An Environmental Analysis of Injection Molding

Alexandre Thiriez and Timothy Gutowski

Department of Mechanical Engineering
Massachusetts Institute of Technology
Cambridge, MA, USA.

Abstract—This environmental analysis of injection molding environmental analysis of the process. It provides a transparent
highlights a few important points. The choice of injection process model that includes all major steps involved in the
molding machine type (hydraulic, hybrid or all-electric) has a production of injection molded products and shows the
substantial impact on the specific energy consumption (SEC). dependency of injection molding on the most important process
The SEC values for hydraulic, hybrid and all-electric machines parameters. This paper presents a summary of our findings
analyzed are 19.0, 13.2 and 12.6 MJ/kg respectively (including along with detail on four major issues:
auxiliaries, compounding and the inefficiency of the electric grid).
For hydraulic and hybrid machines SEC seems to exhibit a 1. Relationship between energy consumption and
decreasing behavior with increasing throughput. This derives throughput.
from spreading fixed energy costs over more kilograms of
polymer as throughput increases. For all-electric machines SEC 2. Differences in environmental performance between
is constant with throughput. When the polymer production stage hydraulic and all-electric machines.
is included in the analysis, the energy consumption values 3. Role of secondary/subsidiary process in the energy
increase up to 100 MJ/kg. The overall injection molding energy accounting.
consumption in the U.S. in a yearly basis amounts to 2.06 x 108
GJ. This value is of similar magnitude to the overall U.S. energy 4. Environmental scale of injection molding.
consumption for sand casting, and to the entire electricity
production of some developed countries II. BACKGROUND
Keywords-Injection Molding; Hydraulic; Hybrid; All-electric; With regards to injection molding life cycle inventories
Specific Energy Consumption (SEC); Life Cycle Inventory (LCI). (LCI), much effort has gone into studying the production of
raw materials (polymers) as well as the product end-of-life
I. INTRODUCTION aspects, such as disassembly separation and recycling.
Amongst the researchers in this area, it is worth mentioning Ian
Plastic components are integral parts in electrical and Boustead, who developed a set of “eco-profiles,” or LCI’s, of
electronic (E&E) products. 8.5% of the plastic production is the most consumed polymers in the plastic industry. He also
dedicated to this market [1]. Although this number might seem created life cycle inventories for injection molded PVC and
small it is larger than the amount of plastic used for the injection molded polypropylene. The former LCI studied two
automotive industry (8%) [1]. In E&E products plastic can injection molding facilities in France that produce PVC fittings
represent from 3% of the total weight in medical equipment to for pipe drainage systems [2]. The latter LCI studied one
33% in small house appliances and 42% in toys [1]. The facility in the U.K. that produces 12 to 76 g polypropylene
majority of this plastic used for E&E products is injection components [3]. These studies are product specific, narrowing
molded in order to attain the specific geometric requirements. on one application and one set of processing parameters. In an
Injection molding involves melting polymer resin together with effort to obtain a range of values typical in injection molding,
additives and then injecting the melt into a mold. Once the and thus more breadth of data, this study incorporates
resin is solidified, the mold opens and the part is ejected. At measurements from machines processing different products
first glance, injection molding may appear to be a relatively and materials. It also provides a transparent outline of all the
benign process with respect to the environment due to its low sub-processes that make up the injection molding process
direct emission levels and apparently low energy consumption. together with their environmental performance.
However, when calculating the environmental cost of injection
molding one must also take into account the ancillary processes Other contributors to the field of injection molding include
and raw materials used in the process. Aside from the raw Mattis et al. 1996 and Boothroyd et al. 2002. Mattis et al. used
material production stage which has substantial emissions, the a 3-D solid modeling environment and numerical analysis to
