1235. The fluid chamber 1240 also includes a flexible sidewall 1245. The flexible wall of the reaction chamber is proximate to the flexible sidewall of the fluid chamber. The reaction chamber and the fluid chamber are not fluidly connected to each other in this embodiment. Instead, it is contemplated that as gas is generated in the reaction chamber, the reaction chamber will expand in volume. The flexible wall 1235 of the reaction chamber will compress the flexible sidewall 1245 of the fluid chamber, causing fluid in the fluid chamber to exit through the outlet 1246. Put another way, the volume ratio of reaction chamber to fluid chamber increases over time as the reaction chamber inflates and the fluid chamber dispenses fluid. It should be noted that a relatively constant volume is required in this embodiment, so that the increasing volume of the reaction chamber causes compression of the fluid chamber. This can be accomplished, for example, by including a rigid backing on the opposite side of the reaction chamber from the flexible wall, or by making the housing from a relatively rigid material.

FIG. 19 illustrates another exemplary embodiment of a device (here, a syringe) that can be used to deliver a high-viscosity fluid using a chemical reaction between reagents to generate a gas. The syringe 1300 is depicted here in a storage state or a non-depressed state in which the chemical reaction has not yet been initiated. The needle is not included in this illustration.

The syringe 1300 includes a barrel 1310 whose interior space is divided into three separate chambers. Beginning at the top end 1302 of the barrel, the syringe includes a reagent chamber 1320, a reaction chamber 1330, and a fluid chamber 1340. These three chambers are coaxial, and are depicted here as having a cylindrical shape. In this embodiment, the barrel of the syringe is formed from two different pieces. The first piece 1380 includes a sidewall 1312 that forms the reaction chamber and provides a space 1313 for the reagent chamber. The sidewall is open at the top end 1302 for a push button described further herein. The fluid chamber is made from a second piece 1390 which can be attached to the first piece.

The sidewall 1312 of the first piece includes an interior radial surface 1314 that divides the first piece into an upper space 1313 and the reaction chamber 1330. The reaction chamber has a smaller inner diameter 1325 compared to the inner diameter 1315 of the upper space. The reagent chamber is located in a separate push button member 1350 that is located within the upper space 1313 of the first piece and extends through the top end 1302 of the barrel. As illustrated here, the push button member is formed from a sidewall 1352 which is closed at the outer end 1351 by a contact surface 1354, and which forms an interior volume into which reagent is placed (i.e. the reagent chamber). A sealing member 1356 (shown here as an 0-ring) is proximate a central portion on the exterior surface 1355 of the sidewall, and engages the sidewall 1312 in the upper space. The inner end 1353 of the sidewall includes a lip 1358 extending outwards from the sidewall. The lip engages an interior stop surface 1316 on the barrel. The reagent chamber is depicted as containing a solvent 1306 in which bicarbonate is dissolved.

A plunger 1370 is located between the reagent chamber 1320 and the reaction chamber

1330. The plunger 1370 is located at the inner end 1324 of the reagent chamber. The plunger includes a central body 1372 having lugs 1374 extending radially therefrom (here shown as four lugs, though the number can vary). The lugs also engage the lip 1358 of the push button member when the syringe is in its storage state. The lugs are shaped with an angular surface 1376, such that the plunger 1370 rotates when the push button member 1350 is depressed. An inner end 1373 of the central body includes a sealing member 1378 (shown here as an O-ring) which engages the sidewall in the reaction chamber.

The reaction chamber 1330 includes a top end 1332 and a bottom end 1334. Another interior radial surface 1336 is located at a central location in the reaction chamber, separating the reaction chamber into a mixing chamber 1335 and an arm/fitting 1333, with the mixing chamber 1335 being proximate the reagent chamber 1320 or the top end 1332. An orifice 1331 within the interior radial surface leads to the arm fitting 1333 which engages the second piece 1390 containing the fluid chamber 1340. The piston 1360 is located at the bottom end of the reaction chamber, i.e. at the end of the arm 1333. Located within the reaction chamber is a dry reagent 1308. Here, the dry reagent is citrate, and is in the form of a tablet. The dry reagent is depicted here as being located upon the interior radial surface, i.e. in the mixing chamber. A gas- permeable I liquid-solid impermeable filter 1337 may be present across the orifice. The filter keeps any dry solid reagent and a liquid inside the mixing chamber to improve mixing.

In addition, a compression spring 1395 is located within the mixing chamber, extending from the interior radial surface 1336 to the inner end 1373 of the plunger. A compression spring stores energy when compressed (i.e. is longer when no load is applied to it). Because the push button member 1350 and the plunger 1370 are fixed in place, the compression spring 1395 is compressed in the storage state. It should be noted that here, the spring surrounds the dry reagent. It is also contemplated, in alternate embodiments, that the dry reagent is attached to the inner end 1373 of the plunger.

Finally, the piston 1360 is also present at the upper end 1342 of the fluid chamber. Again, the piston 1360 can move within the barrel in response to pressure generated in the reaction chamber. The piston can also be described as having a push surface 1362 and a stopper 1364.

The sealing member 1378 of the plunger separates the liquid reagent in the reagent chamber 1320 from the dry reagent in the reaction chamber 1330. While liquid 1306 is illustrated as being present in the push button member, it is also possible that liquid is present in the barrel in the upper space 1313 around the plunger.

When the push button member 1350 is depressed (down to the interior radial surface 1316), the plunger 1370 is rotated. This causes the lugs 1374 of the plunger to disengage from the lip 1358 of the push button member. In addition, it is contemplated that the push button member, once depressed, cannot be retracted from the barrel. This can be done, for example, using a stop surface near the outer end of the barrel (not shown).

When the plunger 1370 is no longer held in place by the push button member, the compression spring extends and pushes the plunger 1370 into the push button member 1350. It is contemplated that the compression spring is sized so that the plunger travels completely through the push button member, but will not push through the contact surface 1354 of the push button member. The liquid 1306 present in the reagent chamber falls into the reaction chamber and contacts the dry reagent 1308. The movement of the plunger into the push button member is intended to cause complete emptying of the contents of the reagent chamber into the reaction chamber. This mechanism can also provide forceful mixing of the wet reagent with the dry reagent, either induced by the spring action, initial chemical action, or both.

In some alternate embodiments, the spring also pushes at least some of the dry reagent into the reagent chamber (i.e. the interior volume of the push button member). For example, the dry reagent could be attached to the inner end 1373 of the plunger, and driven upwards by the spring.

