The Plastic Redesign Project
Overview Coalition Members Contact Us Publications Links Join Members Only Area

October 22, 1999

Designing Plastic Bottles for Recyclability: a Case Study




Design for recyclability -- as it is applied to plastic bottles -- asks packaging engineers to consider how new bottle designs will impact the economics of recycling the package.

It does not demand that recycling considerations dominate over all other considerations. Rather, design for recyclability asks that the reasonable needs of those who handle the package after it is discarded by the consumer are balanced with the needs for the package to perform its intended function of delivering a healthy product to the consumer. That is to say, for the marketplace to work properly, so-called "external" costs must be accounted for, even though they are not recorded on the books of packagers, because others in our economy must bear that cost.

Some examples of bottles designed for recyclability are those with labels that are easily removed in a plastics processors' wash water, and, caps that are easily removed from their bottle or that are made from a resin that is compatible with the bottle resin when remolded.

This does not mean, it should be emphasized, that packaging engineers are pitted against recyclers. For innovation more often than not makes it possible for both sides to win. In the past, the dynamic nature of the packaging industry has often responded to recyclers' needs with adaptations of new bottles that avoid adding more back-end processing costs or carry a cleaner and, therefore, higher value resin to market. Examples are the elimination of base cups on soda bottles and the elimination of multiple incompatible resin layers in ketchup bottles.

The Plastic Redesign Project, with funding from the Environmental Protection Agency and the Wisconsin Recycling Market Development Board, was formed to continue finding win/win solutions between the packaging and recycling industries. It does this by asking package designers to apply their ingenuity to improving bottle performance in ways that do not erode the current fragile economics of plastics recycling and maintain the high value inherent in clean resin.

In 1995, the Plastic Redesign Project brought together cities with recycling programs and packaging companies and developed 13 consensus redesign recommendations that both parties agreed were reasonable. Today, 32 states are participating either individually or through regional associations in the Project, and they are now working on updating those recommendations for new design issues that have evolved since 1995. Many issues have arisen such as heat transfer labels on soft drink bottles and the introduction of a new clear resin -- abbreviated P.E.N. -- for carbonated beverages and hot-fill applications that can be confused with traditional P.E.T. bottles.

Another such issue -- and the one that is the subject of this working paper -- is the use of pigmented opaque milk bottles in place of unpigmented and translucent (natural) plastic milk bottles which, recyclers say, drastically reduces the bottle's scrap value. Ten dairies in different parts of the country are thought to presently use either white or yellow pigmentation in their plastic milk bottles.

According to several of the dairies' public statements, the reason for the switch to pigmented opaque bottles is to protect the milk's flavor from off tastes caused by light, and one dairy has cited vitamin degradation, buttressed by a number of academic reports [1].

There is debate, however, over the validity of these claims in the milk we actually drink. Mr. Robert Byrne of the International Dairy Foods Association has stated that there is no strong evidence that light in a dairy case affects most vitamins in milk and concludes that pigmenting is largely a marketing tool [2,3]. Modern Plastics magazine reports that "[t]he common thread is that each jug design reflects a push by dairies to distinguish their packaging so as to boost flagging milk sales."[4] Solvay Polymers, the major supplier of resin for the dairy industry, states "in practical experience in the production, distribution and consumer cycle, [vitamin A and B2] losses are negligible" and adds that due to added molding difficulties it "does not recommend in-plant container pigmenting until all other alternatives have been fully explored."[4] Plastics in the Environment magazine also reports that while "[p]rotection of flavour and vitamins from light degradation are the reasons generally cited for the design changes[, m]ost market analysts believe that this issue is a red herring and that opaque bottles are essentially a marketing ploy."[4a]

The purpose of this research paper is to, first, evaluate the claims of recyclers that pigmented milk bottles will adversely affect the economics of plastics recycling; and second, evaluate the claim that plastic milk bottles need to be pigmented in order to protect the product from damage by light in the field. This paper also explores alternative measures that might meet the needs of both parties.

