The present invention relates to a bone cement composite comprising at least one cement base material and biodegradable metallic fibers made of a metal alloy .

Amorphous metals in general , but also biodegradable amorphous metals , are well known in the art . Furthermore , fiber- reinforced bone cements are known as well .

US2014 / 0007986A1 describes a composite including a matrix material having an intrinsic strain-to- f allure rating in tension and a reinforcing material embedded in the bulk material . However, said composite requires pre-tensioning of the fibers and does not mention biodegradability or a cementitious matrix material .

CN106273680A discloses a non-biodegradable cobalt-based amorphous alloy fiber composite for sensor applications . Furthermore , US5543187A describes a multilaminate composite material made of an amorphous metallic material and a ceramic material .

CN111773432 discloses the preparation of a medical bone filling material comprising a magnesium-based amorphous- calcium phosphate/calcium silicate composite filler with high biological activity .

US2013/ 150978 discloses bone implant which comprises a magnesium-containing metallic material having a reduced corrosion rate and inorganic bone cement , and to methods and a kit for producing the bone implant . Composites of biodegradable bone cement with biodegradable fibers of various biodegradable materials are known. I. Lodoso-Torrecilla, J.J.J.P. van den Beucken, J. A. Jansen (Calcium phosphate cements: Optimization toward biodegradability. Acta Biomater. 2021;119:1-12) disclose calcium phosphate cements which are reinforced with fibers made of polylactic acid (PLA) , poly ( lactic-co-glycolic acid) (PLGA) , poly(vinyl alcohol) (PVA) , gelatin, chitosan, glass or carbon .

R. Kruger, J.M. Seitz, A. Ewald, F.W. Bach, J. Groll (Strong and tough magnesium wire reinforced phosphate cement composites for load-bearing bone replacement. J Meeh Behav Biomed Mater. 2013;20 (C) :36-44) disclose a composite of a biodegradable magnesium calcium phosphate cement with reinforcement of uniaxially orientated fibers of a polycrystalline magnesium alloy. The mechanical properties of said composite are not sufficient. It was reported that failure of the composite occurs in vicinity of the wirematrix interface. Due to the low fiber-cement interface strength, cracks driven by shear stresses propagate parallel to the fibers when loading the construct in bending mode. Furthermore, due to the reported long and rather thick fibers, combined with their uniaxial orientation, the reported composite does not allow an intra-surgical application. From the reported data it is safe to assume that also a composite of randomly orientated fibers of the same kind would not provide a significant mechanical improvement to plain bone cement without fibers. This assessment is based on the reported weak fiber-cement interface strength and the limited mechanical strength of the used fiber material. Currently available bone cements show excellent biocompatibility and degradability . However, their tensile , shear and bending strength is rather low and as maj or issue , they are exceptionally brittle , which diminishes their applicability in load-bearing applications .

The obj ect of the present invention was therefore , to provide a bone cement composite with enhanced strength and ductility while at the same time being biocompatible .

The problem is solved by the subj ect-matter of independent claim 1 . Further preferred embodiments are subj ect of dependent claims 2 to 15 .

The biocompatible bone cement composite according to the present invention comprises at least one cement base material and metallic fibers made of a metal alloy, said fibers comprising at least three chemical elements in the alloy, wherein at least one of these elements is selected from the group consisting of magnesium and calcium, wherein the fibers are amorphous in a volume fraction of more than 75% , with the remaining part of the fibers being of crystalline atomic structure . The fibers comprised in the bone cement composite are at least partly, preferably completely, coated with a biodegradable degradation-protective layer . Such bone cements are particularly interesting for load-bearing application, since they have outstanding mechanical properties . Said protective layer prevents a chemical reaction between the fibers and the setting solution during setting of the cement .

The biocompatible fiber-reinforced bone cement composite comprises fibers of a metallic alloy which is amorphous in a volume fraction of more than 75% . This highly amorphous , i . e . , disordered, atomic-scale structure of the alloy results in a two to fourfold increased mechanical strength of the here employed fiber material compared to respective crystalline counterparts . The fibers can be randomly oriented, partly oriented, or fully uniaxially oriented within the bone cement matrix . For example , it is possible to orient fibers within the bone cement matrix according to expected stress traj ectories of the later loading .

The fibers are embedded in a cement base material resulting in a fiber-reinforced bone cement composite having signi ficantly enhanced strength and/or ductility, by not compromising biocompatibility and degradability the same time . This allows for an application in load-bearing applications . Furthermore , the bone cement composite according to the present invention can be osteoconductive and osteoinductive by the degradation of the fibers . In addition, due to the small diameter of the fibers the bone cement according to the present invention has an excellent in ectability, which is a clinically important issue .

The fibers contained in the bone cement composite comprise at least three chemical elements as alloying elements , wherein at least one of these elements is selected from the group consisting of magnesium and calcium .

Preferably, the bone cement composite according to the present invention comprises fibers which are amorphous in a volume fraction of more than 90% , or more preferable in a volume fraction of 95% , or particularly preferable in a volume fraction of more than 98 % , with the remaining part of the fibers being of crystalline atomic structure . A high amorphous volume fraction has a direct impact on the strength and ductility of the cement of the invention . The amorphous structure of the fibers contained in the bone cement composite according to the present invention results in multiple times higher fiber-strength compared to crystalline alloys or polymers , which allows for large reduction of fiber length and diameter at the same strengthening ef fect . Shorter, and thinner fibers add to malleability and inj ectability of the bone cement .

Preferably, said protective layer comprises salts , in particular phosphate salts or fluoride salts , oxides , biodegradable polymers , or small biocompatible organic molecules .

