In this hemp fiber composite the silane-treated fiber was held at 20 wt percent while a bio-based polysaccharide filler was stepped up from 0 to 25 wt percent. Every property that improved peaked at the same 15 wt percent filler level. Tensile strength hit 46.5 MPa, flexural strength 69.3 MPa, and impact strength 5.70 kJ/m2. Past that level all of them fell back.
Why a natural gum goes into a hemp fiber composite at all
Unsaturated polyester is the workhorse thermoset matrix. It is cheap, it processes easily, and it holds its dimensions. It is also brittle, and it does not bond well to cellulose.
Natural fiber makes that second problem worse before it makes anything better. Cellulose is covered in hydroxyl groups, so it holds water and refuses to wet out in a nonpolar matrix. The result is a weak interface, poor stress transfer, and mechanical numbers that fall short of what the fiber strength alone would lead you to expect.
Plant-derived gums are interesting here because they are full of hydroxyl groups too. That sounds like a problem, and in one respect it is. It also means the filler can interact with both the fiber surface and the matrix, which is the same job a coupling agent does in a mineral-filled system. A bio-based gum is not a classical compatibilizer, because it does not carry two deliberately different reactive ends. It can still sit at the interface and improve load transfer.
How the fiber was treated before it met the resin
The hemp fiber was treated in two stages before composite fabrication, and both stages matter when you read the results.
| Step | Condition | Why it is done |
|---|---|---|
| Alkali treatment | Sodium hydroxide solution | Remove pectin, hemicellulose, waxes and surface impurities |
| Silane treatment | 2 wt% aminopropyltriethoxysilane in 95:5 ethanol to water | Bridge the polar fiber surface to the nonpolar matrix |
| pH adjustment | Acetic acid to about pH 4.5 | Drive silane hydrolysis before condensation |
| Soak time | 60 minutes at room temperature | Allow the coupling agent to reach the fiber surface |
| Drying | 80 degrees Celsius for 24 hours | Remove water before it competes with the resin |
Silane coupling agents are the standard answer to a polar fiber sitting in a nonpolar matrix, and the mechanism is well understood. The silane hydrolyzes, condenses onto the cellulose hydroxyls, and leaves an organofunctional tail that the polyester can interact with. Getting the pH and the drying right is what decides whether that happens, or whether the silane just self-condenses in solution.
The six formulations tested
| Sample | Polyester resin | Hemp fiber | Polysaccharide filler |
|---|---|---|---|
| HC0 | 80 wt% | 20 wt% | 0 wt% |
| HC1 | 75 wt% | 20 wt% | 5 wt% |
| HC2 | 70 wt% | 20 wt% | 10 wt% |
| HC3 | 65 wt% | 20 wt% | 15 wt% |
| HC4 | 60 wt% | 20 wt% | 20 wt% |
| HC5 | 55 wt% | 20 wt% | 25 wt% |
The filler was ground, sieved below 100 micrometers, then dried and stored with desiccant before use. Curing used 12 wt percent methyl ethyl ketone peroxide catalyst, with hand lay-up followed by compression molding at about 0.51 MPa for 24 hours at room temperature, then a post-cure at 80 degrees Celsius for 2 hours.
Hemp content was deliberately held constant, and that is what makes this a clean filler-loading study rather than a fiber-loading study.
Where the filler optimum actually sits
| Property | HC0 | HC1 | HC2 | HC3 | HC4 | HC5 |
|---|---|---|---|---|---|---|
| Tensile strength (MPa) | 29.2 | 35.8 | 41.2 | 46.5 | 42.8 | 38.9 |
| Tensile modulus (GPa) | 1.62 | Not given | Not given | 2.78 | 2.52 | 2.30 |
| Flexural strength (MPa) | 46.5 | 55.2 | 61.8 | 69.3 | 63.7 | 58.9 |
| Flexural modulus (GPa) | 2.12 | Not given | Not given | 3.28 | 3.02 | 2.78 |
| Impact strength (kJ/m2) | 3.1 | 3.9 | 4.8 | 5.70 | 5.1 | 4.6 |
| Dielectric strength (kV/mm) | 11.2 | 12.8 | 14.5 | 16.3 | 15.1 | 13.9 |
| Water absorption (%) | 0.85 | 1.42 | 1.98 | 2.36 | 2.78 | 3.15 |
The shape of this table is the whole finding for this hemp fiber composite. Tensile strength rose 59 percent from HC0 to HC3, then gave back about 16 percent by HC5. Flexural strength rose 49 percent and gave back about 15 percent. Every mechanical, thermal, and dielectric response peaked at the same formulation.
