Scientific advances and challenges in safety evaluation of food packaging materials: Workshop proceedings

Accepted Manuscript Scientific advances and challenges in safety evaluation of food packaging materials: Workshop proceedings Agnes L. Karmaus, Ron Osborn, Mansi Krishan PII:

S0273-2300(18)30202-2

DOI:

10.1016/j.yrtph.2018.07.017

Reference:

YRTPH 4180

To appear in:

Regulatory Toxicology and Pharmacology

Received Date: 11 July 2018 Accepted Date: 22 July 2018

Please cite this article as: Karmaus, A.L., Osborn, R., Krishan, M., Scientific advances and challenges in safety evaluation of food packaging materials: Workshop proceedings, Regulatory Toxicology and Pharmacology (2018), doi: 10.1016/j.yrtph.2018.07.017. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Scientific Advances and Challenges in Safety Evaluation of Food Packaging Materials:

2

Workshop Proceedings

3

Agnes L. Karmausa, Ron Osbornb, and Mansi Krishanc

4

a

5

b

6

c

7

NW, Suite 600, Washington, DC, 20005, USA

Integrated Laboratory Systems, Inc., 601 Keystone Park Drive, Suite 200, Morrisville, NC, 27560, USA Mars Wrigley Confectionery, Global Innovation Center, Chicago, IL

M AN U

SC

North American Branch of the International Life Sciences Institute (ILSI North America), 740 15th Street

8 9

RI PT

1

Correspondence: Agnes Karmaus ([email protected])

10

Abbreviations: CFR, Code of Federal Regulations; DIPN, di-isopropyl naphthalene; EFSA, European Food

12

Safety Authority; FACET, Flavorings, Additives, Contact Materials Exposure Task; FCMA, food contact

13

materials and article; FCN, food contact notification; FCS, food contact substance; FDA, Food and Drug

14

Administration; FFDCA, Federal Food, Drug, and Cosmetic Act; GC, gas chromatography; GRAS, generally

15

recognized as safe; GRN, GRAS Notification; HPLC, high-performance liquid chromatography; ILSI,

16

International Life Sciences Institute; LCA, life cycle assessment; MOAH, aromatic mineral oil

17

hydrocarbon; MOH, mineral oil hydrocarbon; MOSH, saturated mineral oil hydrocarbon; MS, mass

18

spectrometry; NOL, non-objection letter; PASTA, pack size type association table; PET, polyethylene

19

terephthalate; PFAS, perfluoroalkyl substances; PFOA, perfluorooctanoic acid; PFOS,

20

perfluorooctanesulfonate; QSAR, quantitative structure–activity relationship; RFID, radio-frequency

21

identification; SPC, Sustainable Packaging Coalition; TOF, time-of-flight; TOR, threshold of regulation

AC C

EP

TE D

11

1

ACCEPTED MANUSCRIPT 22 Abstract

24

Packaging is an indispensable component of the food manufacturing and food supply process. This

25

scientific workshop was convened to bring together scientists from government, academia, and industry

26

to discuss the state of the science regarding the safety of food packaging, prompted by rapidly

27

advancing research to improve food packaging that continues to impact packaging technology,

28

toxicology, exposure, risk assessment, and sustainability. The opening session focused on scientific

29

challenges in the safety assessment of food packaging materials. Experts discussed migration of

30

contaminant residues from food packaging, presented emerging analytical methods for safety

31

evaluation, and highlighted the use of improved exposure assessment models and new packaging

32

technologies. The workshop then focused on recycled packaging and sustainability. Experts also

33

discussed application of recycled materials in food packaging, recycling processes, identification of

34

contaminant residues from recycled packaging, and challenges in safety assessment of recycled

35

materials. The workshop concluded with panel discussions that highlighted the challenges and research

36

gaps in food packaging. Overall, there is a need to better understand and define “contaminants in food

37

packaging” for developing appropriate testing methods needed to establish the significance of the

38

migration levels of these contaminants and conduct appropriate safety assessments in this rapidly

39

evolving field.

40

Keywords: food packaging, risk assessment, toxicology, exposure, analytical methods

SC

M AN U

TE D

EP

AC C

41

RI PT

23

42

2

ACCEPTED MANUSCRIPT 43 1. Introduction and Purpose of the Workshop

45

Packaging is a system for preserving the safety and quality of food products throughout the distribution

46

chain to the consumer. In addition to containing food products, food packaging serves other important

47

functions such as protecting food products from outside influences and damage as well as providing

48

consumers with ingredient and nutritional information. Materials such as glass, metals, paper and

49

paperboard, and plastics have traditionally been used in food packaging and, along with inks, adhesives,

50

coatings, and so forth, may come in contact with food, presenting the possibility of unwanted chemical

51

residues migrating into food. To ensure the safe use of packaging materials and food contact materials,

52

regulatory agencies around the globe have conducted research and developed both guidance and

53

regulations for materials intended to contact food. The food packaging safety assessment frameworks

54

developed by the US Food and Drug Administration (FDA), Health Canada, and the European

55

Commission are built on similar principles; however, they differ in application of these principles and

56

migration testing methods for food packaging materials.

57

Workshop participants generally agreed that there can be inconsistent use of commonly used terms,

58

such that defining key concepts related to food packaging is needed. For example, there is a need to

59

better understand and define “contaminants in food packaging” in order to develop appropriate testing

60

methodologies to determine the significance of migration levels for contaminants. Three focal areas

61

addressed in the workshop have been summarized herein to establish a fundamental understanding of

62

food packaging (defined by the presenters; Table 1): (1) defining packaging materials and residue

63

migration, (2) regulatory framework for safety assessments, and (3) sustainability and recycling. These

64

three topics and critical definitions, as presented/discussed throughout the workshop, are summarized

65

in the next three sections. Additional points of interest relating to these topics are included in each

66

section based on topics discussed during panel sessions.

67

AC C

EP

TE D

M AN U

SC

RI PT

44

3

ACCEPTED MANUSCRIPT 68 Table 1: ILSI North America Food and Chemical Safety Committee “Workshop on the Scientific Advances and Challenges in Safety Evaluation of Food Packaging Materials” Program

Introduction Mr. Ron Osborn, ILSI North America Member Scientist

RI PT

Welcome Dr. Mansi Krishan, ILSI North America

Challenges in Safety Assessment of Food Packaging Materials – Toxicology Dr. Jason Aungst, Division of Food Contact Notifications, US FDA

Migration of Contaminant Residues from Food Packaging Dr. Gregory V. Pace, Sun Chemical Corporation

SC

Challenges in Safety Assessment of Food Packaging Materials – Chemistry Dr. Kirk Arvidson, Division of Food Contact Notifications, US FDA

M AN U

Analytical Methods for Evaluating Components of Food Packaging Materials Mr. Tim Begley, US FDA

