3.3. Novatein Production (Cradle-to-Gate)
Although the cradle-to-gate assessment of Novatein production has been reported previously [
18,
19], it is the primary feedstock for this product system, and the most important findings have been included here to provide context for its contribution to the life cycle of the overall system. The cradle-to-grave system assumed that low value by-products have no allocation of impacts from upstream processes, and thus blood has no impacts allocated to it until it is collected for drying (base case for the cradle-to-grave LCA). A discussion on the effect of different allocation methods for dividing farming impacts on the production of Novatein has previously been published [
18,
19]. The effects of alternate allocation scenarios on the overall cradle-to-grave system were considered in the sensitivity analysis for this study.
Blood meal production had four major unit processes that can contribute to GWP and NRPE use; farming, transport of animals, meat processing, and blood drying. Farming had by far the largest amount of GHG emissions associated with it due to methane production in livestock, and the proportion of this attributed to blood meal (rather than other animal products) depends on the allocation scenario chosen.
The allocation scenarios that were investigated included [
18]:
A simple mass-based allocation on fractions of a live animal.
An alternative mass-based allocation considered blood meal as a fraction of other animal products, which excluded wastes and losses (e.g., paunch, manure, and losses during slaughtering).
An economic allocation based on the price of blood meal as a fraction of the price of a carcass.
A mass-based allocation considering blood as a waste product, with no impacts upstream of blood collection attributed to it, but still including the energy used for blood drying to produce blood meal.
Figure 3 shows the contributions from different processes associated with the production of Novatein, under these different allocation assumptions. The production of other additives required to produce Novatein contributed 8.89 MJNRPE/kg Novatein and 0.4 kg CO
2eq/kg Novatein. Additionally, compounding additives with blood meal required to produce Novatein contributed 1.16 MJNRPE/kg Novatein and 0.084 CO
2eq/kg Novatein [
19].
It was concluded that blood meal production had much higher impacts than the rest of the cradle-to-gate system when the simple mass-based allocation was considered. However, for all the other cases, blood drying had the largest impact. Because of the large impact of blood production and drying, assumptions about impacts that were significant when looking solely at the processing of received blood meal had little effect in the overall cradle-to-gate. The final conclusion drawn was that Novatein production is justified if viewed from the point of turning waste into a product [
19], under the assumption that meat production will continue regardless of what blood meal is used for.
3.4. Plug Production, Use, and Disposal (Cradle-to-Grave)
When the entire product life cycle was assessed, it was found that none of the downstream processes were as significant as the cradle-to-gate portion (
Figure 4). The sum of all stages within the downstream processes emitted only 27% of the total NRPE and GWP of the full system. Within the downstream processes, injection molding and packaging showed the largest contributions to GWP impact, each comprising 31% of the total downstream impact, but only 8% of the full cradle-to-grave system (
Figure 4). The next highest impact was rendering, with conditioning and transport having the lowest impacts. Actual plug insertion and the short use-phase during carcass dressing were assumed to have a negligible impact in either category; they were therefore not included in the results.
Downstream from resin production, packaging used the next greatest amount of non-renewable primary energy (
Figure 4). However, this was still only about 15% of the NRPE used for the cradle-to-gate production of resin. Although injection molding is perceived as energy intensive, it required less NRPE than packaging. The plastic pellets are packaged in polyethylene bags and plugs are packaged inside cardboard boxes. If a human operator packs the plugs, there will be no additional electrical power required for this step of the operation, therefore PE bags produced 0.4 g CO
2eq per plug, and adding the emissions associated with cardboard, the net CO
2eq emission per plug for these two factors was 1.48 g CO
2eq.
The energy required by resin packaging is 0.00236 MJ and the combined energy requirements of the polyethylene film and cardboard was 0.0324 MJ; this provided a total NRPE input of 0.0348 MJ per plug. Packaging produced 31% of the gate-to-grave’s emissions and utilized 40% of the NRPE. Due to the packaging’s high emissions, whilst also having the highest NRPE use of any operation in the gate-to-grave section of the study, packing was included in the sensitivity analysis.