FIG. 20 is a bottom view illustrating the interior of the push button member. As seen here, the interior surface 1357 of the sidewall forming the push button member includes four channels 1359 through which the lugs of the plunger can travel. FIG. 21 is a top view of the plunger 1370, showing the central body 1372 and the lugs 1374, which can travel in the channels of the push button member. Comparing these two figures, the outer circle of FIG. 20 is the lip 1358 of the push button member and has an outer diameter 1361. The inner diameter 1363 of the push button member is interrupted by the four channels. The dotted circle indicates the outer diameter 1365 of the sidewall exterior surface 1355. The central body of the plunger has a diameter 1375 which is less than the inner diameter 1363 of the push button member, with the lugs fitting into the channels. This permits the plunger to push the liquid in the push button member out and around the central body. It should be noted that the channels do not need to be straight, as illustrated here. For example, the channels may be angled to one side, i.e. twist in a helical manner. This might be desirable to add turbulence to the liquid reagent and improve mixing.

The combination of the solvent with bicarbonate and the citrate in the reaction chamber 1330 causes gas 1309 to be generated. It should be noted that due to the movement of the plunger, the reagent chamber could now be considered to be part of the reaction chamber. In addition, it should be noted that the dry reagent 1308 in FIG. 19 could be considered as restricting access to the orifice 1331. Upon dissolution, the orifice is clear and gas can enter the bottom end 1334 of the reaction chamber.

Once a threshold pressure is reached, the piston 1360 travels through the fluid chamber 1340, ejecting fluid from the syringe. The needle 1305 of the syringe is visible in this figure.

In some alternate contemplated embodiments, the diameter of the plunger including the lugs is less than the inner diameter 1363 of the push button member. In other words, channels are not needed on the inner sidewall of the push button member. In such embodiments, the barrel sidewall would provide a surface that holds the plunger in place until the push button is depressed to rotate the plunger. The shape and movement of the plunger would then cause turbulence in the liquid as the wet reagent flowed past the lugs into the reaction chamber. It is also contemplated that a stem could be attached to the plunger that extends into the reagent chamber, or put another way, the stem is attached to the outer end of the plunger. The stem may be shaped to cause turbulence and improve mixing.

To speed gas generation in a chemical engine, a conduit comprising apertures such as that shown in Fig. 37 can be utilized. In conduit 3700, flow 3703 is directed into the conduit at an inlet and then flows out through plural apertures 3705 (preferably 5 or more apertures) into a reaction chamber. The conduit-with-apertures configuration is especially advantageous where the reaction chamber comprises a powder where flow from apertures directly contacts the powder. For example, an acid solution (preferably a citric acid solution) flows through the conduit and out from the apertures where it contacts and agitates solid bicarbonate particles. In some preferred embodiments, a plunger (such as a spring-activated plunger) forces a liquid solution through the conduit. In some cases, the conduit contains a solid (preferably solid bicarbonate) which at least partly dissolves as the solution flows through the conduit; this may offer the dual advantage of both enhanced dissolution as well the creation of gas bubbles in the conduit that enhance mixing as they pass through the apertures and into the reaction chamber. Any of the devices described herein for adding a solution into a reaction chamber can be used to direct fluid down this conduit.

It is also contemplated that the speed of the injection could be adjusted by the user. One way of doing this would be to control the speed at which the dry reagent and a wet reagent are mixed. This would adjust the speed of the gas-generating chemical reaction, and therefore the speed at which the force that pushes the piston is generated. This could be accomplished, for example, by adjusting the size of the opening between the reagent chamber and the reaction chamber. For example, an adjustable aperture could be placed beneath the plunger. The aperture would have a minimum size (to accommodate the spring), but could otherwise be adjusted. Another way of adjusting the speed of the injection would be to control the size of the reaction chamber. This would adjust the pressure generated by the chemical reaction (because pressure is force per area). For example, the sidewall of the reaction chamber could move inwards or outwards as desired to change the volume of the reaction chamber. Alternately, the interior radial surface 1336 could include an adjustable aperture to change the size of the orifice 1331 and the rate at which gas can enter the bottom end 1334 of the reaction chamber and push on the piston 1360. Both of these methods could be controlled by a dial on the syringe, which could mechanically adjust the speed of injection as desired by the user. This would allow "on-the-fly" adjustment of the speed of injection. Like other features described herein, this feature of user- controlled adjustable speed is a general feature that can be applied in any of the devices described herein. It is also contemplated that a gas-permeable liquid-solid impermeable filter may be present that separates the piston from the lower chamber in the injection devices described herein. In this regard, the dry powder has been found in some situations to stick to the sides of the chamber. When the piston moves, remaining solvent falls below the level of the powder, such that further chemical reaction does not occur. It is believed that the filter should keep any dry solid reagent and liquid within the lower chamber to improve mixing.

Suitable materials for the injection devices of the present disclosure are known in the art, as are methods for making the injection devices.

The gas-generating chemical reaction used to generate force "on demand", as opposed to springs, which only store energy when compressed. Most autoinjectors hold a spring in a compressed position during "on the shelf storage, causing parts to fatigue and to form over time. Another alternative to compressing the spring in manufacturing is to provide a cocking mechanism that compresses the spring prior to use. This adds another step to the process for using the spring-driven device. In addition, physically disabled users may have difficulty performing the cocking step. For example, many users of protein drugs are arthritic, or have other conditions that limit their physical abilities. The force needed to activate the gas-generating chemical reaction can be far less than that required to activate a spring-driven device or to cock the spring in a spring-driven device. In addition, springs have a linear energy profile. The force provided by the gas-generating chemical reaction can be non-linear and non- logarithmic. The speed of the chemical reaction can be controlled by (i) adjusting the particle size of the dry reagent; (ii) changing the particle shape of the dry reagent; (iii) adjusting the packing of the dry reagent; (iv) using mixing assist devices; and/or (v) altering the shape of the reaction chamber where the reagents are mixed.

It should be noted that silicone oil is often added to the barrel of the syringe to reduce the release force (due to static friction) required to move the piston within the barrel. Protein drugs and other drugs can be negatively impacted by contact with silicone oil. Siliconization has also been associated with protein aggregation. The forces generated by the chemical reaction obviate the need for application of silicone oil to the barrel of the syringe. In other words, no silicone oil is present within the barrel of the syringe.

When a solvent is used to form a medium for a chemical reaction between chemical reagents, any suitable solvent may be selected. Exemplary solvents include aqueous solvents such as water or saline; alcohols such as ethanol or isopropanol; ketones such as methyl ethyl ketone or acetone; carboxylic acids such as acetic acid; or mixtures of these solvents. A surfactant may be added to the solvent to reduce the surface tension. This may aid in improving mixing and the subsequent chemical reaction.