Use of New/Improved Tools and Exposure Assessment Models for Food Packaging Materials Mr. Cian O’Mahony, Crème Global US Regulatory Perspective Dr. Paul Honigfort, Division of Food Contact Notifications, US FDA Global Regulatory Perspective Mr. James Huang, ILSI North America Member Scientist

TE D

Panel Discussion Moderator: Mr. Ron Osborn Panelists: Jason Aungst, Kirk Arvidson, Greg Pace, Tim Begley, Cian O’Mahony, Paul Honigfort, and James Huang

EP

Sustainability and Packaging: Process of Recycling Packaging Materials and Recycled Materials in Food Packaging Dr. Susan Selke, Michigan State University Contaminant Residues in/From Recycled Paper-Paperboard and Plastics: Contaminant Identification, Food Safety Concerns, and Regulatory Controls Dr. Vanee Komolprasert, Division of Food Contact Notifications, US FDA

AC C

69 70

Case Studies: Mineral Oil Saturated Hydrocarbons (MOSHs) and Mineral Oil Aromatic Hydrocarbons (MOAHs) Dr. Stephen Klump, Nestle Case Studies: Di-isopropylnaphthalene, and Poly- and Per-fluoroalkyl Substances (PFAS) Dr. Paul Honigfort, Division of Food Contact Notifications, US FDA Emerging Innovations and Technologies in Food Packaging Dr. Young Kim, Virginia Polytechnic Institute and State University

4

ACCEPTED MANUSCRIPT Panel Discussion Moderator: Mr. Doug Copen, ILSI North America Member Scientist Panelists: Susan Selke, Vanee Komolprasert, Stephen Klump, Paul Honigfort, and Young Kim 71 2. Defining Food Packaging Materials and Identification of Potential Migration Contaminants

73

The fundamental explanations regarding how food packaging materials and potential migration

74

contaminants are defined and identified were presented. Insight on how components of food packaging

75

materials are categorized and analytically identified was presented and context regarding different

76

levels of the supply chain was provided. Overall, the complexity of both the production and evaluation

77

of food packaging materials was emphasized by speakers and panelists. Furthermore, advances in

78

exposure assessments for food packaging residues as well as advances in the generation of new “smart

79

packaging” materials were presented, to shed light on emerging science related to food packaging

80

materials.

81

2.1 Defining substances that migrate from food packaging

82

Speaker: Dr. Gregory V. Pace (Sun Chemical Corporation)

83

The workshop began by highlighting that contact with food by substances can be intentional or non-

84

intentional. Intentionally added substances are chemicals known to be present in supplied materials,

85

which typically have a technical effect and/or have a performance property (FDA, 2017a). Conversely,

86

non-intentionally added substances are chemicals present in the supplied materials which are either an

87

unknown impurity or an unknown reaction by-product (FDA, 2017a). In fact, a guidance document for

88

non-intentionally added substances in food contact materials has been developed to aid in the

89

evaluation non-intentional food contact substances (Koster et al., 2015).

AC C

EP

TE D

M AN U

SC

RI PT

72

90

Substances that migrate from food packaging are a concern because of their potential impact on

91

food properties and/or consumer safety. If these migratory substances do affect the properties or safety 5

ACCEPTED MANUSCRIPT of food, they are considered adulterants or contaminants (see Code of Federal Regulations 21 CFR

93

174.5(a)(1)). The definition of contaminants must be clearly understood for appropriate regulatory

94

review and to optimize the testing needed to establish the significance of the migration levels. For

95

example, differentiating between food and food ingredients can be a source of confusion. Regulation of

96

a contaminant’s technical effect on food is the priority rather than regulating solely based on migration

97

occurring. Identifying all substances in each material used for food packaging is critical for a thorough

98

evaluation of whether any migration to food occurs. The design of packaging materials typically includes

99

characterizing drivers of migration and how they can be controlled to avoid unintended food addition

SC

RI PT

92

(i.e., addition of sealing materials over surfaces to minimize migration). Generally, chemicals with a size

101

of <1000 Da are considered for migration testing, which is accomplished using trial packages. There are

102

multiple sources of contaminant residues in the supply chains, including presses (fountain solutions,

103

press washes, lubricants, and additives), substrates (plasticizers, surfactants, stabilizers, antioxidants,

104

resins, antimicrobials), inks and coatings (resins, polymers, adhesives, pigments, solvents, oils,

105

monomers, additives), and the environment (pesticides, cleaners, temperature effects, fumes).

106

Ultimately, the regulatory framework is chemical substance focused; therefore, the ability to detect

107

chemicals, model the migration and exposure, as well as determine the chemical’s effect (technical

108

effect on food and safety) are required.

EP

TE D

M AN U

100

Identifying chemical substances comprising packaging materials is not trivial. Challenges with

110

proprietary information when evaluating the composition of substances compounds the difficulty, as

111

materials are often produced by multiple entities along the food contact supply chain before the final

112

package comes into contact with packaged food (Figure 1). More specifically, this process requires

113

obtaining relevant documentation and cooperation from chemical substance producer(s) to suppliers,

114

food packaging material suppliers, converters, consumer product groups, and ultimately regulators,

115

enabling a determination of what the consumer may be exposed to from the final packaged retail food

116

product. To ascertain chemical substance identities in packaging materials, it may be necessary to

AC C

109

6

ACCEPTED MANUSCRIPT review technical data sheets, safety data sheets, letters of compliance, or supplier compliance

118

questionnaires; in some cases, it may require obtaining confidential disclosure agreements or a

119

declaration of compliance from suppliers. If the supplier cannot provide the needed documentation or

120

facilitate this process, then a complete regulatory review of the material is not possible. Pursuing

121

regulatory assessment in such cases requires the submitter to provide supplier contact information to

122

the regulator. Disclosure of relevant information to subsequent vendors in the supply chain can be a

123

challenge and may also be a global issue (i.e., which country is the substance/material/article coming

124

from and going to? Is compliance achieved in the country in which the product is intended to be

125

marketed or sold?).

127 128 129

Figure 1. Food contact and packaged food supply chains. These two distinct supply

AC C

126

EP

TE D

M AN U

SC

RI PT

117

chains intersect where packaging leads to the food contact articles coming in contact with processed food.

130

2.2 Analytical approaches for identifying substances that migrate from food packaging

131

Speakers: Mr. Tim Begley (US FDA), Dr. Paul Honigfort (US FDA), and Dr. Stephen Klump (Nestle)

7

ACCEPTED MANUSCRIPT The analysis of food contact materials is more challenging because there are no defined standard

133

methods. Inherently, packaging materials are designed to be solid and are not intended to “come

134

apart,” rendering testing a challenge based on the materials used. There are no official lists of reference

135

chemicals or analytical standards and there is no set of guideline experimental designs to accomplish

136

testing. The guidelines for pre-market submissions to the FDA provide insight on required data, stating

137

that all major impurities and their concentrations must be identified and submitted together with any

138

supporting analytical data and calculations [see Section II.A, item d in FDA (2007): “Concentrations of all

139

major impurities (e.g., residual starting materials, including all reactants, solvents, and catalysts, in

140

addition to byproducts and degradation products). In the case of polymers, concentrations of residual

141

monomers and oligomers should be included.”]. This can be a significant challenge and is generally

142

addressed by using targeted and non-targeted testing.