That is a genuine optimum, not noise. When six independent measures all turn around at the same loading, the mechanism is real, and it is almost certainly dispersion rather than chemistry.
Why interfacial adhesion decides where the optimum sits
The composite was evaluated by dynamic mechanical analysis, and storage modulus at 30 degrees Celsius tells the story most clearly. It went from about 450 MPa for HC0 to 950 MPa for HC1, 1800 MPa for HC2, and 3200 MPa for HC3, then fell to 1300 MPa for HC4 and 700 MPa for HC5.
The damping factor tells the same story from the other side. The tan delta peak, which rises when the interface is weak and energy is lost to internal friction, sat at about 0.135 for HC0, climbed to 0.19 and 0.22 for HC1 and HC2, then dropped to about 0.16 at HC3 and rose again to 0.25 and 0.28 at HC4 and HC5. A low tan delta peak sitting next to high storage modulus is the signature of better interfacial adhesion and restricted chain mobility.
Above 15 wt percent filler the mechanism reverses. Resin viscosity rises, the resin stops penetrating the fiber bundle completely, particles cluster, and voids form. Those clusters are stress concentrators, so a filler that bonded well at 15 percent becomes a defect at 25 percent.
This is the practical lesson for anyone dosing a filler or a compatibilizer into a hemp fiber composite: the loading window is set by dispersion, not by the chemistry of the interface. Better interfacial adhesion lifts the ceiling inside that window. It does not widen the window itself.
What dynamic mechanical analysis showed beyond the peak values
The high-temperature end is worth a look too. At 140 degrees Celsius the optimized composite retained a storage modulus near 90 MPa while the unfilled composite held only about 15 MPa. At 70 degrees the optimized sample was near 1200 MPa and at 100 degrees near 300 MPa.
Cole-Cole analysis plots loss modulus against storage modulus, and it gave a near semicircular arc for the optimized sample, with a peak loss modulus around 1200 MPa at a storage modulus of about 2600 to 2800 MPa. The overloaded samples gave stretched, broken arcs with higher loss modulus, which is the standard signature of a heterogeneous structure.
One number needs a flag. The abstract reports a storage modulus of 4250 MPa at the optimum, while the results section gives about 3200 MPa at 30 degrees Celsius. This summary uses the results-section value, since that is where the measurement conditions get described.
The trade nobody escapes: water absorption rises with filler
Water absorption climbed in a straight line with filler loading, from 0.85 percent at HC0 up to 3.15 percent at HC5. The optimized sample sat at 2.36 percent.
This is the cost of the mechanism that produced the gains. Hydroxyl-rich interfaces move load well, and they move water just as well. A formulation that improves mechanical and dielectric performance by adding a hydrophilic filler will, almost without exception, take on more moisture.
For indoor structural or insulation use that may not matter much. For any application with humidity cycling or outdoor exposure, it is the number that decides whether the formulation survives. It has to be designed against, not measured afterward.
What a formulator should take from this
- Hold the fiber loading constant when you study filler loading. Otherwise the two effects are entangled and the optimum you find is not reproducible.
- Expect the optimum to be set by dispersion and resin viscosity, not by interface chemistry. Watch viscosity as closely as you watch strength.
- Silane coupling agent treatment is doing real work here. Skipping it and hoping the filler compensates is not the same experiment.
- Water absorption moved in the opposite direction to every mechanical property. Plan for that trade from the start.
- Six independent responses turning around at one loading is a strong signal. One response turning around is not.
Limitations
Water absorption rose monotonically across the whole loading range, so the optimized formulation is not the best choice where moisture resistance dominates. The storage modulus at the optimum is reported as about 3200 MPa at 30 degrees Celsius in the results section and as 4250 MPa in the abstract; this summary uses the results value and flags the discrepancy.
Scanning electron microscopy of the fracture surfaces is described by the authors as qualitative morphological evidence, not as quantitative proof of interfacial bond strength. Fiber pull-out, void content, and particle agglomeration were never measured by image analysis, so the dispersion mechanism behind the optimum is inferred rather than quantified. The filler was capped at 25 wt percent because resin viscosity made processing impractical above that level, so the true processing ceiling may sit on either side of it. Only dielectric strength was measured among electrical properties, with dielectric constant and dielectric loss left for later work. Mechanical testing was run at room temperature, and no creep, fatigue, or long-term moisture aging data is reported.