main metric in the whole injection molding process is energy explore the influence of mold design, part design, and some
consumption. The large scale of the injection molding industry process parameters on the process efficiency [4]. Boothroyd et
makes the environmental impacts of this process especially al., whose goals were to develop design-for-manufacturing
critical. In other words, a small increase in the efficiency of the guidelines, developed a set of empirical equations predicting
process could lead to substantial savings for the environment. machine size and processing time of each stage in the injection
molding cycle [5].
This paper investigates injection molding from an
environmental standpoint, yielding a system-level
 If the reader is not familiar with injection molding Thus the choice of machine has a substantial impact on the
technology please refer to [6]. specific energy consumption1 (SEC), or energy consumption
per kilogram of polymer processed. More than 100 energy
III. MACHINE ENERGY CONSUMPTION measurements and calculations were examined from the three
types of machines. This analysis yields average SEC values for
The main division in injection molding machinery lies in hydraulic, hybrid and all-electric machines of 3.39, 1.67 and
how the drives in these machines are powered. The oldest and 1.46 MJ/kg respectively (without accounting for the efficiency
most common injection molding machine type is the hydraulic of the electric grid).
powered machine. This machine uses one or more hydraulic
pumps to power all of the machine’s motions. One can have a For hydraulic and hybrid machines SEC seems to exhibit a
pump for each drive, a centralized pump driving all motion, or decreasing behavior with increasing throughput, as portrayed in
a combination thereof. There are two obvious inefficiencies Fig. 2. This derives from spreading fixed energy costs over
with hydraulic machines. First, for most machines, pumps more kilograms of polymer as throughput increases. The
continue running even while the machine is idle, consuming power in a hydraulic and hybrid can be described as:
power that does not get used in production and thus wasting P = P0 + km&
power. Secondly, there is an intrinsic inefficiency in the where, (1)
architecture of the system. An electric pump transfers work to
the hydraulic circuit, which in turn transfers work to the P0 = fn(hydraulic pumps, computer, etc..)
mechanical components. Each transfer of work entails k = extra SEC to process the polymer
inefficiencies. Why not eliminate one of these transfers? This
is where all-electric powered injection molding machines come where Po is the fixed power requirement (power required when
into place. As their name indicates, these machines use servo the machine is on, but not processing any polymer), m & is the
motors to power each of the mechanical drives. Basically, one throughput or process rate, and k is a processing constant. In
servo motor runs the rotation of the screw, another moves the terms of SEC, this formula can be expressed as:
screw along the injection axis, and a third moves a toggle
clamp to close the mold. Aside from the above mentioned P E P (2)
= = SEC = 0 + k
main servos, there might be others that run secondary m& m m&
functions. These machines exhibit superior efficiency on As throughput increases, SEC approaches the constant k as
average, but are not applicable for high clamping force observed in Fig. 2.
applications due to the instabilities in the toggle clamp
configuration. This is where the hybrid powered machines All-electrics on the other hand have very low fixed energy
come into place. A hybrid machine uses both servo motors and costs (ex: running the computers), and their SEC stays constant
hydraulic pumps. The most common configuration is using the as throughput increases, as portrayed in Fig. 3.
hydraulic pump for clamping and servo motors for screw
movement. These machines sacrifice some of the all-electric
efficiency for the precision of hydraulic clamps. Fig. 1 portrays
the power requirement for a hybrid and an all-electric machine
both running the same part with a cycle time of 14 seconds.
Simple inspection reveals substantial energy savings from
using all-electric over hybrid technology. The reader must note
that the curve for a hydraulic machine would be even higher Name = Magna MM550 w ith and w ithout e-drive.
Shot Weight = 0.68 kg for the PS shots. Unknow n kor others.
than that of the hybrid.
2.6
550 hybrid press (PP)
Ton
120 2.4 550 hybrid press (HDPE)
Buildup
SEC (MJ/kg)