The impacts from transport were solely from the production and combustion of diesel fuel. It was assumed the trucks will operate efficiently, by organizing cargo loads to avoid empty trips (a 0.85 utilization factor was chosen for modeling purposes). The distance travelled is relatively small, and thus by combining the distance travelled with the high utilization factor, the impact from transportation was low. Resin and plug transportation required a total of 0.00171 MJ of NRPE per plug, while the GWP was estimated to be 1.19 × 10−4 kg CO2eq per plug. This equated to only 0.53% of the NRPE use and 0.68% of the GWP of the entire cradle-to-grave LCA.
Although transport has a low impact, it could also vary significantly. There are several factors that can contribute to increased impact. By decreasing non-related cargo (leaving empty space in the truck), the impact from transport can be doubled. Similarly, the utilization factor could be as low as 0.5 (instead of 0.85 assumed here). This would increase the impact by 70%. In addition, the transportation distance could increase greatly if the plugs were to be produced at different locations.
Injection molding required 2.23 MJ of electrical energy to produce 1 kg of injection molded product, a delivered energy requirement of 0.0224 MJ per plug. However, the production of each plug required a runner that had a weight equal to the plug itself. Half the energy was used during production of the runners and sprue, and half was used to produce the plug. Unfortunately, it is not possible to produce the plugs without the additional material required for the runners. Since the runners will be reground for reuse, all energy used during the process was attributed to the plug. The NRPE demand of each plug was 0.0223 MJ/plug, while the GWP was estimated at 1.48 g CO2eq/plug.
The generic injection molding model is based on commodity plastics such as PE. PE tends to have a melting point between 190–240 °C when injection molding, while Novatein is molded between 120–140 °C. From this, it can be assumed that the energy required for Novatein production was overestimated. However, it is unlikely that the delivered energy required to produce Novatein would be twice as high. Novatein requires a longer cycle time than most common polymers (40 s compared to 10 s for the PP plug), in other words, it is heated for longer.
Conditioning (
Figure 4) plays an important role in the production process. For the plugs to function correctly, they must be conditioned for a week at 23 °C and 50% relative humidity, decreasing the total moisture content to 8.4 wt%. It was assumed that the humidifier is constantly running at full capacity, using the maximum amount of electricity to function, resulting in 0.00730 MJ NRPE and 0.5 g CO
2eq/plug CO
2eq per plug. Conditioning was considered a minor impact, although if eliminated, it may simplify the production process substantially.
During use, packaging is removed as discarded. The emissions from the cardboard and PE film have been considered during the packaging phase, so no impacts from packaging are allocated to this phase. After the intestines are removed during dressing, the plugs are entered into the rendering system along with the entrails. The rendering process utilizes energy from the electrical grid and natural gas (burned to turn water into steam). The impact from disposal via rendering was estimated to be 0.02 MJ and 1.2 g CO
2eq per plug. Rendering contributed 25% and 23% to GWP and NRPE (
Figure 4). This means that the rendering operating was the third highest contributor and had a significant effect on the LCA. This highlights some of the unforeseen drawback of bioplastics, even if their apparent benefits seem obvious.
Only NRPE has been reported, not any renewable components of energy production. A significant portion of New Zealand’s electricity is obtained from renewable hydro-electric sources. If alternative assumptions about energy are made, for example 100% coal or gas-fired electricity production, calculated impacts for both NRPE and GWP would be higher. As most of the impacts for injection molding come from electricity usage, this would be the unit process most greatly affected by such assumptions, and was included in the sensitivity analysis.
It should also be mentioned that the properties of Novatein can be varied by adjusting the formulation. A reduced moisture Novatein formula could reduce or eliminate the need for the conditioning step, having the desired mechanical properties to function with the as molded moisture content. However, such formulation tweaks may require a larger amount of blood meal per kg extruded resin, and hence, while reducing impacts from conditioning, may increase impacts from resin production per plug.
