Fiberglass and Other Flame-Resistant Fibers in Mattress Covers

Abstract

Public complaints have raised concerns that some mattresses in the current marketplace may be potential sources of airborne fiberglass. Although mattress foam is often marketed as chemical-free, their cover compositions are not as well understood by the general public. To fill these basic information gaps, the covers of four newly purchased mattresses were sampled and analyzed using polarized light microscopy, SEM-EDS, and FTIR microspectroscopy. Two of the mattress covers contained over 50% fiberglass in their inner sock layers. Up to 1% of the fiberglass had migrated to adjacent fabric layers, representing a potential risk of consumer exposure if the zipper on the outer cover is opened. The observed fiberglass fragments had calculated aerodynamic diameters ranging between 30 and 50 µm, suggesting they are potentially inhalable into the nose, mouth, and throat, but are likely too large to penetrate deeper into the lungs. No fiberglass was observed on the brand new mattresses’ outer surfaces. Synthetic fibers also present in the sock layers were consistent with flame resistant modacrylic containing vinyl chloride and antimony. The use of fiberglass and other chemicals in mattress covers poses a potential health risk if these materials are not adequately contained. The apparent non-inclusion of mattress covers in chemical-free certifications suggests that further improvements are needed in mattress labeling and education of consumers.

Full text



Citation: Wagner, J.; Fowles, J.;
Barreau, T. Fiberglass and Other
Flame-Resistant Fibers in Mattress
Covers. Int. J. Environ. Res. Public
Health 2022,19, 1695. https://
doi.org/10.3390/ijerph19031695
Academic Editor: Paul B.
Tchounwou
Received: 21 December 2021
Accepted: 28 January 2022
Published: 1 February 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
International Journal of
Environmental Research
and Public Health
Article
Fiberglass and Other Flame-Resistant Fibers in Mattress Covers
Jeff Wagner 1, * , Jefferson Fowles 2and Tracy Barreau 2
1
Environmental Health Laboratory Branch, California Department of Public Health, 850 Marina Bay Parkway,
Richmond, CA 94804, USA
2Environmental Health Investigations Branch, California Department of Public Health, 850 Marina Bay
Parkway, Richmond, CA 94804, USA; jeff.fowles@cdph.ca.gov (J.F.); tracy.barreau@cdph.ca.gov (T.B.)
*Correspondence: jeff.wagner@cdph.ca.gov
Abstract:
Public complaints have raised concerns that some mattresses in the current marketplace
may be potential sources of airborne fiberglass. Although mattress foam is often marketed as
chemical-free, their cover compositions are not as well understood by the general public. To fill these
basic information gaps, the covers of four newly purchased mattresses were sampled and analyzed
using polarized light microscopy, SEM-EDS, and FTIR microspectroscopy. Two of the mattress covers
contained over 50% fiberglass in their inner sock layers. Up to 1% of the fiberglass had migrated
to adjacent fabric layers, representing a potential risk of consumer exposure if the zipper on the
outer cover is opened. The observed fiberglass fragments had calculated aerodynamic diameters
ranging between 30 and 50
µ
m, suggesting they are potentially inhalable into the nose, mouth, and
throat, but are likely too large to penetrate deeper into the lungs. No fiberglass was observed on the
brand new mattresses’ outer surfaces. Synthetic fibers also present in the sock layers were consistent
with flame resistant modacrylic containing vinyl chloride and antimony. The use of fiberglass and
other chemicals in mattress covers poses a potential health risk if these materials are not adequately
contained. The apparent non-inclusion of mattress covers in chemical-free certifications suggests that
further improvements are needed in mattress labeling and education of consumers.
Keywords: flame retardants; fiberglass; mattresses; exposure assessment
1. Introduction
Fiberglass, sometimes also referred to as man-made vitreous fibers (MMVF), is a
known respiratory, skin and eye irritant and an asthmagen [
1
]. In occupational settings,
fiberglass exposure has been found to correlate with recurrent chest infections and pul-
monary fibrosis [
2
,
3
]. Fiberglass fibers can vary in diameter, length, and chemical compo-
sition, but are predominantly amorphous (non-crystalline) mixtures of oxides of silicon,
calcium, and other metals. Common commercial uses for fiberglass include insulation and
fire protection [4].
Prompted by historical concerns that chronic inhalation of these fibers could pose lung
cancer risks, the International Agency for Research on Cancer (IARC) reviewed the litera-
ture on at least two occasions, and concluded in 2001 that the evidence for carcinogenicity
from fiberglass exposure met the criteria for Group 3 (unclassifiable as to human carcino-
genicity) [
4
]. This conclusion was supported by a subsequent review of the occupational
epidemiology literature, including case–control, cohort, and meta-analyses, in 2011 [
5
]. The
carcinogenicity of a given fiberglass fiber depends on both its inhalability (aerodynamic
size) and its biopersistence (durability and clearance rate from the lung) [6].
Several anecdotal reports and numerous public complaints have raised concerns that
some mattresses in the current marketplace may be potential sources of airborne fiberglass
exposures. Children and infants represent a special potential risk group, both due to their
increased susceptibility and the possibility that children may play or jump on beds. In one
case investigated by CDPH in 2021, a 6-year old child was found to have persistent skin and
Int. J. Environ. Res. Public Health 2022,19, 1695. https://doi.org/10.3390/ijerph19031695 https://www.mdpi.com/journal/ijerph