SC

M AN U

143

RI PT

132

The application of different chromatographic approaches paired with mass spectrometry (MS), for which time-of-flight (TOF) methods and Orbitrap ion trap mass analysis are the most commonly

145

used, is the best approach to isolate different types of residue chemicals improving detection and

146

analysis of food contaminants. For example, many types of gas chromatography (GC) are available to

147

evaluate volatiles and numerous high-performance liquid chromatography (HPLC) methods can be

148

applied for accurate and sensitive detection of non-volatiles. Important considerations for the proper

149

use of these technologies include confirming that the analyte makes it through the injection system and

150

elutes, and the instrument is calibrated with proper standards. Furthermore, to be acceptable for

151

regulatory assessment, the methods for migration tests must have positive controls at the expected

152

concentrations to demonstrate mass balance until the end of the test. For non-targeted analyses,

153

accurate MS detection must have very high mass resolution (>10,000 Da) and can be very sensitive with

154

approaches offering mass accuracy of as low as ~1 ppm. For determining molecular formulas, a mass

155

accuracy of <3 ppm is required (FDA, 2015). These powerful innovative approaches are continuously

156

becoming more and more sensitive, ameliorating our ability to detect residues and food contaminants.

AC C

EP

TE D

144

8

ACCEPTED MANUSCRIPT 157

An example presented by Mr. Begley described paper coatings, approved as early as 1967, none of which listed certain perfluoroalkyl substances (PFAS) like perfluorooctanoic acid (PFOA) as part of

159

their production process (i.e., the reaction products or degradation products) for decades. This oversight

160

was a by-product of the lack of instrumentation to detect PFAS residues. By 2003, PFAS (namely, PFOA

161

and the degraded perfluorooctanesulfonate [PFOS]) were both confirmed as ubiquitous contaminants of

162

paper coatings in the United States and had been identified in human serum from occupational

163

exposures (Olsen et al., 1999, 2003). PFAS are biopersistent reproductive and developmental toxicants

164

that bioaccumulate, making exposure determination challenging and hence risk assessment difficult. Dr.

165

Honigfort (FDA) further addressed this example with a case study to highlight the impact of improved

166

analytical methods and increased sensitivity to detect contaminant residues. The ability to detect PFAS

167

and their toxicity resulted in a change in authorization status for PFAS in food packaging: the

168

authorizations for the use of PFAS in food packaging were either revoked or voluntarily discontinued.

169

PFAS had previously been authorized for use both via the food additive process (which results in a

170

regulation) and the food contact notification (FCN) process (which results in an effective notification). In

171

2011, the FDA reached a voluntary agreement with manufacturers to no longer sell any food packaging

172

containing any PFAS that was subject to an effective notification (FDA, 2017b). The final ruling to revoke

173

the regulations for PFAS was published in 2016 (FDA, 2016a). The change in regulatory status meant that

174

the presence of PFAS in recycled paper changed from an authorized food additive to a contaminant. To

175

address the ability of a recycled paper manufacturing process to remove contaminants, the FDA

176

generally encourages surrogate testing of recycled materials rather than batch testing.

SC

M AN U

TE D

EP

AC C

177

RI PT

158

Another example was presented by Dr. Klump (Nestle), further discussing how accuracy and

178

limit of detection for technological methods can impact the identification of impurities. This example

179

focused on mineral oil hydrocarbons (MOHs), specifically saturated and aromatic mineral oil

180

hydrocarbons (MOSH and MOAH, respectively). Derived from crude oil, MOHs are typically 75% MOSH

181

and 25% MOAH, with technical-grade MOH containing 10–35% MOAH and highly purified food-grade 9

ACCEPTED MANUSCRIPT MOHs (i.e., white oils) having the MOAH content minimized down to 0–5%. While MOSHs lack evidence

183

for any toxicological concern, the presence of MOAHs presents carcinogenic concern. Further

184

complicating the detection of MOAH or MOSH is the fact that MOHs are mixtures and even in GC-MS

185

chromatograms can only be reported as humps or groups of peaks, which should be confirmed by

186

replicates with ion extraction. Another approach has been to use the presence of di-isopropyl

187

naphthalene (DIPN) as a marker for possible MOH contamination, as DIPN is a contaminant found in

188

recycled paper. Ultimately, these examples highlight that it is critical to appropriately select the

189

detection method for impurities as well as thoroughly characterize the limit of detection and possible

190

limitations leading to false positives.

191

2.3 Advances in exposure assessments

192

Speaker: Mr. Cian O’Mahony (Crème Global)

193

Exposure is a critical component of regulating food packaging materials, posing unique challenges. While

194

the identification and quantification of food packaging residues can be addressed empirically, exposure

195

is much more difficult to quantify precisely. Dietary exposure predictions are generally computed as a

196

function of the amount of food and the concentration of the substance in food. This often requires some

197

assumptions; in Europe, the exposures are estimated as a function of 1 kg food consumed in contact

198

with 6 dm2 packaging with migration at a specific migration limit. However, this approach could be

199

refined, given the wealth of data and study that has gone into exposure modeling as a science. For

200

example, “amount of food” can be replaced with food consumption measurements and “concentration

201

of substance” can be parameterized by using migration information from packaging data. Examples

202

were presented of different migration protocols and exposure scenarios required to assess different

203

articles in contact with packaged infant formula (liquid vs. powder formula, or articles in bottles vs.

204

nipples), as well as assumptions for infant dietary intake (variables like infant body weight used and

205

infant formula intake used). Furthermore, because the use of calories vs. grams results in very different

AC C

EP

TE D

M AN U

SC

RI PT

182

10

ACCEPTED MANUSCRIPT 206

estimates, considerations for using caloric intake vs. daily intake for exposure estimates were raised

207

during the panel discussion. Dr. Arvidson (FDA) confirmed that the adoption of consumption factors to

208

address caloric intake is being considered. The Flavorings, Additives, Contact Materials Exposure Task (FACET) collaboration seeks to

RI PT

209

develop tools to facilitate harmonized approaches for estimating exposure to target food chemicals

211

(Oldring et al., 2014a, 2014b, 2014c). More specifically, FACET’s work includes a harmonized database

212

on food intake data and a database on the occurrence of food chemicals in foods. In addition, migration

213

models for multi-layer packaging, quantitative structure–activity relationship (QSAR) tools for estimating

214

toxicity of food contact substances, and exposure software development for estimating exposure to

215

target food chemicals are all initiatives of this consortium (European Commission Joint Research Centre,

216

2018). To redefine the food dietary exposure model, the EuroMonitor International database was used,

217

which amalgamates European dietary surveys from multiple EU countries as well as market surveys

218

containing migration and packaging information (i.e., food categories, packaging types, pack size,

219

contact areas, market shares, and consumption and demographic information). These data were a

220

challenge to compile due to the proprietary nature of the information. As a result, the information has

221

been maintained in a blinded/coded format, enabling exposure scientists to have anonymized values

222

that can be used to obtain a representative distribution of values to be used for analysis, which cannot

223

be associated with any product or company. These distributions include varying mass/area ratios for

224

products (i.e., bulk store vs. snack-size portions) with the size/geometry and amount of food provided.