Cool 550 hybrid press (PS)

Clamp open- 2.2 550 hydraulic Press (PS)
100
close
Plasticize 2.0
Power Required (kW)

Inject high
80
1.8

60 1.6

Inject low 1.4

40
40 90 140 190
Throughput (kg/hr)
20

0
Figure 2. SEC vs. throughput for a Magna MM550, hydraulic and
hybrid. There is no inclusion of the efficiency of the electric grid.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Tim e (seconds) Source: [8, 9].
MM 550 Hybrid NT 440 All-Electric
1
Here we use a common expression, while in fact energy is not consumed but
Figure 1. - Energy consumed in the injection molding cycle of a hybrid transformed. A more rigorous but less understood definition for SEC would
(electric screw drive) and an all-electric machine. Source: [7]. be “specific exergy consumed”.
 9 application. The polymer is then shipped to the injection
8
All-Electric - 85 tons molder which transforms the polymer into a finished product.
7
The injection molder might also add some additives in the
Hydraulic - 85 tons
process, such as coloring. After being injection molded and
6
packaged, the product is ready to be used by the consumer (and
SEC (MJ/kg)

5 Material: Polypropylene (PP) eventually disposed). The scope of this analysis is “cradle to
4 factory gate” with the exclusion of packaging. Thus it
3 encompasses everything from the creation of the raw materials
2 for polymer production to the injection molding of the product.
1 The system boundaries are portrayed by the dashed square in
0 Fig. 4.
0 5 10 15 20
Throughput (kg/hr) Close to 100 sources were consulted in order to develop
this LCI. Most of these sources are not listed in the references
Figure 3. SEC vs. throughput for an all-electric and a hydraulic machine. section but can be obtained from [10]. The results of the LCI
There is no inclusion of the efficiency of the electric grid. Source: [8]. are summarized diagrammatically in Fig. 5. The reader must
note that with the exception of the polymer production stage
when energy data exhibited variation with type of polymer it
was averaged according to the relative amount of polymer
IV. SUMMMARIZED LIFE CYCLE INVENTORY injection molded in the U.S.
In order to develop a successful life cycle inventory (LCI) it
It is interesting to note how even though the energy
is first necessary to establish the boundaries of the system to be
consumption for injection molding machinery seems low, when
analyzed. In the case of injection molding, the overall process
other stages in the process are included the figure becomes
starts at the polymer production stage. This stage takes raw
substantial. Considering the energy consumption of all stages
materials from the earth and transforms them, with the addition
from the compounder to the injection molder (not including
of energy, into polymers. The raw polymer is then shipped in polymer production), hydraulic, hybrid and all-electric
bulk to the compounder which mixes it with additives in order
machines yield average values for SEC of 19.0, 13.2 and 12.6
to bestow the polymer with the required properties for its future MJ/kg respectively. These values take into account the energy

CRADLE Additives

Naphtha, Oil. Ancilliary Raw

Natural Gas Materials
Compounder

Internal Transport Drying Emissions

Polymer to air,
Thermoplastic Production Delivery Extrusion Pelletizing water &
(Boustead) land
Building (lights,heating, ect..)

Emissions to
air, water, & Polymer Delivery
land
Injection Molder

Energy Production Industry Internal Transport Drying

Emissions
Injection Molding to air,
Emissions to air, water, & land water &
land
Scrap
Anciliary Raw
Materials Building (lights,heating, ect..)

Packaging

Note to Reader: FACTORY GATE

= Focus of this Analysis 1 kg of Injection Molded Polymer

= Also included in the Paper

Service Period

Waste Management
Figure 4. Injection Molding System Boundaries
 ENERGY CONSUMPTION BY STAGE in MJ/kg of shot
Thermoplastic Production
Generic by Amount Extras
HDPE LLDPE LDPE PP PVC PS Consumed Inj. Molded PC PET
avg 89.8 79.7 73.1 83.0 59.2 87.2 81.2 74.6 95.7 78.8
low 77.9 79.7 64.6 64.0 52.4 70.8 69.7 62.8 78.2 59.4
high 111.5 79.7 92.0 111.5 79.5 118.0 102.7 97.6 117.4 96.0

avg 0.19
Polymer Delivery low 0.12
high 0.24

Compounder
Internal Building (lights,
Transport Drying Extrusion Pelletizing heating, ect..)
avg 0.09 0.70 3.57 0.16 0.99
low ----- 0.30 1.82 0.06 -----
high ----- 1.62 5.00 0.31 -----