Int. J. Environ. Res. Public Health 2022,19, 1695 2 of 12
respiratory irritation linked to the suspected leakage of fiberglass from a mattress purchased
in 2018. The resulting clean-up of the home included disposal of the mattress, carpet, and
clothing items from which the fiberglass fibers could not be removed by cleaning. A search
of the Consumer Product Safety Commission website found 128 complaints from 1 January
to 31 December 2020 linked to fiberglass exposure in several different brands of mattresses,
and news media have reported on a growing number of relevant health complaints [7,8].
The inclusion of flame retardants and flame-resistant fibers in mattresses is primarily
driven by California and US flammability regulations, despite the fact that their use in
products poses a potential health risk [
9
]. Changes in California furniture flammability
testing rules have shifted emphasis from the foam to their covers [
10
], so the prevalence of
these additives in current mattress covers is an important unknown. “Certi-PUR-US” is
an industry-based certification program that designates foam products to be free of heavy
metals, PBDEs, TDCPP or TCEP (“Tris”) flame retardants, as well as numerous flame-
retardant additives [
11
]. Consumers are likely to believe CertiPUR-US-certified mattresses
have undergone rigorous testing and are free of hazardous substances. However, the
certification and testing do not appear to include the mattress covers.
The aim of this project was to fill basic information gaps about the physical/chemical
nature of fibers in new mattress covers by analyzing them with microscopy and microspec-
troscopy. We purchased four new mattresses including one child mattress, and sampled
and analyzed their covers using polarized light microscopy (PLM), scanning electron mi-
croscopy with X-ray energy dispersive spectroscopy (SEM/EDS), and Fourier transform
infrared (FTIR) microspectroscopy. We report on the basic structure of these covers, the
size and chemical characteristics of the detected fiberglass materials, and the additional
flame-resistant fibers observed in the analyses.
2. Materials and Methods
The mattresses were purchased and delivered from common online sales outlets in
April and May of 2021. The four purchased mattresses were a Sealy Essentials Joyfulness
8.5” twin (Referred to hereafter as mattress FG-1), a Modway Aveline 6” twin (FG-2), a Zinus
Green Tea 6” memory foam twin (FG-3), and a Graco crib & toddler deluxe (FG-4). The
polyurethane foam in all of these products was advertised as being certified “Certi-PUR-US”
(see Supplementary Materials Table S1).
The constituent components of the four mattress covers were each subsampled by
cutting through each observed layer with a clean, stainless-steel scalpel, then collecting the
pieces into small, labeled, zipper-lock bags. Each piece was subsampled through its entire
layer depth, with square cross-sectional areas of approximately 1 cm
2
. No foam was tested
in this work.
Sample preparation and analysis was conducted using an in-house standard operat-
ing procedure for analyzing the fibrous content of bulk materials, based on EPA 600/R-
93/116 [
12
]. A combination of subsample PLM and whole-sample stereozoom analyses
was used to determine component fractions via calibrated visual estimation and species-
specific optical properties, supplemented with SEM-EDS and ash analyses for confirmation.
The reporting limit for this procedure is 1%, and species detected at less than this con-
centration have been reported as trace (<1%). Although this SOP is oriented towards
determining asbestos content, it also includes specific procedures for identifying fiberglass
and synthetic fibers. CDPH is accredited in the use of this SOP, participates in quarterly
proficiency testing, and maintains a library of over 300 consumer and building material
reference samples.
All subsamples were initially inspected and documented with a Leica S8APO reflected-
light stereozoom microscope and DFC420 digital camera (Leica, Wetzlar, Germany) at
6.3–80×
magnification. This work focused on the non-foam cover components as the most
likely sources of fibers, but the foam components were archived for potential
future analyses
.
PLM was conducted using a Nikon E-600POL and DS-Ri1 digital camera with DS-
Ri1-U2 controller (Nikon Instruments, Melville, NY, USA) at magnifications of 40
×
, 100
×
,