225

The packaging exposure algorithm developed by FACET integrates all packaging that contains a

AC C

EP

TE D

M AN U

SC

210

226

specific substance of concern using a pack size type association table (PASTA). All foods contained in

227

that packaging are then identified and the migration of the substance into each of those foods is

228

obtained using data in a EuroMonitor International database (diffusion coefficients for each

229

material/migrant combination, partition coefficients, and densities of foods and materials are all

11

ACCEPTED MANUSCRIPT considered). To clearly define what food can have which packaging migrants, accounting for the multi-

231

layered structure of packaging, FACET has focused on direct food contact materials to create a tiered

232

system integrating food categorization, wherein pack type is defined by four components: (1) domain

233

(structure of the packing), (2) closure/lidding, (3) outer/wrappings, and (4) inserts. After identifying

234

migration into any foods packaged using any of the food contact materials containing the substance of

235

interest, the consumption of those foods is determined and the exposure to the substance is estimated.

236

The exposures from all products, representative of the participating European countries, are collated to

237

give a distribution of exposure to the substance in the population. Since packaging is generally

238

consistent across European countries, analyses considering differences in consumption across EU

239

populations have been conducted, revealing exposure estimates within the same order of magnitude

240

across the countries. The findings suggest that although there are some differences in consumption,

241

these differences are not profound. This large FACET effort serves as a great example for what could be

242

accomplished when exposure data can be shared and integrated using advanced modeling and

243

informatics approaches.

244

2.4 Advances in the generation of new packaging materials

245

Speaker: Dr. Young Kim (Virginia Polytechnic Institute and State University)

246

New forms of packaging are being created as technological advances and creative solutions are

247

developed to address evolving consumer needs. While traditional packaging is considered passive, new

248

“smart packaging” materials give the products something extra. There are two forms of smart

249

packaging: active packaging and intelligent packaging. Active packaging is a type of smart packaging that

250

provides enhanced protection via an active property of the material (Robertson, 2012). Examples of

251

active packaging include oxygen scavenging packaging and modified atmosphere packaging. The former

252

actively serves to remove residual oxygen from bags to keep foods fresh; such systems include both

253

headspace scavenging and barrier enhancement. Modified atmosphere packaging controls the air

AC C

EP

TE D

M AN U

SC

RI PT

230

12

ACCEPTED MANUSCRIPT composition inside the packaging to affect gas exchange ultimately optimizing the atmosphere inside

255

the packaging to support breathing rate, ripening, and improved shelf-life for products. Furthermore,

256

antimicrobial packaging materials have also been developed, which comprise coating or integrating

257

antimicrobials into packaging materials. Intelligent packaging is the integration of enhanced

258

communication and traceability of product transparency to consumers (Han, 2005). The use of radio-

259

frequency identification (RFID) systems to track all aspects of food and/or packaging production has

260

been integrated for some products, and some even enable consumers to see when and where the

261

product was harvested and packaged. Ultimately, it was emphasized that new “smart packaging”

262

materials are transforming the food packaging industry and will significantly impact the way in which

263

food packaging materials are designed, used, and evaluated in the future.

264

3. Regulatory Framework for Food Packaging Substances

265

Regulation of food packaging substances can be complex, and parts of this critical process can pose

266

significant challenges. Several workshop presentations addressed the regulatory framework, including

267

regulatory mechanisms and how to adequately meet regulatory requirements. Experts presented

268

summaries of the toxicological, chemical, and regulatory infrastructure, respectively. Furthermore, they

269

also provided an international regulatory perspective, revealing challenges for meeting the needs of

270

regulatory agencies on a global level. These presentations all contributed to a thorough examination of

271

the current regulatory framework and requirements for food packaging materials, which are

272

summarized in this section.

273

3.1 The US regulatory approach for food packaging

274

Speakers: Dr. Paul Honigfort (US FDA), Dr. Jason Aungst (US FDA), and Dr. Kirk Arvidson (US FDA)

275

FDA’s jurisdictional authority over food comes from the Federal Food, Drug, and Cosmetic Act (FFDCA)

276

(FDA, 1938) and can be conducted through several regulatory mechanisms, all of which are based on

AC C

EP

TE D

M AN U

SC

RI PT

254

13

ACCEPTED MANUSCRIPT 277

safety. The standard of safety is the same for all regulatory mechanisms, wherein there is no risk/benefit

278

analysis conducted. All mechanisms require a “reasonable certainty of no harm” (21 CFR 170.3). The

279

information necessary to demonstrate safety is dependent on exposure and intended use. Several definitions were emphasized as fundamental to understand which regulatory approval

RI PT

280

process to pursue for food packaging materials (Table 2). Foremost, the FDA (2018a) defines a food

282

additive as something that is intentionally added to food and has a functional effect in food. Any food

283

packaging substance that migrates into food is considered an indirect food additive. Most food additives

284

that are authorized for direct addition to food cannot automatically be used as indirect food additives,

285

with the exception of colorants. Color additives (substances authorized for direct addition to food for

286

the purpose of imparting color) can automatically be used in plastic packaging materials and in that

287

context are considered colorants, not color additives (21 CFR 178.3297). In addition, substances with

288

generally recognized as safe (GRAS) designations for direct addition to food (21 CFR 184) are also

289

considered GRAS for indirect additive use (21 CFR 184.1(a)).

M AN U

TE D

290

SC

281

There are several authorization mechanisms that can be pursued for registering food contact substances. These mechanisms are distinct, and the decision criteria for each are summarized in Table 2.