Subtotal avg 5.51

low 3.25
high 8.01

avg 0.19
Polymer Delivery low 0.12
high 0.24

Injection Molder
Internal Injection Molding Building (lights,
Drying Scrap (Granulating)
Transport (look below) heating, ect..)
avg 0.04 0.70 0.05 0.99
low ----- 0.30 0.03 -----
high ----- 1.62 0.12 -----

Injection Molding - Choose One

Hydraulic Hybrid All-Electric
avg 11.29 5.56 4.89
low 3.99 3.11 1.80
high 69.79 8.45 15.29

Subtotal avg 13.08 7.35 6.68

low 5.35 4.47 3.17
high 72.57 11.22 18.06

TOTAL w/ Hydraulic Hybrid All-Electric

Generic Inj. avg 93.60 87.87 87.20
Molded low 71.65 70.77 69.46
Polymer high 178.68 117.34 124.18

avg 18.97 13.24 12.57

TOTAL w/o
low 8.84 7.96 6.66
Polymer Prod
high 81.04 19.70 26.54

Notes Drying - the values presented assume no knowledge of the materials' hygroscopia. In order words, they are
averages between hygroscopic and non-hygroscopic values. For hygroscopic materials such as PC and PET
additional drying energy is needed (0.65 MJ/kg in the case of PC and 0.52 MJ/kg in the case of PET)
Pelletizing - in the case of pelletizing an extra 0.3 MJ/kg is needed for PP
Granulating - a scarp rate of 10 % is assumed
Figure 5 - Overall System Diagram. The values above account for the efficiency of the electric grid. Multiple sources. Refer to [10] for an extended bibliography.
burden associated with producing the electricity to power the
manufacturing processes2. When the polymer production stage Inj. Molded - Million kg/yr
is included in the scope of the LCI, the energy consumption U.S. Only Global
values increase up to 100 MJ/kg. In the whole LCI, producing 6 Main Thermoplastics 5,571 23,899
All Plastics 12,031 38,961
the polymer has the greatest impact on the environment. After
Table 2. Injection molded polymer totals in kg/year. The subdivision
the polymer production, injection molding machinery and 6 main thermoplastics refers to HDPE, LDPE, LLDPE, PP, PS and
extrusion have the greatest impact. PVC. The complete calculation can be found at [1]. The sources used
are [16, 17, 18].
With regards to emissions, the majority of emissions come
from the polymer production stage. Please refer to [11] if Compounder and U.S. Global
interested in these emissions. In the rest of the LCI, emissions Injection Molder GJ/year GJ/year
can divided into: energy related emissions and processing 6 Main Thermoplastics 9.34E+07 4.01E+08
emissions. Energy related emissions refers to those emissions All Plastics 2.06E+08 6.68E+08
originated from the generation of electricity necessary to run Table 3. Total energy used in Injection Molding. The subdivision 6
the processes. Table 1 presents energy related emissions for main thermoplastics refers to HDPE, LDPE, LLDPE, PP, PS and PVC.
The complete calculation can be found at [1].
the compounder and the injection molder.
Processing emissions arise at the polymer processing sites.
These kinds of emissions are small compared to energy related
ones. For instance extruding polypropylene generates 0.185 g this information, we assume that 70% of the injection molding
of volatile organic compounds (VOC’s), 0.030 g of particulate machines are hydraulic, 15% are hybrids and 15% are all-
matter, 0.0099 g of ketones, 0.0022 g of aldehydes, , and electric. Table 3 shows the results of the U.S. and global
0.0018 g of organic acids per kg of polymer extruded [12]. energy estimate.
As can be observed, the overall injection molding energy
V. ENVIRONMENTAL SIGNIFICANCE consumption in the U.S. in a yearly basis amounts to 2.06 x 108
When compared to other conventional manufacturing GJ. This value includes all steps in the LCI, except polymer
processes, injection molding appears to be on the same order of production. Including polymer production would increase this
magnitude in terms of energy consumption. For instance, number by an order of magnitude. This value (2.06 x 108 GJ)
processes such as sand and die casting have similar energy is of similar magnitude to the overall U.S. energy consumption
requirements (11-15 MJ/kg) [13, 14]. However, when for sand casting (1.62 x 108- 2.28 x 108 GJ, [14]). For the
compared to processes used in the semi-conductor industry, reader to comprehend the scale of the U.S. injection molding
such as chemical vapor deposition and atomic layer deposition, energy consumption, Table 4 provides values of the entire
the impact of injection molding seems insignificant. This is far electricity production of several countries. Without accounting
from the truth, though. In order to understand the real impact for the electric grid, the overall injection molding energy
of a manufacturing system one has to understand how consumption in the U.S. amounts to 6.19 x 107 GJ/year3. This
widespread its use is in the economy. Injection molding is one value can be compared with the values in Table 4.
of the predominant manufacturing processes, and its use is It seems imperative for industry to keep improving the
increasing daily in growing economies like China and India. efficiency of the process, since small savings anywhere in the
Table 2 presents an estimate of the current quantities of LCI can lead to tremendous energy savings on a national scale.
polymer injection molded in the U.S. and in the world. With This seems an intelligent move in a time of raising energy
these values and distribution of the different machine types, the prices.
total energy spent in injection molding can be estimated.
According to Snyder, in 2002 29% of the machines sold in ACKNOWLEDGMENT
the U.S. were electric based rather than hydraulic [15]. With This research was supported by the National Science
Foundation Award DMI 0323426.