Coatings with phosphate salts can be obtained for example by dipping the fibers into a (NH ₄ ) ₂ HPO4 solution or by another form of phosphate conversion coating and subsequent drying . Said coatings have a dense structure and excellent adhesion to the cement base material . Furthermore , they show good biocompatibility .

Coatings with fluoride salts can be obtained for example by hydrofluoric acid immersion treatment . Said coating signi ficantly improves the corrosion resistance of the fibers during setting of the cement while also meeting the requirements of sel f-degradability and biocompatibility .

Biodegradable polymers are preferably selected from the group consisting of polycaprolactone ( PCL ) , poly ( L-lactic acid) ( PLLA) , poly ( DL-lactide-co-glycolide ) ( PLGA) , poly ( ether imide ) ( PEI ) , polyethyleneglycol ( PEG) , polyhydroxybutyrate

( PHB ) , alginate , and chitosan or mixtures thereof .

Small biocompatible organic molecules are preferably selected from the group consisting of glucose , mannose , fructose , sucrose , lactose , maltose and trehalose .

Such coatings can be obtained for example by deposition coating, plasma electrolytic oxidation coating, plasma electrolytic anodi zation coating, anodi zation coating, a fluoric conversion coating, Mg ( OH) ₂ coating, calcium phosphate conversion coating, phosphate conversion coating, hydroxy-apatite coating, organic coating, biodegradable polymer coating, and sol-gel coating .

Preferably, the bone cement composite according to the present invention comprises fibers that comprise at least 45 atomic percent ( at% ) calcium, thus are made of a calcium-rich material . Most preferably, the fibers contained in the bone cement composite according to the present invention comprises between 45 at% to 80 at% of calcium, between 5 at% to 30 at% of magnesium and between 2 at% and 35 at% of zinc . Magnesium, zinc and calcium are essential nutrients in the human body . Therefore , such fibers have a high biocompatibility . Additionally, in contact with water or body fluids the alloy of such fibers starts to slowly degrade with the degradation products being resorbed by the human body . Especially preferred are fibers consisting of alloys selected from the group consisting of Ca57.5Mgi ₅ Zn ₂ 7.5, CassMg^ . sZn^ . s , Ca _{22 . 2} Mg ₂ oZn ₂ 7 .5 , Ca5 ₂ .5Mg ₂ 7.5Zn3o, Ca5 ₂ .5Mg ₂₂ .5Zn ₂ 5 and Ca5oMg ₂ oZn ₂ o , with the subscripted numbers giving the relative amounts of the respective element in atomic percent . Preferably, the fibers contained in the bone cement composite according to the present invention comprises at least 55 at% magnesium, i . e . , are made of a magnesium-rich material . Despite of the high amount of magnesium, the bone cement composite according to the present invention does not show any crack propagation parallel to the fibers or a delamination process . Instead, the fibers break when their mechanical limit is reached in-plane with the cement matrix, which indicates an exceptionally strong fiber-cement interface .

Especially good results are obtained with a bone cement composite , wherein the fibers comprise between 55 at% to 80 at% of magnesium, between 15 at% to 40 at% of zinc and between 1 at% and 10 at% of calcium, preferably between 60 at% to 75 at% of magnesium, between 20 at% to 35 at% of zinc and between 2 at% and 7 at% of calcium . The fibers may additionally contain up to 5 at% of one or more elements selected from the group consisting of copper, nickel , zinc, palladium, and rare-earth elements .

In one embodiment of the present invention the fibers consist of 55 at% to 80 at% of magnesium, between 15 at% to 40 at% of zinc and between 1 at% and 10 at% of calcium, preferably between 60 at% to 75 at% of magnesium, between 20 at% to 35 at% of zinc and between 2 at% and 7 at% of calcium . Such fibers have an excellent corrosion resistance resulting in less hydrogen bubbles during setting of the cement or later during degradation in vi vo . Especially preferred are fibers consisting of alloys selected from the group consisting of MggoZnasCas, Mg ₆₆ Zn ₃ oCa4 , Mg ₆₉ Zn ₂₆ Ca ₅ , Mg ₆₇ Zn ₂₈ Ca ₅ , Mg ₆₉ Zn ₂ 7Ca4 , Mg ₆₉ Zn ₂₈ Ca3 and Mg ₇₂ Zn ₂₄ Ca4 , with the subscripted numbers giving the relative amounts of the respective element in atomic percent .

In one aspect of the present invention, the fibers comprise least 55 at% magnesium or at least 45 at% calcium, and one or more elements selected from the group consisting of copper, nickel , zinc, palladium, and rare-earth elements , preferably magnesium and calcium and one or more elements selected from the group consisting of copper, nickel , zinc and palladium . Preferred rare-earth elements are yttrium and ytterbium .

The bone cement composite according to the present invention comprises fibers having a diameter which is smaller than 100 pm, preferably smaller than 75 pm and most preferably smaller than 50 pm . A smaller fiber diameter simultaneously allows for a decrease of the fiber length without compromising the composite material ' s overall strength . And shorter fibers facilitate intra-surgical application by enhancing malleability and inj ectability of the composite during its setting and intra-surgical application phase . Therefore , preferably, fibers are as short as possible without compromising mechanical parameters of the resulting composite material . The resulting excellent malleability of the bone cement composite according to the present invention allows for the cement to better adapt to a defect ' s site and shape , and short and thin fibers can allow the fiber-cement composite to be applied by inj ection . In addition, such fibers can facilitate shear ( flow) alignment during inj ection, thus result in tailorable mechanical properties of the resulting fiber-cement composite for patient-speci fic application . Preferably, the average fiber length is less than 5 . 75 millimeter, most preferably less than 3 . 8 millimeter and ideally less than 1 . 9 millimeter . Such an average fiber length result in a bone cement according to the present invention with a good malleability and inj ectability .