Impacts from Pigmenting Plastic Milk Bottles on the Recycling Industry

Recyclers' are concerned because unpigmented jugs, which are the majority of recovered HDPE bottles, command a far higher price from end markets. Therefore, if the entire or substantial part of the dairy industry were to convert their currently unpigmented plastic milk bottles to pigmented bottles, recyclers would lose a significant portion of their revenues. In addition to the difference in revenues, there is the related issue of minimum volumes and economies of scale. Handling costs can easily eat up the small margins that exist in the scrap industry. Thus, shifting material from the high volume unpigmented stream to small different colored streams builds inefficiencies in sorting, baling and shipping that add up. Are these losses a major concern?

Plastic reclaimers typically make a profit when resin prices are high, but too often lose money when commodity resin prices are depressed. Because few reclaimers are large entities with substantial diversified resources to weather downturns, the present economics of plastics recycling are tenuous. They feel that the direction of future changes in plastics packaging needs to be towards simpler bottles with higher resale value to insure the survival of plastics recycling. Of course, any losses incurred by the intermediate processors will be passed back to the communities which collect plastic milk jugs.

The nature of their concern should be made clear. It is not that pigmented milk bottles are not recyclable: with one exception, they most certainly are -- technically. Rather, eliminating high-value unpigmented bottles from the recycling stream is said to significantly undermine the economics of plastics recycling over the long term. As reported in the Wall Street Journal, "[t]he problem for community recyclers

"is that milk bottles are their cash cow, and they need high volume to make ends meet. Even without the price difference, industry economics are tenuous right now. Because of the moderate price of oil -- the major component of plastic -- the gap between the prices manufacturers pay for new plastic and for the traditionally cheaper recycled is tiny. So every penny counts." [4b]

What do the numbers show? The reason why unpigmented resin sells for more than pigmented is that unpigmented resin can be used by a wider range of markets and can be readily modified with pigments to virtually any color.

Historic market prices bear this out. The preceding table shows the average price for unpigmented vs. pigmented HDPE bottle resin, the resin used to make milk jugs, paid by end users, as compiled from surveys by Recycling Times between 1994 and the present [5]. The average price for pigmented HDPE was only 57.2% of the average price paid for unpigmented HDPE. Not reflected in average price calculations is the fact that, during times when end markets collapsed, it was impossible to find any market willing to buy pigmented HDPE. That is to say, this 57% discount is a conservative estimate of future price impacts.

Percent Difference Between Unpigmented

and Pigmented Prices for HDPE Bottle Resin

Unpigmented HDPE

Pigmented HDPE





Jan 7.00 3.80 3.20 54.3%
Apr 7.79 4.40 3.39 56.5%
Jul 8.50 4.45 4.05 52.4%
Oct 11.40 5.79 5.61 50.8%
1995 Jan 20.38 10.50 9.88 51.5%
Apr 28.08 15.28 12.80 54.4%
Jul 19.43 10.93 8.50 56.3%
Oct 14.43 9.29 5.14 64.4%
1996 Jan 10.90 6.61 4.29 60.6%
Apr 8.14 5.14 3.00 63.1%
Jul 10.50 6.36 4.14 60.6%
Oct 12.30 7.89 4.41 64.1%
1997 Jan 14.30 8.32 5.98 58.2%
Apr 16.40 9.07 7.33 55.3%
Jul 16.40 9.64 6.76 58.8%
Oct 16.70 9.07 7.63 54.3%
  Average 13.92 7.91 6.01 57.2%


According to American Plastics Council (APC) surveys, of the 171,300 metric tons of postconsumer HDPE bottles recycled in the United States in 1996, 61%, or 104,491 metric tons, were unpigmented, making unpigmented bottles the largest fraction of the HDPE recycling stream.[6] In addition, APC calculates that 67% of HDPE unpigmented resin sales for bottles, or 70,010 metric tons, were to the dairy industry -- as opposed to other industries such as packagers of mineral water [7]. By applying these breakdowns to the total amount of HDPE bottles recycled, the following table shows how overall revenue from sales of recycled HDPE resin could change if all dairies changed to pigmented plastic bottles.