Energy Related Emissions

SEC
Stage CO2 SO2 NOx CH4 Hg
(MJ/kg)
g g g g mg
Compounder 5.51 284.25 1.26 0.51 10.32 0.01
Injection Modler
Hydraulic 13.08 674.82 2.98 1.22 24.49 0.01
Hybird 7.35 379.33 1.68 0.68 13.77 0.01
All-Eletric 6.68 344.57 1.52 0.62 12.50 0.01
Table 1 - Energy-related air emissions for the “compounder” stage and the
“injection molder” stage. Multiple sources. Refer to [10] for calculation

2 3
The electric grid in the U.S. is 30% efficient. Visit [10] for more details. Equivalent to electricity consumption.
 Annual Electricity Production
Within 1 Order of Magnitude
Smaller than U.S. Injection Molding Totals
to U.S. Injection Molding Totals
Country GJ/year Country GJ/year Country GJ/year Country GJ/year
Afghanistan 1.71E+06 Jordan 2.55E+07 Austria 2.19E+08 Iran 4.47E+08
Guatemala 2.22E+07 Nicaragua 8.41E+06 Belgium 2.68E+08 Netherlands 3.18E+08
Honduras 1.37E+07 Nigeria 6.25E+07 Bulgaria 1.49E+08 New Zealand 1.39E+08
Iceland 2.84E+07 Panama 1.78E+07 Czech Rep. 2.52E+08 Poland 4.87E+08
Jamaica 2.26E+07 Slovenia 4.92E+07 Denmark 1.27E+08 Portugal 1.59E+08
Finland 2.56E+08 Saudi Arabia 4.65E+08
Greece 1.80E+08 Switzerland 2.47E+08
Indonesia 3.46E+08 UAE 1.36E+08
Hungary 1.24E+08 Venezuela 3.15E+08
Table 4 - Selected countries with smaller or similar order of magnitude electricity production to the amount of energy spent injection molding (compounder
+ injection molder) in the U.S. Source: [19].