In one aspect of the present invention the bone cement composite comprises a fiber volume fraction between 10 % and 18 % , preferably between 12 % and 16 % . In this range an optimal combination of mechanical strength and ductility can be obtained .

The base material of the bone cement composite ( forming the composite ' s matrix, a term well known in the art ) can be biodegradable or non-biodegradable . Biodegradable base material means that this material will be broken down in vivo and can be replaced over time by new growth of bone tissue , whereas non-biodegradable base material remains in the body unless surgically removed .

A biodegradable base material is preferably selected from the group consisting of magnesium calcium phosphate cements , calcium phosphate cements , brushite cements , hydroxyapatite cements , trimagnesium phosphate ( Farringtonite ) cements , amorphous magnesium phosphate cements , calcium sulphate cements and magnesium- ammonia cements . These cements ' base materials have a good compatibility with the amorphous fibers according to the present invention .

A non-biodegradable base material is preferably selected from the group consisting of poly (methyl methacrylate ) , calcium aluminates and calcium silicates . Such a bone cement composite can be used for example for hip fracture prevention . Even in a non-biodegradable base material biodegradable magnesium-based or calcium-based additions such as fibers are beneficial by initially strengthening the resulting composite material and subsequently forming voids during in vi vo degradation that foster bone-ingrowth . Additionally, leaking-out magnesium or calcium promotes bone growth and bone remodeling .

The bone cement composite according to the present invention can be used in the treatment of a wide variety of orthopedic conditions , in particular in the treatment of bone fractures , bone augmentation, regeneration of bone in large voids , or enhancement of primary fixation of biodegradable or non- degradable implants . By way of non-limiting example , the cement may be inj ected into the vertebral body for treatment of spinal fractures , inj ected into long-bone or flat-bone fractures to augment the fracture repair or to stabili ze the fractured fragments , or inj ected into intact osteoporotic bones to improve strength . The cement is useful in the augmentation of a bone-screw or bone-implant interface . The bone cement may also be used to replace deminerali zed bone matrix materials . Because of its enhanced load-bearing capacity, the cement according to the present invention can provide scaf fold support for various types of vertebral fracture and could be used in tibia plateau reconstruction, wrist fracture reconstruction, calcaneal reconstruction, spine reconstruction and prophylaxis strengthening of the hip bone .

Additionally, the bone cement composite according to the present invention is useful as bone filler in areas of the skeleton where bone is deficient . In this context , the cement is intended to fill, augment, and/or reconstruct maxillofacial osseous bone defects, preferably including periodontal, oral, and cranio-maxillof acial applications. The cements are packed gently into bony voids or gaps of the skeletal system (i.e., extremities, pelvis, and spine) , including use in posterolateral spinal fusion or vertebral augmentation procedures with and without stabilizing hardware, such as pedicle screws and connection rods or expandable intra-vertebral implants. The bone cement may be used to fill defects which may be surgically created osseous defects or osseous defects created from traumatic injury to the bone. In certain embodiments, the cement provides a bone void filler that degrades and is replaced by bone during the healing process.

A further aspect of the present invention relates to the production of fibers. A preferred embodiment of an apparatus is disclosed below.

The metal alloy to be spun comprises at least three chemical elements in the alloy, wherein at least one of these elements is selected from the group consisting of magnesium and calcium. Said metal alloy is heated in a syringe feeding system (Figures 1A and IB) by a heating device 52 until the metal alloy is fully molten, with the syringe feeding system comprising a cylinder 51 and a piston 54. Afterwards, the molten alloy is pressed through a nozzle 60 of the cylinder 51. Outside the nozzle, the melt forms a drop, in which a rotating spinning wheel 56 dips into and drags thin strands of molten alloy out from the drop. The strands of molten alloy being thereby in direct contact with the much colder rotating spinning wheel 56, which is made of metal, is subj ected to a rapid cooling and solidi fication, resulting in fibers with a disordered, or amorphous , atomic-scale structure . The rotating spinning has preferably a sharpened external circumferential edge 63 . Preferably, this edge is non-continuous , forming a plurality of teeth . Only the outermost , sharpened part of the sharpened external circumferential edge 63 dips into the molten alloy drop . Consequentially, the length of the produced fibers corresponds very precisely to the length of the part of the sharpened external circumferential edge that is part of such a tooth .

The apparatus preferably comprises a metallic chamber 10 having a cylindrical portion 12 and a tangentially extending collection tube 14 with a closable port 16 at the end remote from the cylindrical portion 12 . A syringe feeding system comprising the cylinder 51 and the piston 54 ( see Figures 1A and IB ) is mounted inside the chamber 10 right below a rotating spinning wheel 56 and heated by an inductive heating system 52 . Necessary supply lines for a process gas such as argon, for the power of the heating system 52 and for the monitoring of control parameters such as the temperature of the melt exist . Power lines for the heating system 52 enter the chamber 10 via an electrical feedthrough 18 . The chamber 10 can be evacuated by a vacuum pump via an evacuation stub 28 and can be supplied with a flow of an inert or reactive gas via a further feed stub 30 . The wheel 56 is mounted on the inside of and axially parallel to the cylindrical portion 12 and is supported by bearings (not shown) on an axle 20 driven by an electric motor flanged to the rear of the cylindrical portion 12 . The piston 54 can be moved vertically with the help of the li fting rod 62 and a li fting motor 64 that is attached to it . The li fting rod 62 enters the chamber 10 via a linear feedthrough 19 that allows a vertical movement of the li fting rod 62 by at the same time maintaining di f ferent atmospheric conditions inside and outside the chamber 10 . The li fting motor 64 can be equipped with a control system for control of the distance and velocity the rod moves and the force it can exert on the piston and thereby the melt .