3.5. Comparison to Polypropylne
Compared to PP plugs, Novatein resulted in 40% more CO2eq, but required only 70.2% of the NRPE. This discrepancy between the two impact categories comes partly due to approximately two thirds of the PP plug’s non-renewable energy use accounting for the fossil fuels used as feed stock, rather than for energy (electricity). It is interesting to note that this biopolymer resulted in higher GWP potential. It is often assumed that anything bio-based must inherently have a lower GWP, however extensive up and downstream processing is often required for bio-based materials, which drastically alters the situation.
In the modelled LCA, Novatein plugs had greater GWG emissions compared to PP, which highlights the implicit deficiency of an attributional LCA, already shown in the cradle-to-gate study of Novatein. Novatein is produced from an existing agricultural by-product, and it could be argued that no emissions should be attributed to the raw material (blood meal only). Including different allocation scenarios in a sensitivity analysis has shown considerable variance and was further explored in the current cradle-to-grave study.
Petrochemical plastics such as PP do offer advantages in terms of good mechanical properties for their weight and ease of processing, which enable efficient processing into a lightweight part. The biggest advantage of the Novatein plug, however, is the fact that it breaks down during rendering to become a non-toxic part of meat-and-bone meal, which is often used in pet food. When PP plugs go through rendering, they consume energy, but do not break down, and become pollutants in the meat-and-bone meal. This increases the toxicity of the meal, reducing the number of applications for this product. Although this factor does not directly come across in this LCA, it could be investigated further if the goal and scope of this study were broadened to include human or animal toxicity.
3.6. Sensitivity Analysis
The largest variances in this study were the allocation methods of farming impacts, how electricity is generated and distributed, and using cardboard boxes for packaging and formed the basis of the sensitivity analysis. Four different cases were considered and compared to PP and the base case (assuming blood is a waste product, but attributing the impact of drying):
Simple mass-based allocation for Novatein production.
Economic allocation for Novatein production.
Excluding cardboard from packaging.
Coal fired electricity.
The allocation methods for Novatein production were included because Novatein production had such an overwhelming impact on the cradle-to-gate impact categories, and it has been shown to be sensitive to allocation methods [
18]. Packaging also had a significant impact and was included in the analysis as producers can control their packaging choices. Lastly, electricity production varies significantly around the world and since the current process relies heavily on electricity, comparing the current situation with the worst case, was seen as a useful measure. The results for the sensitivity analysis are summarized in
Figure 5.
3.6.1. Allocation Methods for Blood Meal Production
Two allocation methods for farming impacts have been considered for comparison; these were using either a mass or economic allocation method, as well as treating blood as a low value by-product, which is therefore considered to have no impacts prior to blood drying. By allocating some impacts from farming to blood meal, the results from the cradle-to-gate do change quite significantly; therefore, the selected allocation method must be justified.
If blood production were to have impacts allocated from farming on a simple mass basis, where allocation is based on blood being a fraction of the live animal weight, it greatly increases the GWP and NRPE. The impact from Novatein production increased from 1.28 to 14.97 CO
2eq/kg Novatein, and NRPE use increased from 23.73 to 47.98 MJ/kg Novatein [
1]. However, a more suitable method is to apply mass-based allocation where waste and losses are excluded and blood is allocated as a fraction of all animal products, bringing the total impact to 3.76 CO
2eq/kg Novatein, and results in NRPE consumption dropping to 28.11 MJ/kg Novatein produced. These data represent the impacts of blood production, blood drying, and resin production. When implementing this advanced mass-based allocation, not only did the NRPE use increase by 18.5% for resin production, but the CO
2eq emissions increased by 193.0%. When these results were compared to the entire cradle-to-grave analyses, the total impact per plug increased to 0.0425 kg CO
2eq/plug, a 141% increase for a single plug. The NRPE increase was then 13.6% over the initial case.
If blood production were to have impacts allocated to it from the farming process on an economic basis, it would also increase the CO
2eq emissions and energy attributed to Novatein. The impact from Novatein production in the cradle-to-gate phase increased from 1.28 to 2.18 CO
2eq/kg Novatein, and NRPE use increased from 23.73 to 25.27 MJ/kg Novatein resin produced [
19]. NRPE use increased by 6.49% per plug, but again there was a large increase in the CO
2eq emissions at 32.6% over the cradle-to-gate. When these results are compared to the entire cradle-to-grave analyses, the total impact per plug increased to 0.0218 kg CO
2eq, a 23.8% increase for a single plug. The total NRPE increase was then 4.93% over the initial case.