Int. J. Environ. Res. Public Health 2022,19, 1695 3 of 12
200
×
, and 400
×
. Slide mounts were prepared with 1.680 refractive index oil (Cargille
Laboratories, Cedar Grove, NJ, USA) which was certified annually. All images in this
report were acquired using crossed polarizers and a full-wave (red one) compensator.
Under these imaging conditions, fiberglass can be positively identified by an absence of
interference colors and total extinction (indicating an isotropic index of refraction and
amorphous material), in conjunction with an elongated aspect, blunt fiber end terminations,
and a uniform cross-section [
13
]. Fiberglass often possesses a straight, rod-like morphology,
but some varieties of silicaceous fibers (more often called mineral or rock wool) also
exhibit curved fibers. Imaging was also conducted using plane polarized light, crossed
polarizers with no compensator, and dispersion staining conditions, respectively, to check
for other components.
When no fiberglass was detected in any samples for a given mattress cover, all con-
stituent samples were ashed in a muffle furnace at 485
◦
C for 3 h. This procedure removes
any polymeric, combustible, and lower-boiling point components, leaving only non-volatile
remnants such as fiberglass, other minerals, and ash.
A fiberglass standard (NIST Standard Reference Material SRM 1866a, Gaithersburg,
USA) was analyzed as a reference. This SRM contains a range of glass fiber types and
thicknesses ranging from 1–10 µm.
A blank sample containing non-fiberglass synthetic fibers was prepared and analyzed
along with these samples to check for any fiberglass contamination in the laboratory.
Analysis of this blank yielded no fiberglass fibers. This blank also contained crystalline
silica (quartz) particles and was used as a reference for the presence of any quartz in
the samples.
Supplemental SEM/EDS was conducted using an FEI XL30 Environmental SEM
with Noran System 7 EDS (Thermo Fisher Scientific, Madison, WI, USA) and SEMView
8000 electronics
(SEMTech Solutions, North Billerica, MA, USA). Bulk samples were pre-
pared on standard aluminum SEM stubs with double-sided, carbon adhesive tabs and were
carbon-coated to minimize charging. Limited “tape lift” samples were also collected from
the outer covers using one SEM stub/tab per mattress. Samples were imaged at 20 kV and
20–2500
×
using a tungsten source and an atomic number-sensitive back-scattered electron
(BSE) detector for rapid screening differentiation of non-carbonaceous materials.
Limited synthetic fiber and flame retardant identification was performed using a
Nicolet iN-10MX FTIR Microscope with Omnic Picta software and a cooled MCT-A detec-
tor (Thermo Fisher Scientific, Madison, USA). Spectra were acquired from fiber strands
with
4 cm−1
spectral resolution from 700–4000 cm
−1
in transmission mode mounted on a
standard 3-hole slide. Identifications of component mixtures were assigned manually with
the aid of quantitative FTIR spectral searches and in-house and commercial libraries of
over
100,000 spectra
. The estimated limit of detection for flame retardants and other minor
constituents of polymer materials is in the order of 1% [14].
3. Results
3.1. Main Components of Mattress Covers
The main components of each tested mattress cover and their observed compositions
are summarized in Figure 1and Table S2. Three of the four mattresses (FG-2, -3, and
-4) possessed zippers on their outer covers. The various mattress cover components of
FG-1 and FG-2 were mixtures of synthetic fibers, mostly in the form of woven materials
(Figures 2and 3). The mattress FG-1 additionally contained fibrous mat layers beneath
the top cover and inside the bottom cover, as well as a synthetic coating (Figure 2). FG-2
also contained cellulosic (natural) fibers. The outer cover of FG-3 on the back side of the
mattress (Figure 4b) and the inner cover of FG-4 (Figure 5c) were both pre-formed sheets
composed of organic binder and synthetic fibers, and the outer cover of FG-4 had a solid
plastic sheet bonded to its inner surface (Figure 5b).

Int. J. Environ. Res. Public Health 2022,19, 1695 4 of 12
Figure 1. Main components of each tested mattress and observed compositions.
Figure 2.
Low-power reflected-light stereozoom microscope images of FG-1 mattress subsamples
acquired at 10
×
. (
a
) Synthetic, woven outer cover material from top side of mattress. (
b
) Synthetic,
fibrous mat beneath outer cover material from top side of mattress. (
c
) Synthetic, woven outer cover
material from side of mattress. (
d
) Outer cover material from bottom of mattress with fibrous mat,
foam coating, and woven fibers.