processing," Journal of Air & Waste Management Association, 49

REFERENCES (1999): 49-56.
[1] M. M. Fisher, F. E. Mark, T. Kingsbury, J. Vehlow, and T. Yamawaki, [13] S. Dalquist and T. Gutowski, Life cycle analysis of conventional
“Energy recovery in the sustainable recycling of plastics from end-of- manufacturing techniques: die casting, Cambridge, USA: 2004.
life electrical and electronic products,” 2005 IEEE International Unpublished. Available at: <http://web.mit.edu/ebm/publications.htm>.
Symposium on Electronics and the Environment, May 2005. [14] S. Dalquist and T. Gutowski, “Life cycle analysis of conventional
[2] I. Boustead, Eco-profiles of the European plastics industry: PVC manufacturing techniques: sand casting,” ASME International
conversion processes, Brussels: APME, 2002. Visited: 25 Feb. 2005 Mechanical Engineering Congress and RD&D Expo, California, USA,
<http://www.apme.org/dashboard/business_layer/template.asp?url=http:/ Nov. 2004. Available at: <http://web.mit.edu/ebm/publications.htm>.
/www.apme.org/media/public_documents/20021009_123742/EcoProfile [15] M.R. Snyder, "Electric injection machines inspire respect, loyalty,"
_PVC_conversion_Oct2002.pdf> Plastics Machinery&Auxiliaries, Mar.-Apr. 2002. Visited: 18 Jan. 2005
[3] I. Boustead, Eco-profiles of the European plastics industry: conversion <http://www.pma-magazine.com/articles/2002/March/02>.
processes for polyolefins, Brussels: APME, 2003. Visited: 25 Feb. 2005 [16] J.A. Brydson, Plastics Materials, 7th ed. Oxford, Great
<http://www.apme.org/dashboard/business_layer/template.asp?url=http:/ Britain: Butterworth-Heinemann, 1999.
/www.apme.org/media/public_documents/20040610_153828/Polyolefin [17] Probe Economics, Inc, SPI Economic Report 2000,
sConversionReport_Nov2003.pdf> Washington, DC: The Society of the Plastics Industry, Inc., 2000.
[4] J. Mattis, P. Sheng, W. DiScipio, and K. Leong, A framework for [18] I.I. Rubin, Injection Molding Theory and Practice, New
analyzing energy efficient injection-molding die design, Technical York: John Wiley & Sons, 1972.
Report (CSM Report-96-09), California: University of California,
[19] Energy Information Administration, International energy annual 2002,
Berkeley, 1996.
Washington: Department of Energy, 2002. Visited: 23 Feb. 2005
[5] G. Boothroyd, P. Dewhurst, and W. Knight, Product design for <http://www.eia.doe.gov/iea/contents.html>
manufacture and assembly, New York: Marcel Dekker, 2002.
[6] Dominick V. Rosato, Donald V. Rosato, and M.G. Rosato, Injection
NOTE: for the LCI results not all references that were used in the
molding handbook, 3rd ed. Springer - Verlag, 2000. Visited: 05 Aug. calculations were listed. If the reader desires to obtain the references please
2005 refer to [10].
<http://www.knovel.com/knovel2/Toc.jsp?SpaceID=0&BookID=357&
VerticalID=0&NodeID=1884960566&ClickedNodeID=1884960566&Fr
omSearch=true&Action=expand&FilterMode=false&View=4&Collapse
All=false#node1884960566>.
[7] Data file: cm test graphs. Sent by Mark Elsass (Cincinnati Milacron
Supervisor for Technical Service). Received: 18 Apr. 2005.
[8] Ferromatik Milacron, “A communications update for ferromatik
milacron sales professionals,” North America: 2001, unpublished. Sent
by Mark Elsass (Cincinnati Milacron Supervisor for Technical Service).
Received on: Apr. 2005.
[9] The previous reference was received as a word document. In it was
imbedded an EXCEL file.
[10] A.Thiriez, “An Environmental Analysis of Injection Molding”, Masters
Thesis at Massachussetts Institute of Technology, Cambridge, MA,
USA: 2006.
[11] I. Boustead, Eco-profiles of the european plastics industry, Brussels:
APME, 2002-2003 Visited: 20 Mar. 2005 <http://www.apme.org/>.
[12] K. Adams, J. Bankston, A. Barlow, M.W. Holdren, J. Meyer, and V.J.
Marchesani, "Development of emission factors for polypropylene