Difference in Value of HDPE Bottle Stream of U.S.

With and Without Pigmented Milk Bottles

  Without Pigmented Milk Bottles With All Pigmented Milk Bottles

(000 mt)



Total Sales Quantity

(000 mt)



Total Sales
Unpigmented 111.1 13.92 $34,085,124 0 13.92 $0
Pigmented 70.7 7.91 $12,325,583 181.8 7.91 $31,694,358
Totals     $46,410,708     $31,694,358

Thus, in the hypothetical case where every dairy shifts from natural to pigmented plastic bottles, the total value of the recycled HDPE stream will fall from $46.4 million to $31.7 million, or by $14,716,350, a 31.7% decline, based upon the average of the past four years' prices. Losses in scale efficiencies as the volume of natural HDPE declines are more difficult to generalize, but are not insignificant. Ultimately, these losses can be expected to be passed back to local recycling programs.

There is also another area where losses may be incurred if there is a substantial shift to pigmentation of natural HDPE, which is a distinct form of high density known as homopolymer. In those parts of the country where some dairies pigment their milk bottles, recyclers are sorting the pigmented jugs with the rest of the pigmented stream. Because the rest of that stream is a different form of high density known as copolymer, at some point the level of homopolymer in the copolymer will cause problems.

Specifically, copolymer makes a bottle that is more resistant to stress cracking than homopolymer. If too much homopolymer milk jugs are mixed in with the copolymer HDPE, it may no longer be possible to mold bottles from the contaminated pigmented stream. In this eventuality, the mixed resin would be downgraded to sheet or profile applications which pay several cents per pound less than higher grade bottle uses.

The Impact of Light on Milk

While the dairy industry has not previously expressed strong concern over detrimental light effects on milk, researchers in the dairy industry have. They have been concerned that exposure to light adversely affects the taste of and nutrients in milk. With sufficient exposures, laboratory tests do demonstrate that light can affect the milk. Unfortunately, none of the research that we have been able to locate reaches the issue of whether most of the milk actually purchased by consumers -- which is the real issue -- is noticeably affected.

Milk can be affected by light in two distinct ways. First, certain nutrients, such as vitamins A, B2, C and D, contained in or added to milk can be degraded by exposure to light. Second, milk can develop an off-flavor from exposure to light. These potential adverse effects are well documented and governed by a number of factors such as light intensity, light wavelength, distance from light source, exposure time, temperature, and method of packaging.

The question this paper seeks to answer is whether milk packaged in unpigmented translucent HDPE bottles is significantly affected by light during the time and under the conditions to which it is actually exposed. If it is significantly affected, this paper addresses possible alternatives to pigmenting the bottle in order to block harmful light.

Laboratory Tests

Before reaching the question of what happens in the real world, as noted previously, laboratory tests have shown that milk packaged in unpigmented HDPE bottles exhibits nutrient losses and off-flavors after sufficient exposure to fluorescent lighting [8,9,10,11]. Remember when reviewing these lab tests that, by themselves, they do not show that the exposure times at which degradation was observed mimics the light to which a milk bottle is exposed in the field.

Nutrients. The exposure time necessary to produce nutrient degradation varies depending on the nutrient and the intensity of the light source. Under laboratory conditions, significant degradation is found in vitamin A and riboflavin (vitamin B2) levels after less than 12 hours of continuing exposure to fluorescent light of the intensity found in display cases and at distances approximating the bottles closest to the fluorescent light source. Vitamin D is found to be relatively insensitive, and, while vitamin C is significantly affected, milk is not considered a significant source of vitamin C.