292

Obtaining GRAS status via the GRAS notification (GRN) process is generally pursued based on consensus

293

in available data. The GRN process is distinct from other regulatory mechanisms for food additives

294

because the GRN process is voluntary, whereas the other processes are mandatory before producing

295

new food packaging materials. GRN can be obtained for prior sanctioned substances (21 CFR 181) or can

296

be “new use GRAS,” wherein a new use for a prior sanctioned substance is submitted for GRAS

297

consideration. GRN review takes an approximate 180 days and is less regimented, whereas the food

298

additive regulatory mechanisms (including food additive petitions, which result in regulations, as well as

299

FCNs and threshold of regulation assessments) involve an environmental assessment component

300

(mandated by the National Environmental Policy Act of 1969) that is not present in GRN. Food additive

AC C

EP

291

14

ACCEPTED MANUSCRIPT regulations, TORs, and GRNs are generally applicable, meaning any manufacturer of the substance can

302

rely on these authorizations. However, FCNs are manufacturer specific, meaning only the listed

303

manufacturer can rely on an effective FCN—if a new manufacturer wished to sell the same substance for

304

the same use, they must submit their own FCN. The same safety standard (reasonable certainty of no

305

harm) is applied for all food additive authorization mechanisms. The current review process for food

306

additives at the FDA includes a chemist, a toxicologist, and an environmental review scientist assigned if

307

an environmental impact is deemed relevant. Consumer safety officers monitor submissions through the

308

process, ensuring that the submitter answers questions from reviewers and that the FDA fulfills its legal

309

obligations during the review of the submission.

SC

M AN U

310

RI PT

301

Table 2. Current regulatory mechanisms in the United States for food contact substances Decision criteria

Daily concentration > 1 ppm




Food additive petition

Daily concentration < 1 ppm




Food contact notification (FCN)

Daily concentration < 0.5 ppba




Threshold of regulation (TOR)




GRAS notification (GRN)

Food additives

Public availability and

a

312

b

314 315

consensus of data

For TOR submission, there are additional criteria for submission (FDA, 2005). Submission of a GRN to FDA is not required prior to marketing, industry can reach “self-

AC C

313

b

EP

(by available data) 311

TE D

(by exposure)

GRAS

Regulatory mechanism

determination” that the use of a substance is GRAS. It is essential that the statutory

standard for GRAS must be met regardless of whether GRN is pursued vs. self-determination is conducted (see GRAS criteria in 81 FR 5490, August 17, 2016) (FDA, 2016b; FDA, 2018b).

316

More recently, food packaging materials are assessed via the FCN process. Compiling an FDA

317

FCN submission has four main parts: chemistry, toxicology, environmental, and administrative forms

318

(FDA, 2002a, 2002b, 2006a, 2007). The FDA urges making use of pre-notification consultations to assist 15

ACCEPTED MANUSCRIPT throughout the regulatory process in the form of written and/or oral correspondence. Such consultation

320

is made available to facilitate the clarification and interpretation of regulatory requirements, pre-

321

submission review of the safety package, discussion of alternative approaches to determining exposure

322

and/or safety, as well as input on analytical methodology. Engagement between notifiers and regulators

323

is of benefit to both parties.

324

RI PT

319

There are nine different components summarizing the chemical information pieces in the FCN: (1) identity, (2) physical/chemical specification, (3) manufacturing information, (4) impurities, (5)

326

conditions of use, (6) technical effect, (7) stability under the intended conditions of use, (8) migration

327

levels in food, and (9) exposure estimates. The exposure estimates are based on migration levels in food.

328

Sufficient supporting analytical data addressing these nine components are the largest challenge in

329

compiling successful FCN submissions. Supporting analytical data must contain the necessary context

330

and details on processing/analysis to convert “raw” analytical data on the amount of a food contact

331

substance (FCS), and its impurities, that migrates to food to an estimated daily intake. The information

332

should be of sufficient detail such that all calculations can reproduced by an FDA review scientist.

333

Supporting analytical data should include raw chromatograms, calibration curves, validation data, and a

334

description of the data.

335

3.2 Global perspectives on food packaging regulation

336

Speaker: James Huang (ILSI North America Member Scientist)

M AN U

TE D

EP

AC C

337

SC

325

To make comparisons across regulatory frameworks, it is first important to understand the

338

differences in terminology used by the various regulatory agencies. For example, the United States and

339

the European Union have varying definitions for FCSs and food contact materials (FCMs) and articles,

340

respectively. The United States defines FCSs as any substance that is intended for use as a component of

341

materials used in manufacturing, packing, packaging, transporting, or holding food if such use of the

342

substance is not intended to have any technical effect in the food (as consumed) (FFDCA §409(h)6). On 16

ACCEPTED MANUSCRIPT the other hand, the EU Framework Regulation provides a list of 17 groups of materials and articles for

344

food contact (Annex I, Regulation (EC) No. 1935/2004) (European Parliament, 2016). This definition, in

345

contrast to that of the United States, does not address the concept of intended use, the requirement of

346

inertness, or the hierarchy of the supply chain (substance > material > article), nor does it specify the

347

connection to food. Furthermore, delineating food contact vs. packaged food and their supply chains

348

(Figure 1) is critical; although these supply chains intersect, the approach to safety is different for each

349

respective supply chain among regulatory agencies.

SC

RI PT

343

The parameters used to establish safety are distinct across regulatory agencies, and harmonizing the

351

requirements is not necessarily feasible due to the differences in the approaches used by the respective

352

agencies. This creates compliance gaps, which in turn are regarded as safety gaps. This problem

353

becomes further complicated when more countries with disparate compliance structures are also

354

considered, creating a challenge due to packaging being regulated by different compliance systems. To

355

highlight major differences in regulatory approaches, S-polyethylene terephthalate (PET), a polyester

356

used as a special glycol introduced in 2010 by Japan, was presented as a case study. Ultimately, both the

357

United States and European Union determined this polymer to be safe for food contact and safety

358

parameters were established for compliance purposes. FDA FCN 1135 came out in 2012, specifying that

359

S-PET, a polymer of three monomers, has an intended use specifically for materials with a maximum

360

thickness of 5 mm, among other specifications (FDA, 2012). Meanwhile, in the European Union,

361

European Food Safety Authority (EFSA) opinion FCM 1052 was released in 2014 followed by regulation

362

in 2016 with guidance stating that the substance must not exceed 5 ppm (mg/kg food) and that the

363

migration of oligomers less than 1000 Da may not exceed 50 µg/kg food (EFSA, 2014). The EFSA

364

guidance also requires that any final material or article containing S-PET must have a well-described

365

method to determine oligomer migration for compliance assessment by a competent authority, further

366

stating that a sufficient sample be supplied upon request if the method requires a calibration sample.

AC C

EP

TE D

M AN U

350

17

ACCEPTED MANUSCRIPT 4. Food Packaging Sustainability and Recycling

368

4.1 Considerations for sustainability and recycling of food packaging

369

Speaker: Dr. Susan Selke (Michigan State University)

370

To address sustainability of food packaging materials, it is important to understand that sustainability

371

and recycling are two distinct concepts. Sustainability is generally defined as operating in ways that

372

meet current needs without adversely impacting the ability of future generations to meet their own

373

needs. Three major components contributing to sustainability under this definition are the economic,

374

social, and environmental effects. The Sustainable Packaging Coalition (SPC) (2011) has a definition

375

specifically for sustainable packaging, which provides more detail for each of these components and

376

relates to packaging. Conversely, for the purpose of discussing packaging, recycling is defined as the

377

collection, sorting, processing, and marketing of materials after their initial use.