The alloy to be spun is placed inside the melting cavity 58 that is formed by the assembled piston 54 and cylinder 51 and heated by a heating device 52 . A temperature sensor 64 , which is placed inside the piston 54 and right below the melting cavity 58 and the metal alloy to be processed, is used in interaction with the heating device 52 to control the temperature of the metal alloy . A piston 54 presses the molten alloy through the noz zle 60 of the cylinder 51 onto the rotating spinning wheel 56 . The spinning wheel has preferably a sharpened external circumferential edge 63 , which has a plurality of teeth . Such a wheel enables the convenient production of thin and short fibers . The noz zle 60 of the syringe feeding system comprising the cylinder 51 and the piston 54 has preferably a noz zle opening of circular shape . The diameter of the opening can be chosen within relatively wide limits , e . g . , 0 . 5 millimeters to 10 millimeters to control the si ze of the molten alloy drop and the rate of flow of the molten alloy from the noz zle onto the sharpened external circumferential edge of the wheel in interaction with the li fting velocity of the piston 54 . A further aspect of the present invention relates to the coating of the fibers . Preferably, said fibers are partly or fully coated by a method selected from the group consisting of deposition coating, plasma electrolytic oxidation coating, plasma electrolytic anodi zation coating, anodi zation coating, fluoric conversion coating, Mg ( OH) ₂ coating, calcium phosphate conversion coating, phosphate conversion coating, hydroxy-apatite coating, organic coating, biodegradable polymer coating, and sol-gel coating . This coating allows to prevent a chemical reaction between the fibers and the setting solution during setting of the cement . Furthermore , it can delay degradation onset during in vivo application or slow down degradation rate in the beginning of an in vivo application .

Figures :

Figure 1A and IB show a schematic drawing of an apparatus for the fiber production . Figure IB provides a cross-sectional view of the main components that are needed to form the molten alloy and control the fiber spinning process .

Figures 2A and 2B show the di f ference of the fibers before and after the treatment with diammonium hydrogen phosphate coating solution . A color change occurs due to the treatment .

Figures 3A and 3B show scanning electron microscopy images of an uncoated short fiber made from Mg ₆ oZn ₃₅ Ca5 ( in at % ) ( Figure 3A) and sections of fibers made from the same alloy after coating by immersion in diammonium hydrogen phosphate solution ( Figure 3B ) . Figures 4A and 4B show an embodiment of the present invention with uncoated fibers of a fiber volume fraction of 10 % and a cement powder to liquid ratio of 2 . 5 grams per milliliter . The material ' s volume increased signi ficantly during setting of the cement , and macropores formed on the surface as well as in the bulk .

Figures 5A-5C show an embodiment of the present invention with coated fibers and cement powder to liquid ratio of 3 grams per milliliter . The set fiber-reinforced bone cement was embedded in bakelite and ground with sandpaper, and the images were taken with a light microscope . The shown embodiments comprise coated fibers in a volume percentage of 10 % ( Figure 5A) , 15 % ( Figure 5B ) and 20 % ( Figure 5C ) .

Figures 6A and 6B show scanning electron microscopy images of cross sections of embodiments that were produced using a cement powder to liquid ratio of 3 grams per milliliter and coated fibers in volume percentage of 10 % ( Figure 6A) and 15 % ( Figure 6B ) . The set fiber-reinforced bone cements were embedded in bakelite and ground with sandpaper before taking the images . The samples show no signs of pores .

Figures 7A and 7B show graphical visuali zations of determined mechanical properties of fiber-reinforced bone cements according to the present invention with respect to the fiber volume fraction . Coated fibers were used . The presented values are given as mean values with their respective standard deviation shown as error bars . "Reference" refers to plain bone cement without fiber addition .

Figure 8 shows typical graphs for bone cements according to the present invention with coated fibers of di f ferent fiber volume fractions as tested in 3-point bending . Additionally, plain bone cement without fiber addition was tested in the same way and is displayed labeled as "Reference" .

Figure 9A- 9F shows mean curves (black) with the corresponding standard deviations ( grey) of 3-point bending tests of bone cements without fibers ( reference material ) and with coated fibers of di f ferent fiber volume fraction . The reference material was chosen to be represented only by one typical curve , because the mean curve would indicate a false ductile behavior . All other samples do show ductile behavior .

Figure 10 shows a scanning electron microscopy image of a typical fracture surface after a 3-point bending test of a fiber-reinforced bone cement composite according to the present invention . Several broken fibers are visible , in some cases with short pullout sections .

Figure 11 shows a comparison of fiber-reinforced bone cement composites according to the present invention with coated and non-coated fibers tested in 3-point bending . Both examples were prepared with a fiber volume fraction of 10 % and a powder to liquid ratio of 2 . 5 grams per milliliter . The example with uncoated fibers shows signi ficant lower mechanical strength and a signi ficant lower flexural modulus .

Figure 12 shows the cell viability of the di f ferent materials .

Figures 13A and 13B show cell number of the di f ferent materials . Examples

Fiber Production

An alloy with composition Mg ₆ oZn ₃₅ Ca5, with the subscripted numbers giving the relative amounts of the respective element in atomic percent, made of high-purity components was used to produce highly amorphous fibers. This alloy has a density of about 3.062 grams per cubic centimeter and a melting temperature of about 390°C.