Vitamin A. There have been two separate findings with respect to vitamin A. Vitamin A found as a natural component of whole milk is not significantly affected by fluorescent light [9,12,13]. However, synthetic vitamin A that is added to low fat milk has been shown to be light sensitive. Laboratory tests vary, but overall, prolonged exposure to fluorescent light produced large losses in synthetic vitamin A of 90% over a 24 hour period under high-intensity fluorescent lighting. Under more moderate light conditions and shorter exposure times of 4 to 6 hours, smaller, but significant, losses were found in the vitamin A content of low fat milks packaged in unpigmented plastic bottles [10]. This shorter exposure time may be more realistic for the majority of milk sold from large supermarket dairy cases having high milk turnover.

Riboflavin (B2). Riboflavin has also been shown to be sensitive to exposure to fluorescent light. Laboratory tests of exposure of milk packaged in one gallon unpigmented HDPE bottles to fluorescent light have shown the effect on riboflavin to be less severe than that of vitamin A. Twelve to 18% losses in riboflavin were reported in milk after 12 hours of exposure to fluorescent light at average dairy case intensity [9]. These results are consistent with other studies at similar levels of light exposure [10].

Vitamin C. Vitamin C is extremely sensitive to light-induced degradation. This fact, however, has not created much concern among dairy researchers because milk in not a significant dietary source of vitamin C [10]. In addition, two researchers stated in phone conversations that there are so many sources of vitamin C in the American diet that adding vitamin C to milk would serve little nutritional purpose [11].

Vitamin D. Only one publication addressed vitamin D degradation. When milk with added vitamin D was exposed to light, vitamin D losses reached only 5% after two days [10]. These laboratory results are much lower magnitude than those for vitamin A and riboflavin.

Taste. Off-flavors, sometimes described as burnt cabbage taste, are thought to occur when riboflavin is "excited" by light energy, especially in the violet and blue portion of the visible light spectrum (i.e. 400 - 500 nanometers). This serves as a catalyst for the amino acid methionine to oxidize to methional altering the flavor [9]. Off-flavors were detected by dairy experts -- but not necessarily consumers -- in less exposure time than nutrient degradation, generally after 6-12 hours exposure. In comparison, milk packaged in paperboard did not exhibit light-induced off-flavors to any significant degree [8,9,10,12,13,14,15].

Other tests attempted to determine whether consumers could detect off-flavors deliberately created in the laboratory, without reference to whether those conditions exist in the field. The National Dairy Council states that "[i]t has been shown that consumers cannot distinguish an off-flavor in milk unless they have the opportunity to compare different milk samples" [10]. When given a reference point and a glass of milk that was deliberately exposed to create off-flavors, consumers preferred the glass without off-flavors. A taste comparison by 2,000 Vermont consumers showed that 73 percent preferred non-exposed milk to light-exposed milk [8].

Field Analysis

There have also been several attempts to find out whether the milk that is actually purchased by consumers in typical natural HDPE bottles shows noticeable deterioration in flavor and nutrient content in a significant number of instances. Unfortunately, while the tests detected impacts in some areas, none of the field oriented tests really answered this key question.

Nutrients. One field study, performed by the University of California-Davis to look at riboflavin impacts, randomly selected and purchased milk from actual supermarket dairy cases. Test results showed little difference in riboflavin content between milk packaged in unpigmented HDPE and milk packaged in paperboard [15]. The difference between this field result and laboratory results may be due to the fact that most milk is purchased in supermarkets which have high traffic and product turnover. Consequently, bottles do not remain on the display shelf for long periods of time, and, while in the display case, they move from the back of the shelf to the front as customers remove the front bottles for purchase. Light exposure may only be significant when the bottle sits on the front position on the shelf, not behind. Unfortunately this study did not test for vitamin A which has been shown in laboratory tests to be more sensitive than riboflavin to light degradation.

Flavor. On the other hand, the Davis study did find light-induced off-flavors. But since this was determined by a panel of dairy experts, it is not known whether the off-flavor would have been detected by the typical consumer. Among dairy experts, however, 45% of the milk samples packaged in natural HDPE bottles were found to exhibit light-induced off-flavor compared to 4.1% of the milk samples packaged in paperboard which is relatively opaque.