The following examples are for purposes of further illustrating the present disclosure.

The examples are merely illustrative and are not intended to limit processes or devices made in accordance with the disclosure to the materials, conditions, or process parameters set forth therein.

EXAMPLES

A test rig 1000 for carrying out experiments is shown in FIG. 10. A standard prefilled syringe 1040 was filled with 1 ml fluid. Starting at the left of the figure, a prefilled syringe 1040 was fitted with a 19 mm long and 27 gauge thin walled needle 1006 and stopped with a standard stopper 1066. This syringe 1040 acted as the fluid chamber. Connected to the prefilled syringe was a reaction chamber syringe 1030. A piston rod 1064 and a push surface 1062 were used to apply force from the chemical reaction to the stopper 1066. A one-way pressure valve 1050 was used to allow injection of solvent from a second "injector" syringe 1020 that acted as the reagent chamber. The set-up was clamped into the test fixture 1010. A graduated pipette (not shown) was used to measure the volume delivered versus time.

EXAMPLE 1

Two fluids were tested, water (1 cP) and silicone oil (73cP). Water served as the low- viscosity fluid, silicone oil served as the high-viscosity fluid. One of these two fluids was added to the prefilled syringe 1040 depending on the experiment. To the reaction chamber syringe 1030 was added 400 mg NaHC0 ₃ and 300 mg citric acid, as dry powders. The injector syringe 1020 was filled with either 0.1 ml, 0.25 ml, or 0.5 ml water. The water was injected into the reaction syringe 1030 (the volume of the reaction syringe was adjusted based on the volume to be delivered by the injector syringe). The delivered volume versus time and total delivery time were measured. The pressure was calculated using the Hagen-Poiseuille equation and assumed there was 0.6 lb frictional force between the stopper 1066 and the prefilled syringe 1040.

Alternatively, the force on the prefilled syringe 1040 was determined by placing a load cell at the exit. The results are shown in Table 1 and were based on a minimum of at least three runs. The chemical reaction syringe provided delivery of 1 mL water in 5 seconds or less. The delivery time for the higher viscosity fluid depends on the volume of water injected.

Surprisingly, the delivery time is faster when the volume of water is greater. This is surprising because water, which doesn't participate in the reaction, serves to dilute the reagents. Reaction kinetics, and production of C0 ₂ , decrease as the concentration of reagents decreases. The results indicate the importance of the dissolution kinetics. The dissolution is faster for higher volume of water. Using 0.5 mL of water, high viscosity fluid can be delivered in 9 seconds. Thus, in some preferred embodiments, the invention provides for delivery of substantially all of a solution (of at least 0.5 ml, or 0.5 to 3.0 mil or 1 ml) having a viscosity of at least 20cP, preferably at least 40 cP and in some embodiments below about 70 cP within 20 seconds or within 15 seconds or within 10 seconds with a chemical engine comprising a starting volume (prior to expansion) of less than 1.0 ml, in some embodiments within the range of 0.3 to 1.0 ml; in some embodiments, the ratio of the volume of water in the chemical engine to volume of medicament is less than 2: 1, preferably less than 1: 1, less than 0.5: 1 and in some embodiments in the range of 1: 1 to 0.3: 1. Throughout this description, viscosity is measured (or defined) as the viscosity at 25 °C and under the delivery conditions.

Table 1

FIG. 22 is a graph showing the pressure versus time profile for delivery of silicone oil when 0.1 ml (triangle), 0.25 ml (square), and 0.5 ml (diamond) of water was injected into the reaction chamber. This graph shows that a nearly constant or diminishing pressure versus time profile could be obtained after a ramp-up period, although the impact of volume expansion dominated at longer times. These pressure versus time profiles were not exponential. A constant pressure versus time profile may allow for slower, even delivery of a high-viscosity drug, as opposed to a sudden exponential burst near the end of a delivery cycle. EXAMPLE 2

Sodium chloride (NaCl) was used to enhance the release of gaseous C0 ₂ from the reaction solution in the reaction chamber, accelerating the increase in pressure. In control experiments, citric acid and NaHC0 ₃ were placed in the reaction syringe. A solution of 1.15 M NaHC0 ₃ in water was injected into the reaction syringe from the injector syringe. The empty volume in the reaction syringe was kept constant through all experiments. In experiments demonstrating the concept, NaCl was added to the reaction syringe. The chemical reaction was used to deliver 1 ml of water or silicone oil from the prefilled syringe. The delivered volume versus time and total delivery time were measured. The pressure was calculated using the Hagen- Poiseuille equation and assumed there was a 0.6 lb frictional force between the prefilled plunger and the syringe. Note that bicarbonate was present in the water injected into the reaction syringe, so that gas could be generated even if solid bicarbonate was not present in the reaction syringe itself. The results are shown in Table 2.

Table 2.

Salt served to significantly enhance the delivery rate, particularly for systems that used smaller amounts of reagent. A high viscosity fluid could be delivered in 6 to 8 seconds using the chemical reaction. This is significantly faster than what can be achieved with standard auto- injectors that employ mechanical springs.

FIG. 23 shows the delivered volume versus time profile for experiment Nos. 5 and 6 of Table 2. A high viscosity fluid was delivered in 20 seconds using a system having a footprint of less than 1 cm3. The delivery rate (i.e. slope) was also relatively

constant. The small footprint enables a variety of useful devices. EXAMPLE 3

Several different reagents were tested to show the influence of powder morphology and structure on the pressure profile. Morphology, in this case, refers to the surface area, shape, and packing of molecules in the powder. The same bicarbonate (sodium bicarbonate) was tested. Three different morphologies were created - one as-received, one freeze-dried, which was produced by freeze drying a 1.15 M solution, and one where the bicarbonate was packed into a tablet "Layering" of the reagents in the reaction chamber was also examined; where layering refers to preferential placement of reagents within the reaction cylinder. In another example, Alka Selzter, which contains citric acid and sodium bicarbonate in a matrix, was used.

The following reagents were used: as-received baking soda (NaHC0 ₃ ), citric acid, freeze- dried baking soda, Alka-Seltzer, or as-received potassium bicarbonate (KHC0 ₃ ). The as-received baking soda was also tested as a powder, or in a tablet form. The tablet form had a decreased surface area.

The freeze-dried baking soda was formulated by preparing 125ml of saturated baking soda aqueous solution (1.1SM). The solution was poured into a 250ml crystallization dish and covered with a Kimwipe. The solution was placed in a freeze dryer and was ramped down to - 40°C and held for two hours. The temperature remained at -40°C, and a vacuum was applied at 150 millitorrs (mTorr) for 48 hours. Alka-Seltzer tablets (Effervescent Antacid & Pain Relief by Kroger) were broken up using a mortar and pestle into a coarse powder.