To produce the fibers, an in-house-built wire spinner was used. An original, simple version of this wire spinner has been reported in (B. Zberg, E. R. Arata, P. J. Uggowitzer, and J. F. Ldffler, "Tensile properties of glassy MgZnCa wires and reliability analysis using Weibull statistics," Acta Materialia, vol. 57, no. 11, pp . 3223-3231, 2009, doi: 10.1016/ j . actamat .2009.03.028 ) and was modified as described above and visualized in Figures 1A and IB.

The amorphous structure of the fibers can be achieved by rapid cooling. This rapid cooling and solidification of the melt is carried out during the spinning process when the melt gets in contact with the spinning wheel. The above-mentioned spinning wheel construction and the small diameter of the fibers facilitate to reach the required high cooling rates. The maximum casting thickness for this alloy to achieve an amorphous state after solidification is around 2-4 mm, which is well above the targeted fiber thickness. This process allows to freeze the disordered state of the liquid alloy and hinders a long-range periodic arrangement of the metal atoms. The maximum casting thickness (also called critical casting thickness) is related with the minimum cooling rate (also called critical cooling rate ) in rapid solidi fication according to the equation where T _c is the critical cooling rate in Kelvin per seconds , f is a constant that has an empirically determined value ranging from about 20 to about 40 in centimeters squared times Kelvin per second, and d _c is the critical casting diameter in centimeters . For the here employed alloy this equation results in a critical cooling rate of about 125 to 1000 Kelvin per seconds . To achieve the desired amorphous state the material needs to be solidi fied at a cooling rate that is higher than the critical cooling rate .

A block of the above-mentioned alloy (Mg ₆ oZn ₃₅ Ca5, with the subscripted numbers giving the relative amounts of the respective element in atomic percent . ) was cut into pieces of si ze 12 x 12 x 25 millimeters ( around 10- 12 grams of weight ) . After cutting, the pieces were ground until visible staining (potentially oxides ) on the surface was mainly removed, and subsequentially cleaned with ethanol .

The wire spinning process was conducted inside of a high vacuum chamber . The small metal samples were placed on the piston and moved into the cylinder to the height of the heating system in the form of a copper induction coil . Afterwards , the vacuum pump was started, and the chamber was flushed with argon ( 400 millibars ) two times starting at a pressure < 10 ^-1 millibars . Before the chamber was ultimately filled with argon, a third time to 800 mbar, it was ensured with a high-vacuum pump that the pressure reached a value lower than 10 ^-5 millibars . The sample was then molten with inductive heating of the water-cooled copper coil and temperature was measured with a thermocouple placed inside the piston . As soon as the desired temperature of about 430 ° C was reached, the molten sample was pushed out of the noz zle on top of the cylinder with the help of the motori zed li fting rod that pushed the piston up . A small droplet of sample formed at the opening, before the rotating spinning wheel with a multitude of teeth started to extract the melt and ej ect the spun short fibers into the collection tube . Each tooth segment measured about 2 millimeters of the wheel ' s sharpened external edge . Thereby fibers of a length of about 2 millimeters and narrow length distribution were produced conveniently . The motori zed li fting rod and the attached piston move upwards at a constant velocity and thereby provide a constant supply of molten material to the spinning wheel . At the end of the spinning process , the heating was stopped for the system to cool down . Subsequently, the fibers were extracted from the collection tube and stored in a plastic beaker . Afterwards , the fibers were sieved in a sieve shaker ( Fritsch Laborgeratebau, W . -Germany) with a mesh si ze of 0 . 355 millimeters to get rid of all parts that are not fibers .

Continuous fiber production

In a similar manner, long, continues fibers were produced . However, instead of a spinning wheel with a multitude of teeth, a spinning wheel with a continuous sharpened external edge (without teeth) was used . The rotating speed of the spinning wheel was set to 1750 revolutions per minute . Fiber characterization:

To verify the amorphous state of the fibers, x-ray diffraction measurements (X' Pert-Pro with Cu-Ka source) in a 0-20 set up were performed. Each spectrum was recorded for 1 hour with a Seller slit. The measurements were recorded in an angle range of 5-70°. The x-ray diffraction measurements showed similar spectra for both the short and the long fibers. In both cases, no crystalline phase was detected, so it could be concluded that the fibers were amorphous. Furthermore, the geometry and surface quality were investigated and characterized by scanning electron microscopy (Hitachi SU-70) and light microscopy.

Coating of the short fibers

After the sieving, the short fibers were placed into a 3.5 molar diammonium hydrogen phosphate solution for 4 hours under constant stirring. After this treatment, the fibers were washed with deionized water several times until the waste water was not opaque anymore. Furthermore, the fibers were rinsed with ethanol three times and dried in air at room temperature .

As soon as the diammonium hydrogen phosphate solution was added to the fibers, many bubbles started to form originating from the fibers. The gas bubbles moved to the liquid's surface and an ammonium smell was present. Moreover, the solution turned opaque immediately after the fibers were added. Furthermore, during the treatment the color of these fibers changed from a shiny-silver to a mat-brownish color (Figures 2A and 2B) . An analysis of the coated fibers by scanning electron microscopy revealed that the surface experienced a signi ficant change . The scanning electron microscopy images show that fibers after the immersion treatment feature a rough surface ( Figure 3B ) and not smooth as before the treatment ( Figure 3A) . The described treatment leads to coating of the fibers . This coating covers the entire length of the fibers as well as the face sides at each end of a fiber .

Chemical analysis performed by energy dispersive x-ray analysis of a coated fiber showed that the coating comprises mainly the elements magnesium, phosphorus , oxygen, calcium and zinc . Table 1 lists the elements detected in the fiber coating and its relative amounts in atomic percentages . Interestingly, the elements calcium and zinc are strongly underrepresented in the coating with respect to their relative amount in the fiber material .