Int. J. Environ. Res. Public Health 2022,19, 1695 5 of 12
Figure 3.
Low-power reflected-light stereozoom microscope images of FG-2 mattress subsamples
acquired at 10
×
. (
a
–
c
) Three sections of the synthetic, woven outer cover materials, including
fiber-filled bead (b). (d) Inner cover material with natural, woven fibers.
Figure 4.
Low-power reflected-light stereozoom microscope images of FG-3 mattress subsamples
acquired at 10
×
. (
a
) White, synthetic, woven outer cover material from top side of mattress. (
b
) Beige
outer cover material from bottom side of mattress consisting of synthetic fibers in polymeric binder.
(
c
) Inner sock material, showing woven threads, each uniformly wrapped in white synthetic fibers.
(d) Same as (c), but inner fiberglass core revealed inside each thread.

Int. J. Environ. Res. Public Health 2022,19, 1695 6 of 12
Figure 5.
Low-power reflected-light stereozoom microscope images of FG-4 mattress subsamples
acquired at 10
×
. (
a
) White, synthetic, woven outer surface of outer cover material. (
b
) White,
synthetic inner plastic sheet surface of outer cover material. (
c
) Inner cover material consisting of
synthetic fibers in polymeric binder, and short straight fiberglass fibers sticking into surface (indicated
by arrows). (
d
) Inner sock material, showing woven threads consisting alternately of white synthetic
fibers and shiny/clear fiberglass bundles.
In addition to these polymeric cover materials, FG-3 and FG-4 both possessed inner
“sock” layers that were accessible beneath a zipper in their outer covers. The sock compositions
of both were dominated by straight bundles of fiberglass (Figures 4,5and S1). PLM confirmed
no fiberglass content in FG-1 or FG-2, but the positive detection of fiberglass for FG-3 and
FG-4 (Figures 6and S2–S6). The fiberglass bundles in FG-3 and FG-4 were intermixed to
different degrees with synthetic (polymeric) fibers.
Other than polymeric binders, the non-fibrous components of the tested materials
were minimal, on the order of 1%. No crystalline silica (quartz) particles were observed in
these samples by PLM.
3.2. Fiberglass Characteristics
In sample FG-3, each fiberglass bundle was completely encased in a synthetic fiber
sheath layer such that no fiberglass was visible except at the cut ends, and the cloth
maintained much of its structural integrity after being cut. This sock material was approxi-
mately 50% fiberglass by mass. In FG-4, however, bare fiberglass bundles were interwoven
separately with less numerous woven synthetic fiber strands. This sock material was
approximately 75% fiberglass by mass and was observed to visibly degrade and shed
fiberglass pieces, especially where it was cut. Although cutting does not represent normal
bed usage, the FG-4 sock’s degradation (Figure 6d) suggests a relatively lower structural
integrity compared to sample FG-3.
The 400
×
PLM imaging revealed the fiberglass fibers in FG-3 and FG-4 to be fairly
uniform, straight, and 5–10
µ
m in thickness (Figures 6f and S6), consistent with previous
surveys of fiberglass in textiles [
1
] and the thickest fibers in the NIST SRM mixture. The
longest intact fiberglass bundles were very long, comparable to the sizes of subsamples
removed for analysis (3–5 cm), and may have been even longer in the original material.
However, the numerous fiberglass fragments shed from FG-4 during inspection were much
shorter, many approximately 2 mm long. The shortest fiberglass fragments observed at
400x PLM were approximately 50 µm long.