Baking soda tablets were prepared by pouring 400mg of as-received baking soda powder in a die to produce a tablet with a 1 cm diameter. The die was swirled around to move the powder to give an even depth across the 1 cm. The die was placed in a press and held at 13 tons pressure for 1 minute. Tablets weighing 40 mg and 100 mg were broken from the 400mg tablet. The Apparatus and Plan

The previously described test rig was used. The 3ml injection syringe was filled with

0.5ml of de-ionized water. The 10ml reaction syringe was connected to the injection syringe by luer locks and a valve, and then clamped down tightly in the apparatus. A load cell was attached to the plunger rod so the reaction syringe plunger presses on it during the test. This recorded the applied force from the reaction while displacing the fluid in the prefilled syringe.

The fluid from the prefilled syringe was displaced into a graduated syringe which was video recorded. This observed the change in volume of the fluid over time. The fluids were water (lcP) or silicone oil (73cP), which were displaced through a 27 gauge thin-walled prefilled syringe and had a volume of 1 milliliter (ml).

Two measurements were acquired while during each test: the force on the prefilled syringe using a load cell and the change in volume of the prefilled syringe by measuring the dispensed volume with time. The average volume vs. time curve was plotted to show how each reaction changed the volume in the prefilled syringe. The pressure vs. time curve using the Hagen-Poiseuille equation was provided by calculating the flow rate from the volume vs. time curve. To account for the friction in the prefilled syringe, 94,219 Pa was added (which is equivalent to 0.6 lb). This calculated the pressure inside the prefilled syringe (3 mm radius) so the hydraulic equation was used (PiAi=P ₂ A ₂ ) to calculate the pressure inside the reaction syringe (6.75 mm diameter). This was used to check the measurement made by the load cell.

Another pressure vs. time curve was produced by using the force in the load cell measurement and dividing by the area of the reaction plunger. We found this to provide morereproducible data than the calculation by Hagen- Poiseuille.

To determine how the pressure changed with volume, pressure vs. volume curves were produced. The pressures used were those calculated by the load cell measurements. The reaction volume was calculated using the change of volume in the prefilled syringe. The volume of the reaction syringe (VR) could be determined from the dispensed volume in the prefilled syringe (Vp) at time t.

Finally, the reaction rate while dispensing the fluid was found by using the ideal gas law where PR is the pressure calculated from the load cell, VR is the volume of the reaction syringe, R is the universal gas constant (8.314 Jmol^K ^"1 ), and T is the temperature, 298K.

The Tests

The baseline formulation was 400 mg of baking soda, 304 mg of citric acid, and 0.5 ml of de-ionized water as described in Example 1. This formulation produces 4.76x10-3 moles of C0 ₂ assuming 100% yield. The ingredients of all tests were formulated to produce the same 4.76x10- 3 moles of C0 ₂ assuming 100% yield. Four sets of tests were performed.

The first set used as-received baking soda (BSAR) and freeze-dried baking soda (BSFD). Their relative amounts were varied in increments of 25%. 304 mg citric acid was also included in each formulation. Table 3A provides the target masses of the baking soda for these tests. Table 3A.

The second set used as-received baking soda and Alka-Seltzer. The amount of as- received baking soda was varied in increments of 25%. The stoichiometric amount of citric acid was added. Alka-Seltzer is only approximately 90% baking soda/citric acid. Therefore, the total mass of Alka-Seltzer added was adjusted to obtain the required mass of baking soda/citric acid. Table 3B provides the target masses of each ingredient for these tests.

Table 3B.

The third set used as-received baking soda and as-received potassium bicarbonate. The mass of citric acid was maintained at 304 mg throughout the tests. Due to the heavy molar mass of potassium bicarbonate (100.1 g/mol as opposed to baking soda's 84.0 g/mol), more mass is required to generate the same moles of C0 ₂ . Table 3C provides the target masses (in mg) of each ingredient for these tests.

Table 3C. Test Baking Soda Potassium Bicarbonate

100% BS 400 0

50% BS 200 239

100% KHC03 0 477

The fourth set used the baking soda tablets. The stoichiometric amount of citric acid was used. No other reagents were added. Table 3D provides the target masses (in mg) of each ingredient for these tests.

Table 3D.

Results of the Tests

First Set: as-received baking soda (BSAR) and freeze-dried baking soda (BSFD).

The freeze-dried baking soda powder appeared coarse relative to the as- received baking soda powder. It was also less dense; 400 mg of the freeze-dried powder occupied 2 ml in the reaction syringe, whereas the as-received powder only occupied 0.5 ml. Due to the volume of material, the smaller volume of water (0.5 ml) was insufficient to fully contact all of the freeze- dried baking soda. Thus, in "100%", "75%", and "50" experiments, the bicarbonate was not fully wetted, dissolved or reacted. Therefore, a fifth formulation was run where the freeze-dried sample was inserted first. It was followed by the citric acid and then the as-received powder. It was labeled as "50% BSAR Second". This formulation permitted the injected water to come into contact first with the freeze-dried powder, then contact and dissolve the citric acid and the as- received powder. The time needed to displace 1 ml of the silicone oil is listed in Table 3E.

Table 3E.

Formulation Time (sec) 100% BSAR 10

75% BSAR 13

50% BSAR 11

50% BSAR Second 22

100% BSFD 14

The volume vs. time graph is seen in FIG. 24. It appeared that the 100% freeze-dried powder was initially faster than the 100% as-received powder but slowed over time. The as- received powder took 10 seconds to displace 1 ml, and the freeze-dried powder took 14 seconds. As expected, the trials with mixed amounts were found to have times between the two extremes.

The pressure vs. time graph is given in FIG. 25. The formulations with 100% BSAR showed a maximum pressures nearly 100,000 Pa higher than those with 100% BSFD. In comparison, using "75% BSAR" gave a faster pressure increase and slower decay. For ease of comparison, the pressures were normalized and plotted in FIG. 26 and FIG. 27 (two different time periods).

The 100% BSAR had an initial slow reaction rate compared to the 75% BSAR and 50% BSAR formulations. This suggests the freeze-dried baking soda (BSFD) dissolves and reacts faster, and this is seen in FIG. 22. However, FIG. 21 shows that as the freeze-dried baking soda content increases, a lower maximum reaction pressure is obtained. Due to the low density of the freeze dried powder, 200 mg of the freeze-dried baking soda occupies 1 ml of space, so the 0.5ml of de-ionized water cannot contact all of the material before the generated gas moves the plunger, leaving the dry powder behind stuck on the side of the chamber; not all of the freeze- dried baking soda was reacted and lessC02 was produced. It was estimated for the 100% BSFD trial that only a quarter of the reagent dissolved. This phenomenon may be reduced by modifying the chamber devices, for example where the gas-generating chemical reaction occurs in a rigid chamber, with the generated C02 diffusing through a filter to push the plunger.