Bottom line
A hemp fiber composite with silane-treated fiber held at 20 wt percent and a bio-based polysaccharide filler stepped to 25 wt percent peaked at 15 wt percent filler on every measure tested: 46.5 MPa tensile, 69.3 MPa flexural, 5.70 kJ/m2 impact, 16.3 kV/mm dielectric strength, and about 3200 MPa storage modulus at 30 degrees Celsius. Above that loading, viscosity and clustering reversed the gains. Water absorption rose throughout, which is the trade attached to a hydroxyl-rich interface.
FAQ
What is a hemp fiber composite
A hemp fiber composite uses hemp bast fiber as the reinforcement inside a polymer matrix, in this case an unsaturated polyester resin. Hemp brings high strength per unit weight and low density, and it grows back every season. The difficulty is that its cellulose surface is polar and holds water, while the polyester matrix does not.
Why does the filler dosage have an optimum at all
Below the optimum, filler particles disperse well and improve load transfer across the interface. Above it, resin viscosity climbs, the resin stops wetting the fiber bundle completely, and particles cluster into voids. Those clusters act as stress concentrators, so strength falls even though more filler was added.
How much did the properties improve at the optimum
At 15 wt percent filler, tensile strength rose 59 percent to reach 46.5 MPa and flexural strength rose 49 percent to reach 69.3 MPa against the unfilled composite. Impact strength reached 5.70 kJ/m2 from a 3.1 kJ/m2 baseline, and dielectric strength rose by about 45 percent to reach 16.3 kV/mm, and tensile modulus peaked at 2.78 GPa.
What happened above 15 percent filler
Every mechanical property gave back part of its gain. Tensile strength fell from 46.5 MPa to 38.9 MPa and flexural strength from 69.3 MPa to 58.9 MPa at 25 wt percent. Storage modulus at 30 degrees Celsius fell from about 3200 MPa to about 700 MPa, which is the clearest sign that dispersion had broken down.
Does silane treatment matter in this system
Yes, and it is doing the work usually assigned to a silane coupling agent in filled thermosets. The silane hydrolyzes, condenses onto the cellulose hydroxyls, and leaves an organofunctional tail that the polyester matrix can interact with. Without that step the polar fiber and the nonpolar matrix stay poorly bonded to each other.
What did storage modulus show
Storage modulus at 30 degrees Celsius rose from about 450 MPa unfilled to 950 MPa, then 1800 MPa, and about 3200 MPa at the optimum. It then dropped to 1300 MPa and 700 MPa at the two highest loadings. At 140 degrees Celsius the optimized sample still held near 90 MPa, against only about 15 MPa for the unfilled one.
What does the damping factor tell you
The tan delta peak rises when the interface is weak and energy is lost to internal friction inside the material. Here it sat near 0.135 unfilled, climbed through 0.19 and 0.22, fell back to about 0.16 at the optimum, then rose again to 0.25 and 0.28. Low damping alongside high stiffness points to better interfacial adhesion across the fiber-matrix boundary.
Is a plant gum the same as a compatibilizer
Not in the classical sense. A true compatibilizer carries two deliberately different reactive ends, one for each phase. A plant gum is hydroxyl-rich and can interact with both the fiber and the matrix, so it improves load transfer, but it is better described as an interfacial filler than as a designed compatibilizer.
Why does water absorption get worse
The same hydroxyl groups that let the filler bond well to cellulose also bind water molecules readily. Absorption rose from 0.85 percent unfilled to 2.36 percent at the optimum formulation and 3.15 percent at the highest loading. Any formulation that gains strength in this particular way will also take on more moisture over its service life.
Is the storage modulus figure consistent
Not entirely, and that is worth knowing before you quote the number. The results section gives about 3200 MPa at 30 degrees Celsius, while the abstract reports 4250 MPa. This summary uses the results-section value because that is where the measurement temperature is actually stated. The gap between the two should be resolved against the full data set before the number is quoted.
What is missing from the reported data
Fracture surface analysis is qualitative rather than quantitative, so void content, pull-out length and agglomeration were never measured by image analysis. Creep, fatigue and long-term moisture aging are absent. Dielectric constant and dielectric loss were left for later work, and loading was capped at 25 wt percent.