Table 1 : Chemical composition of the fiber coating as detected with energy dispersive x-ray analysis . The main elements detected are magnesium, phosphorus , and oxygen . Production of Fiber-reinforced Cements and Characterization

Bone cement preparation

Composites of bone cement and fibers according to the present invention were produced with different fiber volume fractions. The cement material was produced as follows:

The individual components

- calcium hydrogen phosphate (CaHPCy, >97%, Fluka Chemie AG, Buchs, Schweiz) ,

- calcium carbonate (CaCOs, 98+% extra pure, Lot: 1879814, Fisher Chemical, Lough-borough, UK) ,

- magnesium hydrogen phosphate trihydrate (MgHPO ₄ -3H ₂ O, 99%, Lot: 10227928, Alfa Aesar, Kandel, Germany) and

- magnesium hydroxide (Mg (OH) >95 %, Fluka Chemie AG, Buchs, Schweiz) were prepared in amounts according to Table 2 and filled together in a 375 milliliter aluminum oxide crucible. To ensure good mixing the powder mixture was thoroughly stirred with a spatula and slightly tapped on the ground for compacting. The crucible was then transferred into a furnace (Nabertherm Schweiz AG) and heated to 1100 °C with a ramping of 45 minutes. It was kept at that temperature for 5 hours before it was slowly cooled down to room temperature inside the furnace. After the resulting sintered powder cake had cooled down, it was crushed in an agate mortar to a fine powder and sieved in a sieve shaker (Fritsch Laborgeratebau, Germany) with a mesh size of 1 millimeter. Afterwards, the powder was dry-milled in a planetary ball mill (Changsha Tianchung Powder Technology Co. Ltd.) for 2 hours at 200 revolutions per minute. The beakers were 1 liter in volume and made of ZrO ₂ . Per beaker, 350 grams of zirconia balls of diameter 10 millimeter were added together with 125 grams of sieved powder. After ball milling, the powder was removed and stored in a dry place.

Table 2: Amounts of each cement component for one batch of bone cement as produced in the present example. The amount is given in mol.

A stock solution of a 3.5 molar diammonium hydrogen phosphate ( (NH ₄ ) ₂ HPO4, 99%, ACS reagent, Lot: A0421666, Acros Organics, Waltham, Massachusetts, USA) was prepared and used as the liquid part for the cement preparation and is required to achieve a setting of the cement. Therefore, it is also called setting solution.

Several samples were prepared with different ratios of cement powder to liquid setting solution. The ratio of cement powder to liquid setting solution in grams per milliliter is well known in the art as "powder to liquid ratio" (PLR) . Powder to liquid ratios in a range from 2.5 grams per milliliter to 4 grams per milliliter in steps of 0.5 grams per milliliter were realized. The samples with a powder to liquid ratio of 3 grams per milliliter were made by adding 9 grams of cement base material mixing with 3 milliliters of 3.5 molar diammonium hydrogen phosphate stock solution. The other samples were mixed according to their respective powder to liquid ratio. After mixing cement powder and stock solution, the mixture was immediately filled in a planetary mixer (Thinky Mixer ARE-250) without any grinding balls for 30 seconds at 2000 revolutions per minute. The resulting paste was then transferred into a silicon rubber mold with a plastic spatula. After the filling, the molds with the cement were slightly tapped on the table several times, to enhance the compaction. Each sample was individually prepared to ensure that all samples experience identical curing conditions with respect to time. For the curing process, a glass lid was put on the open surface of the mold to achieve a plane top surface. In a first phase, the samples were cured with the glass lid overnight. The next day, the glass lid was carefully removed, and the samples then continued the curing in air for another 24 hours. The whole curing process took 48 hours .

The crystalline phases of the cement after hardening were determined by x-ray powder diffraction (X' Pert-Pro with Cu-Ka source) in a 0-20 set up. The spectrum was recorded for 1 hour with a Seller slit. The measurement was recoded in an angle range of 5-70°. To find the containing cement phases and their respective phase fractions, a Rietveld refinement was performed on the spectra. The following phases were detected: struvite (NH ₄ MgPO ₄ • 6H ₂ O) , stanfieldite

(Ca ₄ Mg ₅ (PO ₄ ) ₆ ) , farringtonite (Mg ₃ (PO ₄ ) ₂ ) , and brushite (CaHPO ₄ -2H ₂ O) (Table 3) . However, the evidence for brushite is rather low and there is a chance that the detected values are artefacts. Table 3 : Crystalline cement phases and their respective phase fractions in percent as detected in the reali zed bone cements with respect to the powder to liquid ratio ( PLR) in grams per milliliter . The error of these numbers is in the range of a few percentage points .

Bone cement composite with uncoated fibers Fibers according to the present invention were prepared as described above by employing a melt spinning process . No modi fications were made except of sieving . Then, the uncoated fibers were in a first step mixed with powder of magnesium calcium phosphate cement that was prepared as described above for 30 minutes in a planetary mixer ( Thinky Mixer ARE-250 ) at 200 revolutions per minute , without grinding balls . Subsequently, in a second step, the resulting powder- fiber mixture was mixed with 3 . 5 molar diammonium hydrogen phosphate solution.

The resulting paste was then transferred into a silicon rubber mold with a plastic spatula. After the filling, the molds with the cement were slightly tapped on the table several times, to enhance the compaction. Each sample was individually prepared to ensure that all samples experience identical curing conditions with respect to time. For the curing process, a glass lid was put on the open surface of the mold to achieve a plane top surface. In a first phase, the samples were cured with the glass lid overnight. The next day, the glass lid was carefully removed, and the samples then continued the curing in air for another 24 hours. The whole curing process took 48 hours.