In the 50% BSAR Second trial, when the freeze-dried baking soda was added first followed by the citric acid and as-received baking soda, much of the powder remained solid, resulting in a lower pressure. The low initial reaction was most likely caused by the 0.5 ml of water diffusing through the 1ml of freeze-dried baking soda powder before reaching and dissolving the citric acid. This test was the closest of the trials in this set to providing a constant pressure profile.

The maximum pressure obtained was at approximately 0.8ml C0 ₂ volume for the 50% BSAR and 75% BSAR formulations. These formulations also had the fastest rate in the pressure vs. time graphs (see FIG. 26). The remaining formulations had maximum pressures at approximately 1.2 ml C0 ₂ .

Interestingly, when looking at FIG. 23 and FIG. 24, the "50% BSAR Second" showed a distinct pressure vs. time profile (Pa s in FIG. 23), but had approximately the same pressure vs. volume profile as the 100% BSFD in FIG. 24. Referring back to Table 3E, it took approximately 8 seconds longer for the "50% BSAR Second" to displace the 1ml of silicone oil, so its pressure curve is "drawn out" relative to the 100% BSFD, and it had a different flow rate. The results indicate that is is possible to reduce the pressure drop that accompanies piston motion (and volume expansion of the reaction chamber) by including bicarbonates with two different dissolution rates, where the different dissolution rates are provided by their morphology and/or position in the reaction chamber.

Table 3F shows the reaction rates for production of C02 versus time fitted to y = ax +bx.

Table 3F.

The 100% BSAR, 75% BSAR, and 50% BSAR curves have approximately the same linear reaction rate. The "50% BSAR Second" forms a second order polynomial. The " 100% BSFD" appears to be parametric; it has the same linear rate as 100% BSAR and the other two, and then the slope suddenly decreases after 5 seconds and converges with "50% BSAR Second."

Second Set: as-received baking soda and Alka-Seltzer. The volume vs. time graph is seen in FIG. 28 for silicone oil, and in FIG. 29 for water as the injected fluids respectively. The time needed to displace 1 ml of each fluid is listed in Table 3G. The error in time measurement is estimated to be half a second.

Table 3G.

Time (sec)

The times for displacement of water are difficult to compare because they are all within one second of each other. The volume profiles for 100% BSCA, 25% BSCA, and 100% Alka- Seltzer had the fastest times to displace 1ml of silicone oil. The 100% BSCA appeared to start slowly and then speed up. The 50% BSCA and 75% BSCA were found to have the slowest times. They appeared to slow down as the displacement proceeded.

The pressure vs. time graph is given in FIG. 30 for silicone oil, and in FIG. 31 for water as the injected fluids respectively. The 100% BSCA had the slowest initial pressure rise. This was expected, since Alka-Seltzer is formulated to allow fast diffusion of water into the tablet. The 75% BSCA and 50% BSCA had the second and third greatest maximum pressure, respectively, for silicone oil. However, these two formulations took the longest to displace 1 ml of silicone oil. Their pressures also had the slowest decay. This is most likely due to increased friction in the syringe.

The curves in FIG. 31 for water are within a reasonable error of each other. However, they were greater than the estimated pressures by Hagen-Poiseuille, which calculated the maximum pressure to be 51,000 Pa by the 100% Alka-Seltzer formulation. High friction was not observed during testing.

Normalized pressure vs. time graphs are provided for silicone oil in FIG. 32. The pressure decay rate for silicone oil is provided in Table 3H.

Table 3H. Formulation Decay Rate (Pa/s)

100% BSCA 6,854

75% BSCA 4,373

50% BSCA 3,963

25% BSCA 9,380

100% Alka-Seltzer 10,695

For silicone oil, the 100% BSCA and the 75% BSCA had the same normalized pressure increase, but different decays. As explained above, the 75% BSCA may have undergone more friction causing the change in volume to slow and hold pressure longer. The same was true for the 50% BSCA, which had the same decay as 75% BSCA. Surprisingly, the pressure increase for 50% BSCA fit just between 100% BSCA and 100% Alka-Seltzer. This may indicate that friction does not affect the pressure increase. The 100% Alka-Seltzer and 25% BSCA had the same pressure profiles with the fastest pressure increase and fastest decay. The 100% BSCA also appeared to have the same decay as these two formulations.

For water, it was found that higher ratios of Alka-Seltzer to BSCA resulted in relatively less pressure decay. The 100% Alka Seltzer had a fast pressure increase but quickly decayed along with 100% BSCA" and 75% BSCA. However, 25% BSCA and 50% BSCA had fast pressure increase and less pressure decay than the other formulations.

For silicone oil, the 100% Alka-Seltzer, 50% BSCA, and 75% BSCA all peaked at approximately 1.2 ml of C0 ₂ volume. The 25% BSCA peaked at approximately 0.8 ml. The 100% BSCA did not reach maximum pressure until approximately 1.6 ml. This was slightly different than the " 100% BSAR" in the first set of tests, which used the exact same formulation but reached its maximum pressure at a C0 ₂ volume of 1.2 ml.

Table 31 shows the reaction rates for C02 production during injection of silicone oil fitted to y=ax ² +bx. Table 31.

All formulations except 100% Alka-Seltzer formed linear reaction rates for silicone oil. The high friction in the prefilled syringe used to test 75% BSCA and 50% BSCA caused a high pressure, which may have reduced the reaction rate to 3xl0 ^"5 mol/s. The 100% BSCA and 25% BSCA had the same reaction rate at 4xl0 ^"5 mol/sec. 100% Alka-Seltzer resulted in a second order polynomial. It initially had the same reaction rate as the other formulations, but the slope decreased in the last few seconds. When the reaction was finished, the solution was much thicker than the other formulations.

The 100% BSCA was slightly slower than the previous experiment with freeze-dried baking soda, 100% BSAR (see Table 3F), by lxlO ^"5 mol/sec. This may have caused the slower time to displace the silicone and possibly the maximum pressure at a greater C0 ₂ volume at 1.6 ml.

Third Set: as-received baking soda and as-received potassium bicarbonate.

The volume vs. time graph is seen in FIG. 34, for silicone oil. The time needed to displace 1 ml of each fluid is listed in Table 3 J.

Table 3J.