SC

M AN U

378

RI PT

367

Evaluation of the environmental aspects of sustainability must consider the whole packaging system and the entire life cycle inventory of the packaging-product system (Figure 2). While environmental

380

impacts tend to be lower with recycled materials (recycled content usually results in greater

381

sustainability), not all materials are good candidates for recycling, so reducing the amount of material

382

may be more beneficial. For example, plastics generally use less energy for production and distribution,

383

with a number of life cycle assessments (LCAs) (ISO, 1998) confirming lower environmental impact for

384

plastic bottles over other materials. However, in considering rigid vs. flexible plastic packaging, the net

385

impact was often found to be lower for flexibles (which use much less total material despite the fact

386

that they are typically composed of a combination of materials rendering them unlikely to be recycled)

387

than for rigid materials (which use more material but are generally easy to recycle and have high

388

recycling rates).

AC C

EP

TE D

379

18

RI PT

ACCEPTED MANUSCRIPT

SC

389

Figure 2. Life cycle inventory considerations for assessing the environmental component

391

of sustainability for packaging materials (adapted from life cycle stages; Socolof et al.,

392

2001).

M AN U

390

It is also important to consider that the impact of production is usually greater than that of disposal;

394

therefore, focusing on end-of-life can be misleading when evaluating sustainability (S. Selke, personal

395

communication) (Burek et al., 2018; Dhaliwal et al., 2014; Kang et al., 2013). For example, plastic bottles

396

use less energy for production and distribution, which may render them more sustainable than glass,

397

despite glass being more readily recycled and having more recycled content; consideration of the energy

398

input up front can influence sustainability. Thus, LCAs are used to facilitate comparisons of whole

399

systems to avoid unintended consequences; an inventory of what goes in and what comes out of a

400

system is tabulated and put into a broader conceptual framework to evaluate what this means in terms

401

of environmental impact (Scientific Applications International Corporation and Curran, 2006).

EP

AC C

402

TE D

393

The recycling process includes the collection, sorting, processing, and marketing materials/products.

403

The use of recycled content usually results in greater sustainability (decrease in overall system impacts)

404

if material types and amounts are similar, if contamination is not excessive, and if degradation of the

405

material due to use and reprocessing is minimal; this is not easy to achieve. If any of the steps fail, then 19

ACCEPTED MANUSCRIPT 406

recycling has not occurred and a new form of waste is produced. Therefore, it is important to consider

407

how functional, costly, and available recycling is. These factors also impact motivation of households

408

and businesses to recycle or contribute to the sustainable solution. Challenges facing sustainable recycling solutions are vast. Combinations of materials being difficult

RI PT

409

to separate (i.e., full-body shrink labels on bottles, direct printing, new plastic resins, coatings, single

411

materials to multiple layers, etc.) pose substantial processing considerations. The contamination

412

likelihood increases with commingled collection of a variety of items, which needs to be weighed against

413

the price, approachability, and motivation of households and businesses to recycle and contribute to a

414

sustainable solution. For instance, consider residential recycling curbside vs. drop-off sites vs. business

415

recycling, all of which are voluntary and require various combinations of hand labor and mechanized

416

systems for sorting materials. While drop-off sites have the advantage of potentially collecting more

417

types of recyclables and being amenable to multi-dweller properties and businesses sort recyclables by

418

material category, the need for processing by a processor or transfer station must be integrated because

419

there is no standard structure for this system. Additionally, consideration of the cost to implement

420

recycling, shipping distances, emissions, energy used for processing of recyclables, and waste generation

421

should also be addressed. Finally, post-processing to make recycled materials suitable for food contact

422

may not be environmentally preferable to use of virgin materials; therefore, availability of other markets

423

for the recycled materials is a key consideration.

AC C

EP

TE D

M AN U

SC

410

424

The production of recycled materials for packaging is generally pursued when there is a market for

425

the materials, yet users do not typically create systems to use these materials unless there is adequate

426

supply for their needs; all of this must be done at reasonable scale, which can be particularly

427

problematic for processing new materials. Thus, suitability of materials for processing and the use of

428

recycled materials vary by material and type of product. There are open and closed loop systems; a

429

closed loop system entails materials being recycled into the same material (i.e., glass bottles back into

20

ACCEPTED MANUSCRIPT glass bottles), whereas an open loop system repurposes the materials. Metal and glass are generally

431

amenable to closed loop systems, as the recycled materials are generally as safe as virgin materials for

432

direct food contact. Both materials can be melted at very high temperatures during processing, avoiding

433

contamination as most organic materials will be destroyed, and there is no decrease in the performance

434

properties for the recycled product. On the other hand, steel cans are typically recycled via an open loop

435

system due to sheet steel requiring a rather pure alloy and being made in very few facilities. In terms of

436

sustainability, it is not efficient to ship post-consumer steel cans to these sparsely available facilities

437

when more approachable options are available; most steel cans are recycled into rebar or a variety of

438

other non-food products.

SC

M AN U

439

RI PT

430

The final consideration for integrating the use of post-consumer recycled content for packaging is the effect recycling has on the material integrity. For example, recycling affects the strength of

441

corrugated boxes, and reduced strength may require an increase in box weight (more paper fiber);

442

optimizing recycled content depends on requirements for the final product. Paper made from a mix of

443

feedstocks is generally weaker, use and reprocessing typically results in some fiber damage and

444

weakening, and temperatures are much more moderate and are not high enough to destroy organic

445

contaminants. For plastics, the effort required for critical sorting steps to be conducted appropriately

446

must be considered. There are a wide variety of plastics, and many are incompatible with each other;

447

sensitivity to degradation during use and reprocessing varies by resin type (i.e., melting temps). Sorting

448

for subsequent food-use to make food-grade plastics is done separately; as such recycling streams may

449

be a mix of food-grade and non-food-grade plastics. Food-grade post-consumer recycling requires

450

control over source materials and intensive processing (high-temperature washing with detergent,

451

vacuum treatment, and sometimes even depolymerization/repolymerization, which may not be

452

economical).

453

4.2 FDA’s regulatory approach for recycling

AC C

EP

TE D

440

21

ACCEPTED MANUSCRIPT Speaker: Dr. Vanee Komolprasert (US FDA)

455

The US FDA review for recycled materials for food packaging is voluntary and is conducted on the

456

recycling process itself rather than for the recycled product. An informal non-objection letter (NOL) can

457

be obtained from the FDA for the effective recycling process that demonstrates effective cleaning and

458

production of recycled material that is suitable for food contact use. The NOL is not legally binding, does

459

not necessarily endorse the recycled materials/products, and applies to the process regardless of

460

manufacturer/recycler. Sublicensing of a recycled process does not involve the FDA, and a company that

461

is sublicensing a NOL process does not need to obtain a new NOL as long as the process (especially the

462

cleaning steps) is exactly the same as the original NOL process.