In summary, there is a body of laboratory research that concludes that milk packaged in natural HDPE bottles will exhibit nutrient degradation and light-induced off-flavors after exposure to dairy case lighting for sufficient periods of time. Thus, the potential exists for milk sold at retail to be negatively affected by dairy case lighting and no research was found that refutes this potential.

However, field studies did not show riboflavin degradation to be significant in the field given the short milk turnover and stock rotation at a typical supermarket dairy case. Field effects on vitamin A, which is more sensitive, were not recorded. Light-induced off-flavors, on the other hand, have been detected in significant portions of the commercial milk market by experts, but whether or not these off-flavors are detectable to the typical consumer is unclear.

According to Rob Byrne of the Washington-based International Dairy Foods Association, quoted earlier, there is no strong evidence that the light in the dairy case affects most vitamins in milk packaged in plastic or paper [2]. In a phone conversation with Mr. Byrne, he stated that light does affect milk nutrients and flavor but that as far as milk purchased by consumers in unpigmented HDPE bottles, there is not a significant light effect on nutrients. He also stated that most consumers do not recognize light-induced off-flavor except in extreme cases and concluded that pigmenting is largely a marketing tool.

On the other hand, Mr. Byrne said that if a dairy is adding vitamin C, which is extremely light sensitive, pigmenting the bottle may be justified [3]. This statement is validated in conversations with researchers in the dairy field, but typically they have discounted the nutritional need to add vitamin C to milk because vitamin C can be obtained naturally and as an additive in a variety of foods in the American diet [11].

Options for Minimizing Potential Light Effects

Although the technical case has not yet been made to support the need for pigmenting plastic milk bottles, if flavor and nutrient degradation is perceived or is later shown to be a significant problem for milk consumers, there are options available for minimizing light effects. One option is to package milk in opaque containers such as paperboard or pigmented HDPE bottles. A second option, changing lighting practices, does not necessitate pigmenting bottles.

Hopefully, a spirited dialog will uncover other options, as well. But, in the meantime, we have compared the relative costs of the two options that have been proposed.

Pigmenting bottles using titanium dioxide blended in with the HDPE resin has been tested in the laboratory and shown to significantly reduce detrimental light effects. It also is approved by the FDA for food application packaging. A recent examination of bottle costs showed that pigmentating using this method currently costs dairies an additional 0.6¢ to 2.5¢ per one gallon HDPE bottle with the majority of dairies paying close to 1¢ [9,11,16]. With approximately 4,902 million gallons of milk sold in unpigmented HDPE bottles in 1995, shifting the entire diary industry to pigmented HDPE bottles would entail an additional cost of 1¢ on each of 5,611 million gallon and half-gallon bottles used to package that milk [17]. This translates into an additional industry wide cost of $56,611,000.

Pigmenting all HDPE milk bottles has additional negative affects of $14,716,350 on the plastics recycling industry as described earlier in this paper. The total added cost to society, therefore, is $56,611,000 plus $14,716,350, or $71,327,350.

A less expensive option for the dairy industry, and without adverse consequences for recyclers, is to have retailers of milk decrease the intensity of flourescent lights used in milk display cases and either purchase yellow "bug" bulbs or cover white bulbs with a yellow light filter. The yellow filter blocks much of the light in the 400nm to 500nm light range that is of greatest concern in causing flavor and nutrient degradation. This method of decreasing milk exposure to harmful light has been shown to be effective under laboratory testing [14]. A sampling of commercial lighting stores indicates that the incremental cost of a yellow vs. cool white 40 watt flourescent bulb is $10.20 and the bulbs typically last more than two years. A yellow shield costs $5.35 and typically lasts the life of two or more bulbs.