Samples of different powder to liquid ratio and different fiber volume fractions were produced (Table 4) . All percentages in Table 4 are given in volume percent of fibers to cement base material.

Table 4: Number of samples produced with uncoated fibers with the corresponding powder to liquid ratio (PLR) and the realized fiber fractions in volume percent. After the resulting composite ' s setting process was finished, it was found that the composite signi ficantly gained in volume , caused by large pores that formed inside the material ( Figures 4A and 4B ) . By variation of the powder to liquid ratio and fiber volume fractions it was found that the pore formation was more pronounced with lower powder to liquid ratios and higher fiber volume fractions . It can be safely assumed that the large number of pores causes strongly reduced mechanical strength . Therefore , a structural application can be rather excluded as a potential field of application . However, for non- or low-loadbearing applications and some types of slow-degrading bone cements , for example apatitic calcium phosphate cements , the addition of such uncoated fibers and the resulting pores are beneficial to increase the degradation rate and/or boneingrowth of the resulting compound compared to the original bone cement without fibers . It is also conceivable to use instead of the here described highly amorphous magnesium- based fibers powder of magnesium or calcium or a magnesium- based alloy or a calcium-based alloy to achieve a similar pore- forming ef fect during setting of the cement .

Bone cement composite with coated fibers

Fibers according to the present invention were prepared as described above by employing a melt spinning process . Then these fibers were coated as described above by immersion in a 3 . 5 molar diammonium hydrogen phosphate solution . Subsequently, the coated fibers were mixed with magnesium calcium phosphate cement powder that was prepared as described above for 30 minutes in a planetary mixer (Thinky Mixer ARE-250) at 200 revolutions per minute, without grinding balls. After removing the power-fiber mixture and placing it in a glass beaker, 3.5 molar diammonium hydrogen phosphate solution was added.

The resulting paste was then transferred into a silicon rubber mold with a plastic spatula. After the filling, the molds with the cement were slightly tapped on the table several times, to enhance the compaction. Each sample was individually prepared to ensure that all samples experience identical curing conditions with respect to time. For the curing process, a glass lid was put on the open surface of the mold to achieve a plane top surface. In a first phase, the samples were cured with the glass lid overnight. The next day, the glass lid was carefully removed, and the samples then continued the curing in air for another 24 hours. The whole curing process took 48 hours.

Samples of different powder to liquid ratio and different fiber volume fractions were produced. Table 5 lists all the samples produced with the coated fibers.

Table 5: Number of samples produced with coated fibers with the corresponding powder to liquid ratio (PLR) and the realized fiber fractions in volume percent.

Figure 5A to 5C show samples of the realized bone cement composites according to the present invention with coated fibers and a powder to liquid ratio of 3 grams per milliliter and with a fiber volume ratio of 10 % ( Figure 5A) , 15 % ( Figure 5B ) , and 20 % ( Figure 5C ) . The shown samples were embedded in bakelite and ground with 1200 and 4000 grid sandpaper . The images were taken with a light microscope .

Scanning electron microscopy images of these samples show furthermore excellent integration of the fibers into the cement matrix . No gaps or pores are visible in the samples with a fiber volume ratio of 10 % ( Figure 6A) and a fiber volume ratio of 15 % ( Figure 6B ) .

3-Point Bending Tests

Samples of fiber reinforced bone cements that were prepared as described above were tested in a universal testing machine ( Schenck Trebel ) in 3-point bending with a span length of 60 millimeters . Before the mechanical test , the setup was carefully aligned and fixed in this position . The samples had a dimension of about 75 x 10 x 7 millimeters . The surface that was covered with glass during the curing of the cement was placed facing down . This surface was clean, and no apparent defects were present . Therefore , it can be safely assumed that the result was not altered by stress concentrations from defects at the surface . The samples were tested immediately after the 48-hour curing period in dry condition and no further treatment . Testing was performed in speed-controlled mode at a testing speed of 0 . 5 millimeters per minute. Testing was continued until sample failure, or stopped when the force dropped below 10 newtons.

From the obtained data flexural strength and flexural modulus were calculated. The following formulas were used.

Flexural strength:

Flexural modulus:

L ³ m F ^{flex =} 4^ where F _max is the maximum recorded force, L is the span length, w is the width of the sample, h is the height of the sample and m is the gradient (i.e., slope) of the initial straight-line portion of the load deflection curve.

Table 6 presents the obtained results on mechanical properties of fiber-reinforced bone cement according to the present invention with respect to the fiber-volume ratio.

Table 6: Mechanical properties of tested fiber-reinforced bone cements as determined by a 3-point bending test. Fibers were coated in a process as described above. Flexural strength and flexural modulus are listed with respect to the fiber volume fraction. The values are stated as mean values and their standard deviation.

It was found that the flexural modulus depends only moderately on the fiber volume fraction up to a fiber volume fraction of about 18 % ( Figure 7A) . The highest flexural modulus was obtained with 13 % of fiber volume fraction . Notable is that initially with 10 % fiber volume fraction the flexural modulus apparently decreased compared to the reference , which was plain bone cement without fibers . Moreover, at higher fiber volume fractions the flexural modulus drops signi ficantly .

Furthermore , it was found that the addition of 10 % fibers apparently decreases the flexural strength compared to the reference material . However, at a fiber volume fraction of 13 % the composite shows a signi ficant higher flexural strength compared to the reference material without fibers and to the composite with a fiber volume fraction of 10 % ( Figure 7B ) . At this fiber volume fraction also the highest flexural strength of all investigated samples was found . A fiber volume fraction of 20 % led to a signi ficant drop in flexural strength . These properties are visually represented as bar plots in Figures 7A and 7B .