The 100% KHCO ₃ was the fastest to displace the 1 ml of silicone at 6.33 seconds. The 100% BS and 50% BS displaced the same volume at a time of 8.00 seconds. The pressure vs. time graph is given in FIG. 35. The pressure decay rate for silicone oil is provided in Table 3K.

Table 3K

The 100% BS formulation only reached a maximum pressure of approximately 250,000 Pa, while the other two formulations had a maximum pressure of approximately 300,000 Pa. The 100% KHCO ₃ and 50% BS formulations (each containing potassium bicarbonate) continued increasing in pressure for a few seconds after the 100% BS reached its maximum. The 50% BS formulation initially had less pressure as expected but was able to maintain a higher pressure after 6 seconds compared to the 100% KHCO ₃ . The results showed that using a mixture of sodium and potassium bicarbonate can produce higher pressures and slow decays.

The 100% BS had a peak pressure somewhere between 0.6 and 1.8 ml of C0 ₂ . The curves for 50% BS and 100% KHCO ₃ were different from the other pressure vs. volume graphs seen previously. Instead of peaking at approximately 1.2 ml of C0 ₂ volume and decaying, they continued increasing in pressure at greater C0 ₂ volumes. The 50% BS and 100% KHC03 formulations peaked at approximately 2.0 and 3.2 ml of C0 ₂ volume, respectively.

Table 3L shows the reaction rates fitted to y=bx.

Table 3L.

They appeared to be linear reaction rates with 100% BS at 5x10 ^~5 mol/sec (the same rate from the experiments above). 100% potassium bicarbonate has twice the rate as baking soda..

Fourth Set: baking soda tablets. The volume vs. time graph is seen in FIG. 43 for silicone oil. The time needed to displace 1 ml of each fluid is listed in Table 3M.

Table 3M.

Time (sec)

For both silicone oil and water, using 400 mg and 100 mg baking soda tablets and the stoichiometric citric acid resulted in nearly straight lines. Packing the baking soda into dense tablets significantly decreased reaction rates, and thus increased injection times, relative to other baking soda experiments. Table 3N shows the reaction rates fitted to y=ax +bx.

Table 3N.

For silicone oil, the 400 mg BS tablet showed the linear reaction rate as 4 xlO ^"6 mol/sec. The 100 mg baking soda tablet was linear for almost 87 seconds until it suddenly stopped producing gaseous C0 ₂ . The reaction rate for the 40 mg tablet was a second order polynomial and very slow. It reached a total of 2xl0 ^"5 moles C0 ₂ and stayed steady with some fluctuation possibly caused by the C0 ₂ moving in and out of solution. Due to the small reaction rate in water, only the 400 mg tablet was used.

The results of Example 3 showed the ability to create different pressure versus time profiles when the dissolution kinetics are modified.

EXAMPLE 4

The test rig was used to test silicone oil and a 27 gauge thin- wall needle. The

stoichiometric reaction and results are shown in Table 4 below. Table 4.

EXAMPLE 5

The prototype test device illustrated in Fig. 19 was tested using silicone oil. A pre-filled syringe acted as the fluid chamber from which fluid was ejected. Next, a connector was used to join the pre-filled syringe with the reaction chamber. The reaction chamber included a mixing. A piece of filter paper was placed inside the reaction chamber to cover the orifice to the arm. A spring was then placed inside the mixing chamber. Next, a plunger was used to separate the dry reagent in the reaction chamber from the wet liquid. The next piece was the push button, which included an interior volume for the liquid reagent. The push button included a hole (not visible) that was used to fill the volume with liquid reagent. A screw was used to fill the hole in the push button. A cap was fitted over the push button to provide structural support, and also surrounds a portion of the reaction chamber. Finally, a thumb press was placed on top of the cap for ease of pressing. Both the reagent chamber and the reaction chamber were completely filled with liquid solution and dry powder, respectively.

The syringe was tested in both the vertical position (reagent chamber above reaction chamber) and the horizontal position (the two chambers side-by-side). The reagents and results are shown in Table 5 below. Table 5.

Assuming adequate mixing, the potassium bicarbonate is the limiting reactant, with citric acid at an excess of 89 mg. This assumption was found to be incorrect because liquid was found in the top chamber and powder was found in the bottom chamber when disassembled. When the syringe was laid in a horizontal position, and the chambers were completely filled, the silicone oil was displaced in 17 seconds. This illustrates that the device can work in any orientation. This is helpful for permitting patients to inject into their abdomen, thigh, or arm, which are the most common locations for self-injection. Example 6 Mixed Bicarbonates

In the test apparatus of Example 1 was modified to contain a mixture of 50:50 molar mixture of sodium and potassium bicarbonate. Delivery of the silicone oil was accommodated in a faster time (just under 8 sec) and the pressure versus time was flat. The flow increased and then reached a plateau in just under 2 seconds. The use of mixed bicarbonates allows for a system that has different reaction kinetics to control the pressure profile.

Example 7: Use of a Nucleating Agent to Enhance Release of CQ2

Sodium Chloride (NaCl) was used to enhance the release of gaseous C02 from the reaction solution into the reaction chamber, accelerating the increase in pressure. In control experiments, citric acid and NaHC03 was placed in the reaction syringe. A solution of 1.15 M NaHC03 in water was injected into the reaction syringe. The empty volume in the reaction syringe was minimized through all experiments and was dictated by the density of the powder. In experiments demonstrating the concept of the invention, NaCl was added to the reaction syringe. The chemical reaction was used to deliver 1 mL of water or silicone oil. The delivered volume versus time and total delivery time were measured. The pressure was calculated using Hagen-Poisuielle equation, considering the area of the plungers, and assumes there is a 0.61b frictional force between the prefilled syringe plunger and the syringe.

Salt serves to significantly enhance the delivery rate, particularly for systems that use smaller amounts of reagent. A high viscosity fluid can be delivered, for example, in 6 to 8 seconds using the chemical reaction. This is significantly faster than what can be achieved with standard auto-injectors that employ mechanical springs. The delivered volume versus time is shown for the smallest chemical-reaction. A high viscosity fluid was delivered in 20 seconds using a system having a footprint of less than 0.5 cm . The small footprint enables a variety of useful devices.

Example 8 -- Use of Reagents with Two Dissolution Rates to Modify Pressure versus Time Profile

Reagents with two different dissolution rates were created by combining NaHC03 with two different morphologies (preferably, different surfaces areas). For example mixtures can be prepared using bicarbonates obtained from different sources, or treating a portion of the bicarbonate prior to combining with an untreated portion. For example, a portion can be freeze- dried to increase surface area. High surface area NaHC0 ₃ was produced by freeze drying a 1.15 M solution. Reagents with two different dissolution rates were also created by combining as- received sodium citrate / NaHC0 ₃ and formulated sodium citrate / NaHC03 (Alka Seltzer). The results show the ability to create different pressure versus time profiles when the dissolution kinetics are modified.