Using the milk sales data above and throughputs of a typical grocery store dairy case that uses two 40-watt fluorescent bulbs [18], the industry-wide cost of adding yellow shields over these bulbs totals $95,000. Although broad assumptions were required in order to arrive at this figure, the result illustrates the magnitude of the difference between dairy case lighting changes at $95,000 and pigmenting all HDPE milk bottles, along with the attendant loss of revenues to recyclers, at $71,327,350.

According to two professors that have been involved in this issue for many years, dairy researchers attempted to implement these measures in the early 1980's. The effort did not succeed partly due to the fragmented structure of the dairy industry and partly due to counter efforts of marketing specialists to make milk more noticeable to consumers and improve milk sales. Whether or not a more concerted national effort to change lighting practices in milk dairy cases could succeed has not been determined.


1. Jan H. Schut, "Pigmented white jug aims to sweeten U.S. milk sales." Modern Plastics, Dec. 43-46, 1997 and Statement of H. P. Hood on Pigmented Milk Bottle, dated November 4, 1997.

2. S Toloken, "Recyclers worried over opaque milk bottles," Plastic news (Oct 27 '97).

3. Conversation with Rob Byrne of the International Dairy Foods Association, 11/11/97.

4. Solvay Polymers, "The Effects of Light on Packaged Milk," Technical Publication 9/97.

4a. "Milk Jugs: More Design Changes Appear," Plastics in the Environment, 12/97.

4b C. Gentry, "'Recyclable' Milk Cartons Might Not Be Recycled," Wall Street Journal (New Eng. ed.), 12/3/97.

5. The Markets Page, Recycling Times (1994 - 1997).

6. American Plastics Council, Resource Recycling Update (1995-6).

7. Provided by Ron Perkins of the American Plastics Council.

8. Bradley, R.L., Jr. "Effects of light on alteration of nutritional value and flavor of milk: a review." J. Food Protection 43: 314-320, 1980.

9. Dimick, P.S. "Photochemical effects on flavor and nutrients of fluid milk." Can. Inst. Food Sci. Technol. J. 15(4): 247-256, 1982; and Paul S. Dimick, PhD, "Protecting Milk's Vitamins from Light: Does It Matter?", paper presented to the Society of Chemical Industries, UK, 1995.

10. National Dairy Council, "Scientific Status Report #9: Effect of Light on Milk".

11. Conversations with Professor Robert Bradley, University of Wisconsin, Department of Food Science; Professor Paul Dimick, Pennsylvania State University, Department of Food Science; and Professor Bruce Harte, Michigan State University, School of Packaging.

12. Hedrick, T.I., and L. Glass. "Chemical changes in milk during exposure to fluorescent light." J. Milk Food Technol. 38: 129-131, 1975.

13. Senyk, G.F., and W.F. Shipe. "Protecting your milk from nutrient losses." Dairy Field 164: 81-85, 1981.

14. Ochtel, F.C., C.M. Stine. M.L. Richmond, Z.E.M. Saad, B.R. Harte. "The effect of fluorescent light on riboflavin and flavor quality of 2% milk packaged in high density polyethylene containers." J. of Packaging Technol. 1:4 124-127, 1987.

15. Reif, G.D., A.A. Franke, and J.C. Bruhn. "Retail dairy foods quality - an assessment of the incidence of off-flavor in California mild." Dairy and Food Sanitation 3: 44-46, 1983.

16. Conversation with David Boyden of Solvay Polymers, Inc., 12/22/97

17. Milk Industry Foundation. "Milk Facts" for 1995.


Professor Robert Bradley, University of Wisconsin, Department of Food Science

Professor Paul Dimick, Pennsylvania State University, Department of Food Science

Professor Bruce Harte, Michigan State University, School of Packaging

Fazila Seker, University of Wisconsin, School of Engineering, Department of Chemistry

Gail Wheeler, H.P. Hood, Inc.

Home Page
Project Overview - Publications
Coalition Members - Members Only - Contact Information
Links - Join the Plastic Redesign Project