Figure 8 shows typical 3-point bending graphs for all investigated specimens . The reference material ( containing no fibers ) was the only investigated material that failed in a clearly brittle manner . The graphs of the fiber-reinforced bone cements show occasionally occurring abrupt drops in stress , which can be attributed to single failing fibers .

Figures 9A- 9F show the obtained results as mean curves in black and standard deviations in gray for all tested materials of di f ferent fiber volume fractions . In case of the reference material only one representative curve is depicted since the mean curve would indicate a ductile failure which was never observed . Only the reference samples with no fibers clearly showed brittle failure . All other samples showed ductile failing behavior . All investigated specimens failed by showing only one visible crack . Multiple crack formation was not observed .

Figure 10 shows a scanning electron microscopy image of a typical fracture surface after the 3-point bending test of fiber-reinforced bone cement . The analysis of the fracture surface revealed that most of the fibers broke directly at the fracture surface , and only few appeared to be pulled out . This indicates an excellent fiber-cement interface strength and furthermore leads to the conclusion that even shorter fibers of the same diameter would lead to equally good mechanical properties . Shorter fibers would further improve malleability and inj ectability of the fiber-reinforced bone cement during application .

A comparison of the fiber-reinforced bone cements with coated and uncoated fibers shows a signi ficant di f ference in their mechanical behavior ( Figure 11 ) . Fiber-reinforced bone cements with uncoated fibers exhibit a signi ficantly lower flexural strength and a lower flexural modulus compared to fiber-reinforced bone cements with coated fibers . On the positive side , fiber-reinforced bone cements with uncoated fibers also show a ductile failing behavior, although not as pronounced as fiber-reinforced bone cements with coated fibers .

Injectability Study

Bone cement composites according to the present invention with coated fibers with di f ferent powder to liquid ratios ( PLR) were tested on their inj ectability . The bone cement composites were prepared as described above . Recorded was the consistency of the paste after mixing with the setting solution, how big the minimum syringe opening was at which the paste was still inj ectable , how well it was inj ectable with this opening, i f the push-out force increases during the inj ection, and i f glycerol would enhance the inj ectability i f it was used instead of stock solution . All tests were done by hand and qualitatively assessed rather than quantitatively .

It was found that the inj ectability improved signi ficantly with the addition of glycerol to the mixture of powder and diammonium hydrogen phosphate solution . With this combination the material was extrudable with no force increase and no separation of liquid and powder phases . It is equally conceivable that other liquid additions of high viscosity can improve inj ectability in a similar manner .

Bone cement in vitro extracts and cell viability :

Bone cement composites according to the present invention with coated fibers were subj ected to extraction according to ISO 10993- 12 and tested for in vitro cytotoxicity according to ISO 10993-5.

The following samples were used. Fxx indicates specimens with xx % fiber volume fraction. FOO served as reference of plain calcium phosphate cement without fibers, and Monetite and Brushite served as well-known controls.

- FOO = 6 mm diameter, 1.5 mm thickness, 84.82 mm2 surface area, sample number (n) = 50

- F10 = 6 mm diameter, 1.5 mm thickness, 84.82 mm2 surface area, n = 53

- F13 = 6 mm diameter, 1.5 mm thickness, 84.82 mm2 surface area, n = 56

- F15 = 6 mm diameter, 1.5 mm thickness, 84.82 mm2 surface area, n = 53

- F18 = 6 mm diameter, 1.5 mm thickness, 84.82 mm2 surface area, n = 62

- Monetite = 6 mm diameter, 1.2 mm thickness, 79.17 mm2 surface area

- Brushite = 6 mm diameter, 1.2 mm thickness, 79.17 mm2 surface area

Experimental Methods

All the materials, except the monetite and brushite, were sterilized by gamma radiation. The monetite and brushite samples were sterilized by immersion in isopropanol in combination with ultrasound for 10 minutes. After that, the samples were briefly washed in PBS and let them dry in a sterile environment. After sterilization, all the materials were submerged in alpha-MEM medium (containing 10% of FBS and 1% of pen-strep) for 24 h, at 37 °C and 5% CO2. The extraction ratio utilized was 125 mm ² * ml ^-1 . One day prior the start of the experiment, 10.000 cells per well for six technical replicates for each group were seeded in a 96-well plate and allowed them to adhere and form a cell layer overnight. After that, the cells were cultured for three days with the conditioned media produced from the different materials. Then, two different cell assays were performed. A 10% Alamar Blue HS + medium solution was used to assess cell viability as a function of their metabolic activity and a LDH detection assay was used to count the cells number in each well. Normally, this kind of assay is used to measure cytotoxicity but, in this case, the cells were completely lysed right before the measure allowing to actually estimate their biomass. Moreover, a standard curve (Figure 13B) was built performing the same procedure in a group containing different known numbers of cells. This standard curve was then used to extrapolate the number of cells in the other groups. All the data were first explored in excel and then a One-way ANOVA plus Bonferroni post hoc correction was performed in GraphPad Prism

Results

Cell Viability (Figure 12) : All the different materials (from FOO to F18) show a substantial improvement (statistically significant) compared to the control group (cells only) . FOO, F13 and F15 have the highest values, while F10 and F18, while showing improvement over the control, also have the lowest values among all groups.

Cells Number (Figures 13A and 13B) : The only group amongst the different material compositions to show an increase in cell proliferation is F13 (p < 0.001) . The other groups seem to behave just like the control. Only F18 performed poorly, with a statistical decrease (p < 0.01) in cell proliferation.