M AN U

SC

RI PT

454

FDA’s recycled paper and paperboard regulation pertains to the process and material source

464

control (21 CFR 176.260). Briefly, input material cannot contain any substance that is deleterious and

465

could migrate to food, requiring processing or pre-sorting to avoid any such substances. Recycled paper

466

materials are generally prepared by re-pulping with water to recover the fiber with the least possible

467

amount of non-fibrous pulp substances. There have been very few voluntary paper recycling

468

submissions submitted to FDA in the past decade.

There is no 21 CFR section that explicitly addresses post-consumer recycled plastics for food

EP

469

TE D

463

contact applications; rather, there is a default to 21 CFR 174.5, which contains general provisions

471

applicable to indirect food additives. Recycled plastics must be safe for food contact use, meaning that

472

recycled materials should be derived from food-grade material and should meet all existing applicable

473

authorizations. The FDA (2006b) has issued a "use of recycled plastics in food packaging: chemistry

474

considerations" guidance document (developed in 1992 and updated in 2006). This guidance addresses

475

some of the known chemistry issues (namely, the migration of chemicals contaminants from post-

476

consumer recycled plastics to food). Guidance establishes an acceptable upper limit of dietary

477

concentration at 0.5 ppb to chemical contaminants from recycled material (adopted from 21 CFR

AC C

470

22

ACCEPTED MANUSCRIPT 170.39, TOR exemption) and recommends surrogate testing protocols for evaluating the efficacy of a

479

proposed recycling process to remove various unknown chemical contaminants. Combined surrogate

480

testing and migration modeling are typically used to estimate migration levels and to demonstrate a

481

dietary concentration to each contaminant of <0.5 ppb, the TOR limit that FDA equates to negligible risk

482

for a contaminant. The surrogate testing protocol is a means to evaluate the efficacy of a proposed

483

recycling process in cleaning recycled plastic for food contact use. This process involves an intentional

484

contamination of recycled plastic with known chemicals (surrogates), chosen to simulate various

485

physical and chemical properties of potential unknown contaminants. The intentionally contaminated

486

plastic is then subjected to the proposed recycling process and analyzed to measure residual levels of

487

surrogates in the recycled materials. Further clarification explained three distinct plastic recycling

488

processes: (1) primary recycling, wherein the use of pre-consumer, post-industrial, or industrial scrap

489

and salvage are used to form new packaging; (2) secondary recycling, which is physical reprocessing

490

typically involving aqueous washing, drying, re-melting, and reforming; and (3) tertiary recycling,

491

involving chemical processing (i.e., hydrolysis, methanolysis, and glycolysis, which are applied to PET),

492

depolymerization of post-consumer plastic starting materials, purification of starting materials, and

493

repolymerization of the starting materials to form regenerated polymers.

494

5. Conclusions and Key Workshop Outcomes

495

This workshop provided a forum to discuss the state of the science on safety assessments, detection and

496

quantification of exposure, sustainability, and risk assessment frameworks as they relate to the safety of

497

food packaging. The workshop brought together scientists from government, academia, and industry to

498

discuss advances in the food packaging field, identify research gaps, and gain consensus on best

499

practices.

AC C

EP

TE D

M AN U

SC

RI PT

478

500

The US FDA Division of Food Contact Notifications indicated a shift in priorities to include

501

significantly more post-market activities, changes in the market place, and advances in science that have 23

ACCEPTED MANUSCRIPT created competing priorities as well as challenges to incorporate the best science and most recent

503

information in exposure assessments. Challenges in the application of new science for safety

504

assessments of food packaging materials include validating and interpreting the data from high-

505

throughput screening, in vitro assays, computational toxicology, thresholds of toxicological concern, and

506

understanding the mode of action and development of adverse outcome pathways. FDA has initiatives

507

to update the chemistry guidance, including making updates to references, applying migration modeling

508

to consumption factors, and reorganizing the guidance to follow the formatting of Form 3480 (K.

509

Arvidson, personal communication). Challenges with migration modeling and appropriateness of certain

510

models are being addressed (i.e., QSAR models). For example, some challenges with the application of

511

QSAR approaches have been met through collaboration with other FCS centers as well as QSAR model

512

developers. Participation in the development of QSAR models helps ensure that the FDA can validate

513

any QSAR predictions submitted in support of a notification.

SC

M AN U

514

RI PT

502

Strategies to improve the reliability of analytical methods for measuring and evaluating migration residues were discussed and establishing some standard analysis methods or reference

516

materials was advised. The main challenges specifically in the analysis of food contact materials include

517

a lack of analytical standards, reference materials, or standard methods of analysis for many

518

components of food contact materials. False negatives and limits of detection are also of great concern

519

when identifying food contaminants and residues, with advances in analytical methodologies allowing

520

for progressively lower limits of detection leading to the unexpected detection of chemicals. Workshop

521

panelists noted that it is critical to quantify uncertainty and contextualize the amount of chemical

522

detected with a threshold of toxicological concern to determine significance, and that educating the

523

public on this concept is required to mitigate issues surrounding perceived risk vs. real risk.

524 525

AC C

EP

TE D

515

With a lack of consistency in approaches to recycling (what materials are collected and how), there is a serious need for ongoing education. The consensus among workshop participants was that not

24

ACCEPTED MANUSCRIPT everything should be recycled, despite societal pressures. There are challenges with dye contaminants

527

among others, and it is paramount to consider the entire life cycle by applying LCAs. The possibility that

528

contaminants in plastic materials intended for recycling may remain in the recycled material and could

529

migrate into food is one of the major considerations for the safe use of recycled plastics, illustrating the

530

need for the development of new recycling technologies. Emerging innovations in packaging, such as

531

smart packaging, will continue to grow, as additional functionality will create new paradigm in terms of

532

safety, security, traceability, sustainability, marketing, and shelf-life of products.

533

Acknowledgments

534

The authors thank the speakers for reviewing the manuscript.

535

Funding: The workshop was developed and supported by the ILSI North America Food and Chemical

536

Safety Committee. Agnes Karmaus received funds from the ILSI North America Food and Chemical Safety

537

Committee for her work on this article.

538

Competing interests statement: The author(s) declared the following potential conflicts of interest with

539

respect to the research, authorship, and/or publication of this article: Ron Osborn is an employee of

540

Mars Wrigley Confectionery.

541

References

542

Burek, J., et al., 2018. Environmental sustainability of fluid milk delivery systems in the United States. J.

SC

M AN U

TE D

EP

AC C

543

RI PT

526

Ind. Ecol. 22, 180-195.

544

Dhaliwal, H., et al., 2014. A life cycle assessment of packaging options for contrast media delivery:

545

comparing polymer bottle vs. glass bottle. Int. J. Life Cycle Assess. 19, 1965-1973.