None of the allocation scenarios examined change the technological scope of the system, or the total emissions and energy use of the NZ sheep and beef sector. Rather, they are different ways of attributing existing impacts between existing co-product streams.
3.6.2. Removing Cardboard
One of the unit operations that can be drastically altered in this LCA is the packaging phase, which would lead to a decrease in both CO
2eq emissions and energy use. Packaging required the highest amount of energy in the gate-to-grave phase (
Figure 4) and emitted the same amount of CO
2eq emissions as injection molding (
Figure 4). Therefore, a plausible way to decrease the impact of the plug’s life cycle is to remove unnecessary packaging.
Another possibility is to further reduce the impacts by maximizing the number of plugs in each bag before moisture sorption becomes a problem after the bags are opened. This means considering how long the plugs can be exposed to humid conditions before the moisture content will affect the physical properties too much, and then sizing the packaging so that the plugs are used before the plugs lose their effectiveness. This results in a reduction of the total amount of PE film that is required. However, for the purpose of this sensitivity, the plastic bags were kept the same size, and only the elimination of cardboard was inspected.
CO2eq emissions from cardboard boxes equated to 22.6% of the gate-to-grave, and 6.07% of the entire LCA. The NRPE used by the cardboard packaging equated to 26.1% of the gate-to-grave and 6.95% of the total LCA. By removing the use of cardboard packaging all together, not only did the NRPE use and CO2eq emissions decrease, but the impact from transport decreased as well. Although transport had a negligible emission of 0.68% CO2eq of the entire LCA, and only utilized 0.53% of the NRPE, if the locations of the resin and production plants changed drastically, removing cardboard could well help to decrease transport impacts significantly.
3.6.3. Coal-Based Electricity
It is useful to consider the data for delivered energy (
Table 2) in case the study needs to be adapted for different regions or countries that produce their electricity in different ways. For resin production, one of the sensitivity cases investigated was using 100% coal-based electricity to produce Novatein. For the delivered energy to be converted to cumulative energy demand (CED) for New Zealand electricity, it must be multiplied by a factor of 2.36, of which, 42.8% of the CED is non-renewable. On top of this, the CO
2eq emissions for the grid mix come to just 0.02797 kg CO
2eq/MJ CED. However, for coal-based electricity, the conversion is 2.77, and 100% is non-renewable. Additionally, the CO
2eq emission for coal-based electricity is 0.09788 CO
2eq/MJ CED. In other words, not only is the CED higher for coal-based electricity, but 100% of the CED is non-renewable, and the impact per MJ is also much higher.
For this sensitivity analysis, it was assumed that transport and packaging production are outside of the control of our system and would remain at their original values. However, resin production, injection molding, conditioning, and rendering were calculated using the NZ grid mix, and therefore their impacts would be altered by assumptions about electricity generation.
The most striking observation was that injection molding now has a larger NRPE use and GWP than the rest of the operations in the gate-to-grave (
Figure 6). Similarly, the conditioning chamber had a GWP increase of 178% and an NRPE use increase of 311%. This can be expected, as both operations had NRPE and GWP impacts based purely on electricity consumption (
Table 2); therefore, it makes sense that their values should increase by the exact the same ratio.
The resin production and rendering processes both included the use of natural gas for blood drying and to process the plugs at the end of their lives. The cradle-to-gate had an NRPE use increase of only 20.6%, with GWP increasing by just 39.5%. The rendering process had an NRPE increase of 22.3%, with GWP increasing by 38.9%. They did not show the same increase, as rendering was a unit operation on its own, and resin production included several process steps which do not utilize natural gas. As this sensitivity case only affected electrical energy supply, this is expected. Additionally, both processes had emissions that do not stem from energy use but are related to vapor mass streams that are released into the atmosphere.