Example 9 - Minimization of Pressure Decrease with Expansion of Piston

Chemical engines were created to deliver fluids with viscosity from 1 to 75 cP and volumes of 1 to 3 mL in less than 12 seconds through a 27 gauge needle. In the experiments of this example, the dry chemical were premixed in a jar and then added to the reaction syringe (B). The reaction syringe consisted of either a 10 mL or 20 mL syringe. The plunger was fully depressed on the powder so that no additional empty volume was present. A solution was added to the reaction syringe; as C0 ₂ was generated, the plunger rod pressed against the plunger of the PFS and delivered fluid. Six engine formulations were considered, as shown in the Table.

Formulation Solution Dry Chemicals in Amount of Time to Time to

Added to Reaction Chamber Fluid Deliver 20 Deliver 50 cP Reaction Delivered cP Fluid (s) Fluid (s) Chamber

1 0.5 mL 107 mg citric acid 1 mL 4 7

saturated 200 mg NaCI

KHC03

2 0.75 mL 160 mg citric acid 1 mL 2.5 5

saturated 100 mg NaCI

KHC03

3 1.0 mL 800 mg KHC03 (s) 3 mL n/a 9.5

saturated 610 mg citric acid

KHC03

4 1.0 mL water 1130 mg KHC03 3 mL n/a 11 (s)

610 mg citric acid

5 2.5 mL water 2530 mg KHC03 3 mL n/a 6.5

(s)

1500 mg citric acid

The force versus time profiles are shown for the different chemical engines in Figs. 40-42. Formulations 1 and 2 were used to deliver 1 mL of fluid through a 27 gauge thin wall needle; i.e. a standard PFS. Fluids with viscosity of 25 and 50 cP were examined. Faster delivery is achieved when the amount of reagent is increased. The use of potassium bicarbonate allows for substantially less reagent to be employed than when sodium bicarbonate is used.

Formulations 3, 4, and 5 were used to deliver 3 mL of 50 cP fluid through a 27 gauge thin wall needle. The target flow rates were higher than those targeted for Formulations 1 and 2. In this case, simply scaling up the reaction to larger amounts (Formulation 3) results in substantial initial overshoot in the force, due to rapid reaction. The overshoot was reduced by employing 100% solid reagents (mixture of citric acid and potassium bicarbonate in the reaction chamber) and water during the injection. This method provided a steady-state of C0 ₂ as the water dissolved the potassium bicarbonate and made bicarbonate ions available. Formulation 5 exhibited a flat delivery profile and delivered 3 mL of 50 cP fluid in 6.5 s.

Example 10 Addition of Convection Agents

The addition of a small amount (for example, <10 mg for a 1 mL engine) of slowly dissolving or insoluble particles was found to be effective in substantially increasing the rate of gaseous C0 ₂ generation and the maximum gaseous C0 ₂ generated, substantially increasing the power density of the engine. Surprisingly, we found the surface energy and surface topology of the particles to have only a minor effect, as a variety of slowly dissolving or insoluble particles work, including diatomaceous earth, Expancel™ (polyacrylonitrile hollow microspheres), calcium oxalate, and crystalline oxalic acid. In this case, slowly dissolving means that the particle is slow relative the reagents in the engine. The presence of these particles can be determined experimentally, by dissolving the solid reagents and looking for the presence of particles or determining the identity of the materials present and comparing their solubility products. The density can be either lower than or higher than water. We believe that these agents act alone, or in concert with gaseous C0 ₂ leaving the fluid, to set-up mixing fields, similar to those that might be found in fluidized beds. The mixing fields increase the collisions between reagents and between the convection agents and reagents. These increased collisions serve to release kinetically trapped C0 ₂ that may exist at surfaces and crevices, such as on the container or on bicarbonate surfaces.

Our results indicate that these reagents are not serving primarily as nucleating agents, though that could be a minor factor. The reagents are effective in conditions where the reaction chamber is fixed to constant volume, or allowed to expand. Under constant volume

circumstances, the pressure does not decrease, and the solution is never supersaturated, as, for example, might be seen in a pressurized carbonated beverage that is opened to atmosphere.

Furthermore, the addition of nucleating surfaces, for example porous aluminum, is ineffective. The reagents must be present as particles in the engine.

Experiments were carried out in one of two set-ups.

Constant Volume: The constant volume setup was used to compare different chemical reaction times. Reagents were placed in a 2 ml reaction chamber; liquid reactants were added to the chamber by using a syringe and injecting into the reaction chamber at the desired time of injection and the valve was closed. Pressure was measured by a pressure transducer and temperature by thermocouple. As the reaction occurred, pressure from the chamber was channeled into the air cylinder through a small tube. The air cylinder was used as an equivalent to a piston/plunger in a real injection system. For the constant volume setup, the air cylinder was initially moved 1.5 inches to the end of the injection position for a 2 ml syringe and kept at that location throughout the test. This gave a total volume of the reaction chamber and the air cylinder was 9.9 ml. A load cell was used to measure force as a back up to pressure, but could have been calculated from the pressure and area of the piston in the cylinder. The advantage of using a constant volume setup to look at initial chemical selection was that it allowed a good comparison of pressure versus time profiles without having to take into account the differences in volume that a real system would have.

Test Conditions

The following chemicals were tested as received: sodium bicarbonate, potassium bicarbonate, citric acid, tartaric acid, oxalic acid, calcium oxalate, diatomaceous earth, and Expancel™ . Anodized aluminum oxide was prepared by anodization of aluminum in oxalic acid to create porous surface structure and high surface energy.

Reactants were loaded into either the reaction vessel or as a solution to the syringe. The reaction vessel generally contained the acid, as a solid or solution, with or without any additives. The syringe contained the bicarbonate solution. The reactants (bicarbonate and acid) were measured out as powders on an analytical balance in stoichiometric ratios. The masses of reactants are displayed in the following Table.

Table. Masses of reactants used for testing. 1 niL of water was used to dissolve the bicarbonate and supply them a medium where the reaction could evolve within the reaction chamber of the ChemEngine.

There were 4 different types of convection agents added to the chemical formulation, a nucleation surface was added to the reaction chamber, and external vibration was also added to the reaction chamber in separate experiments.