546

EFSA, 2014. Scientific Opinion on the safety assessment of the substance, 2,4,8,10-

547

tetraoxaspiro[5.5]undecane-3,9-diethanol, β3,β3,β9,β9-tetramethyl-, CAS No 1455-42-1, for use

548

in food contact materials. EFSA J. 12, 3863. 25

ACCEPTED MANUSCRIPT 549 550 551

European Commission Joint Research Centre, 2018. The FACET Project. http://expofacts.jrc.ec.europa.eu/facet/. European Parliament, 2016. Food Contact Materials - Regulation (EC) 1935/2004. http://www.europarl.europa.eu/RegData/etudes/STUD/2016/581411/EPRS_STU(2016)581411_

553

EN.pdf.

554

FDA, 1938. Federal Food, Drug, and Cosmetic Act (FD&C Act).

RI PT

552

https://www.fda.gov/RegulatoryInformation/LawsEnforcedbyFDA/FederalFoodDrugandCosmeti

556

cActFDCAct/default.htm.

558 559

FDA, 2002a. Guidance for Industry: Preparation of Food Contact Notifications (Administrative). https://www.fda.gov/Food/GuidanceRegulation/ucm081818.htm.

M AN U

557

SC

555

FDA, 2002b. Guidance for Industry: Preparation of Food Contact Notifications for Food Contact Substances (Toxicology Recommendations).

561

https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/In

562

gredientsAdditivesGRASPackaging/ucm081825.htm.

563

TE D

560

FDA, 2005. Guidance for Industry: Submitting Requests Under 21 CFR 170.39 Threshold of Regulation for Substances Used in Food-Contact Articles.

565

https://www.fda.gov/RegulatoryInformation/Guidances/ucm081833.htm.

567 568 569 570

FDA, 2006a. Guidance for Industry: Preparing a Claim of Categorical Exclusion or an Environmental Assessment for Submission to the Center for Food Safety and Applied Nutrition.

AC C

566

EP

564

https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/In

gredientsAdditivesGRASPackaging/ucm081049.htm.

FDA, 2006b. Guidance for Industry: Use of Recycled Plastics in Food Packaging (Chemistry

571

Considerations).

572

https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/In

573

gredientsAdditivesGRASPackaging/ucm120762.htm. 26

ACCEPTED MANUSCRIPT 574

FDA, 2007. Guidance for Industry: Preparation of Premarket Submissions for Food Contact Substances (Chemistry Recommendations).

576

https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/In

577

gredientsAdditivesGRASPackaging/ucm081818.htm.

578 579 580

RI PT

575

FDA, 2012. Inventory of Effective Food Contact Substance (FCS) Notifications: FCN No. 1135, Mitsubishi Gas Chemical Company, Inc. https://www.accessdata.fda.gov/scripts/fdcc/?set=fcn&id=1135. FDA, 2015. Acceptance Criteria for Confirmation of Identity of Chemical Residues Using Exact Mass Data for the FDA Foods and Veterinary Medicine Program.

582

https://www.fda.gov/downloads/ScienceResearch/FieldScience/UCM491328.pdf.

585 586 587 588 589

M AN U

584

FDA, 2016a. 3M Corporation; Filing of Food Additive Petition: 21 CFR Part 176 (81 FR 25625). Fed. Regist. 81, 25625-25627.

FDA, 2016b. Substances Generally Recognized as Safe: 21 CFR Parts 20, 25, 170, 184, 186, and 570 (81 FR 5490). Fed. Regist. 81, 54959-55055.

FDA, 2017a. Code of Federal Regulations Title 21.

TE D

583

SC

581

https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=174.5. FDA, 2017b. Update on Perfluorinated Grease-Proofing Agents. https://www.fda.gov/Food/IngredientsPackagingLabeling/PackagingFCS/Notifications/ucm3084

591

62.htm.

593 594 595 596 597

FDA, 2018a. Food Ingredients and Packaging Terms.

AC C

592

EP

590

https://www.fda.gov/Food/IngredientsPackagingLabeling/Definitions/default.htm

FDA, 2018b. Ingredients, Additives, GRAS & Packaging Guidance Documents & Regulatory Information. http://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/Ing redientsAdditivesGRASPackaging/default.htm Han, H. J., 2005. Innovations in Food Packaging, 1st ed. Academic Press, San Diego, CA.

27

ACCEPTED MANUSCRIPT

599 600 601 602 603

ISO, 1998. ISO 14041:1998 Environmental Management -- Life Cycle Assessment -- Goal and Scope Definition and Inventory Analysis. https://www.iso.org/standard/23152.html. Kang, D., et al., 2013. Comparison of bacon packaging on a life cycle basis: a case study. J. Cleaner Prod. 54, 142-149.

RI PT

598

Koster, S., et al., 2015. Guidance on Best Practices on the Risk Assessment of Non-Intentionally Added Substances (NIAS) in Food Contact Materials and Articles. ILSI Europe, Brussels, Belgium. Oldring, P., et al., 2014a. FACET: Light metal packaging methodology. Coatings Tech. 11, 30-40.

605

Oldring, P. K., et al., 2014b. Estimates of dietary exposure to bisphenol A (BPA) from light metal

SC

604

packaging using food consumption and packaging usage data: a refined deterministic approach

607

and a fully probabilistic (FACET) approach. Food. Addit. Contam. Part A Chem Anal. Control

608

Expo. Risk Assess. 31, 466-89.

609

M AN U

606

Oldring, P. K., et al., 2014c. Development of a new modelling tool (FACET) to assess exposure to chemical migrants from food packaging. Food. Addit. Contam. Part A Chem Anal. Control Expo.

611

Risk Assess. 31, 444-65.

612

TE D

610

Olsen, G. W., et al., 2003. Epidemiologic assessment of worker serum perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) concentrations and medical surveillance examinations. J. Occup.

614

Environ. Med. 45, 260-70.

615

EP

613

Olsen, G. W., et al., 1999. Serum perfluorooctane sulfonate and hepatic and lipid clinical chemistry tests in fluorochemical production employees. J. Occup. Environ. Med. 41, 799-806.

617

Robertson, G., 2012. Food Packaging: Principles and Practice, 3rd ed. CRC Press, Boca Raton, FL.

618

Scientific Applications International Corporation, Curran, M. A., 2006. Life-Cycle Assessment: Principles

619

and Practice. National Risk Management Research Laboratory, Office of Research and

620

AC C

616

Development, US Environmental Protection Agency, Cincinnati, OH.

28

ACCEPTED MANUSCRIPT 621

Socolof, M. L., et al., 2001. Desktop Computer Displays: A Lifecycle Assessment, Vol. 1 (EPA-744-R-01-

622

004a). https://www.epa.gov/sites/production/files/2014-

623

01/documents/computer_display_lca.pdf. Sustainable Packaging Coalition, 2011. Definition of Sustainable Packaging.

RI PT

624 625

https://sustainablepackaging.org/wp-content/uploads/2017/09/Definition-of-Sustainable-

626

Packaging.pdf.

AC C

EP

TE D

M AN U

SC

627

29