1. Water insoluble with high density - Diatomaceous Earth (DE)

2. Water insoluble with low density - Polyacrylonitrile (PAN) hollow micro-spheres

3. Water slightly soluble - Calcium Oxalate

4. Water highly soluble - Sodium Chloride and Oxalic Acid

5. Nucleation surface - Anodized Aluminum Oxide

6. Mechanical vibration

All of the convection agents (1-4) were added as dry particles to at loading between 5 mg and 50 mg to the chemical formulation. The nucleation surface (anodized aluminum oxide) was added into the reaction chamber.

As shown in the Figure 43, all of the convection agents increased the rate of pressure accumulation over the baseline formulation. In this case, the baseline formulation is the same amount of potassium bicarbonate, citric acid, and water, but no other chemicals present.

Mechanical vibration also increased the rate of pressure accumulation over the baseline formulation. A nucleation surface had very little effect on the rate of pressure accumulation. The data in Fig. 43 are provided to show that the convection agents functioned to increase the collision rate between reactants, and between reactant and products, to release C0 ₂ into the gas phase at an increased rate. The data show that addition of the convection agents operated by a differentiated mechanism, and increased the reaction chamber pressure more quickly from that of a nucleation agent. The data in Fig. 44 show that mechanical vibration (70 Hz from a Vibra- Flight™ Controller) and convection agents have a similar effect on the rate at which the system displaces a plunger at constant force.

Fig. 45 shows the surprising result that relatively smaller quantities of convection agents resulted in faster C0 ₂ generation. Experiments showed that the presence of about 10 mg of diatomaceous earth resulted in significantly faster C02 generation than either 5 mg or 50 mg per ml. Thus, some preferable compositions comprise between 7 and 15 mg of a convection agent or agents.

Example 11 : Power Density

Power density of a variety of chemical engines were measured either at a constant force or a constant volume.

Constant Force Setup:

Constant Force: A similar setup was used for constant force as constant volume (described above), but the stage that the load cell was attached to was able to move. The air cylinder was initially closed so that the initial volume of the reactant chamber and connectors was 2.3 ml. The air cylinder was allowed to move 1.4 inches (3.56). This distance was chosen because it was the amount that a piston would have to travel to empty a standard 2 ml syringe. The stage was initially pressurized to 18 lbs, which corresponded to injecting 2 ml of 50 cp fluid using a standard 2 ml syringe with a 27 gauge, TW needle in 8 seconds. Additional measurements were obtained at 9 lbs of backpressure. This corresponds to injecting 1 ml of 50 cp fluid using a standard 1 ml syringe with a 27 gauge, TW needle at 8 seconds. The powder reactants were placed into the reactant chamber and the liquid reactants were in the syringe. The liquid reactants were injected into the chamber and the valve was closed. Pressure, force, and temperature were measured until the air cylinder reached its travel distance, which was determined using an LVDT (linear variable differential transformer) that was attached to the stage. The apparatus is shown in Fig. 46. The syringe labeled water in Fig. 46 could alternatively be an aqueous solution comprising either acid or bicarbonate.

This test apparatus is applicable to test almost any chemical engine. Chemical engines that are integrated devices can be tested by placing the entire device in the test apparatus. When testing an integral system including a fluid compartment, the average force can be measured directly or calculated using the Hagen-Poiseuille equation. Chemical engines that are detachable from a fluid compartment are first detached prior to testing.

In the Table below, the water was added to a mixed powder of citric acid and bicarbonate in a 1:3 molar ratio.

Power Density ratio:

Power Density was calculated at different backpressures and initial volumes. Time is measured starting at the initiation of the reaction which is the time when acid and carbonate are combined with a solvent.

For our case, Power Density=Average force*distance to end of travel/(time to deliver* Volume of initial reactants)

The volume used was the volume of the reactants all the reactants dissolved and after C02 has escaped. Open space within the reaction chamber that is not occupied by reactants or solvent is not taken into account for our calculations.

1 mL water, 403

mg citric acid 529 m sodium bicarbonate 1.4 42.5 5.52 195,472 Control

315 mg potassium bicarbonate,

0.5 mL water 202 mg citric acid 0.7 45.8 16.8 138,709 6.0

0.5 mL water, 315 mg potassium bicarbonate,

202 mg citric 10 mg diatomaceous earth,

19.8 acid mixing 0.7 45.7 5.1 455,594

0.5 mL water,

202 mg citric

Control acid 264 mg sodium bicarbonate 0.7 44.5 60** 23,000

Did not reach full displacement, so a displacement of 3.05 cm at 60 s was used

**Did not reach full displacement, so a displacement of 2.17 cm at 60 s was used

Example (Row 1):

Power Density = Average Force * Displacement / Time / Volume of Reactants

111,778 W/m ^A 3 = 99.3 N * 3.56 cm*(0.01m/cm) / 22.59 s / 1.4 mL*(0.000001m ^A 3/mL)

(The numbers do not match the table exactly because they were rounded for this example.)

In each of the above experiments, the molar ratio of bicarbonate to citric acid is 3:1 (in general, for all of the systems described in this application, preferred formulations have a molar ratio of bicarbonate to citric acid in the range of 2 to 4, more preferably 2.5 to 3.5. The present invention can be characterized by power density at room temperature, measured and calculated as described above and at a nominal back pressure of 9 lbs (40 N). At these conditions, the invention preferably has a power density of at least 50,000 W/m ³ , more preferably at least 100,000 W/m ³ , more preferably at least 250,000 W/m , more preferably at least 400,000 W/m ³ , and in some embodiments an upper limit of 1,000,000 W/m ³ , or about 900,000 W/m ³ . Alternatively, the invention can be defined in terms of a power density by comparison with a control formulation that is subjected to the same conditions. The control formulation contains 1 mL water, 403 mg citric acid, and 529 mg sodium bicarbonate. This control formulation is appropriate for reaction chamber volumes of about 2 mL; the power density of controls in chemical engines which are larger or smaller than 2 ml should be tested with a control formulation that is adjusted in volume while maintaining this proportion of water, sodium bicarbonate and citric acid. Again, as measured at a nominal back pressure of 9 lbs (40 N) and at constant force, the inventive chemical engine preferably has a power density ratio of at least 1.4, more preferably at least 3 when compared to the control; and in some embodiments a maximum power density ratio of 10 or a maxium of about 5, or a maximum of about 4.4. In preferred embodiments, displacement begins with 2 sec, more preferably within 1 sec of the moment when acid, carbonate and solvent (water) are combined. It should be noted that the term "control" does not imply a conventional formulation since conventional formulations for chemical engines were much more dilute. The control is typically tested to full displacement; however, in cases where full displacement is not achieved within 30 seconds, the control defined as the displacement within the first 30 seconds.