Int. J. Environ. Res. Public Health 2022,19, 1695 7 of 12
Figure 6.
(
a
) Fiberglass (straight) in FG-3 inner sock material with synthetic fibers (wavy) (
100×PLM
).
(
b
) Rare fiberglass fragment in FG-3 bottom, outer cover material (100
×
PLM). (
c
) Fiberglass in
FG-4 inner sock (80
×
stereozoom). (
d
) Fiberglass shed onto foil surface after FG-4 inspection
(
10×stereozoom
). (
e
) Fiberglass in FG-4 inner cover material with synthetic fibers (100
×
PLM).
(f) Higher magnification of fiberglass morphology from FG-4 inner sock (400×PLM).
SEM/EDS analyses of the inner sock layers from samples FG-3 and FG-4 confirmed
the same fiberglass morphology as was observed with PLM and 5–10 um-thick fibers
(
Figures 7and 8
). The fiberglass in both samples exhibited the typical mineral fiber compo-
sition of primarily silicon (Si), aluminum (Al), and calcium (Ca).
Analyses of the cloth layers adjacent to the fiberglass layers in samples FG-3 and FG-4
revealed instances of loose fiberglass adhered to them, likely due to migration from the
inner socks. For FG-3, only a trace amount of fiberglass (5 fibers) was found adhered to the
bottom outer cover (Figure 6b), and no fiberglass was observed in the top cover. Adhered
fiberglass fragments were much more commonly found adhering to the inner cover of FG-4
(Figures 5c and 6e), representing approximately 1% of the material by mass.
No fiberglass was observed adhered to the outer cover of FG-4, however. It is possible
that the migration of fiberglass from the inner cover was prevented to some extent by the
solid polymer sheet sublayer of the outer cover.

Int. J. Environ. Res. Public Health 2022,19, 1695 8 of 12
Figure 7.
SEM-EDS from FG-3 inner sock. (
a
,
b
) Cut bundle of fiberglass wrapped in synthetic fibers.
(
c
) One fiberglass fiber and two synthetic fibers with characteristic bright regions. (
d
) EDS acquired
from fiberglass region marked in (
c
) showing Si, Al, and Ca. (
e
) EDS from bright spot in synthetic
fiber marked in (c) exhibiting Cl and Sb.
Figure 8.
SEM-EDS from FG-4 inner sock. (
a
) Cut bundle of fiberglass and separate bundle of
synthetic fibers. (
b
) Single fiberglass fibers with blunt ends and 8 um thickness. (
c
) Fiberglass fibers
and synthetic fibers. (
d
) EDS acquired from fiberglass region marked in (
c
) showing Si, Al, and Ca.
(e) EDS from bright spot in synthetic fiber marked in (c) exhibiting Cl and Sb.

Int. J. Environ. Res. Public Health 2022,19, 1695 9 of 12
3.3. Other Flame Resistant Fibers
The ashing of samples from FG-1 and FG-2 resulted in near-total volatilization of
most components at 485
◦
C, confirming they did not possess major fiberglass content like
FG-3 and FG-4. Some non-flammable fibers remained in FG-1, however. The fibrous mat
material from the top and bottom of mattress FG-1 survived 485
◦
C, with little change in
appearance (Figure S2c,f), and the fiber coating in the bottom outer cover also remained
(Figure S2f). Only charred, shriveled fiber remnants from the inner cover survived the
ashing of FG-2 (Figure S3c,d). The limited FTIR of fibers from FG-1 variously matched
polyester, polypropylene, and viscose (Rayon), respectively (Figure S7). Additional small
peaks matching antimony trioxide and silicic acid were inconclusive, but may help explain
these fibers’ persistence at high temperatures.
SEM/EDS BSE of synthetic fibers in both FG-3 and FG-4 revealed them to be carbona-
ceous, with sub-micron, brighter inclusions characteristic of manufactured polymers with
inorganic additives (Figures 7and 8). EDS acquired from these bright regions revealed
enriched chlorine (Cl) and antimony (Sb), consistent with a flame-retardant product with
metal synergists [
15
–
17
]. FTIR analyses of these synthetic fibers from the inner sock layers
of FG-3 and FG-4 matched library spectra for modacrylics with vinyl chloride and antimony
trioxide flame-retardant additives (Figure 9). Although these FTIR peaks were also small,
the detection of Cl and Sb with SEM-EDS confirms the presence of flame retardants in the
FG-3 and FG-4 fibers. Additional peaks in the FTIR spectra from FG-3 suggest they were
blended with PET (polyester terephthalate) fibers. In general, most of the detected polymer
fiber types in the covers of FG-1, FG-3, and FG-4 were listed on their tags, but with no
flame-retardant specification (Table S1).
Figure 9.
(
a
) FTIR microscope image of analyzed area of 200 um thickness, synthetic fiber bundle
from inner sock material in FG-4. (
b
) FTIR microspectroscopy results (red spectrum) from (
a
). Also
shown is a spectrum from the inner sock of FG-3 (orange). Both are good matches with the library
spectra for modacrylic fibers containing vinyl chloride and antimony trioxide additives (blue). In
addition, the synthetic fibers from FG-3 contain minor peaks matching PET (green).

Int. J. Environ. Res. Public Health 2022,19, 1695 10 of 12
4. Discussion
Our investigation confirmed the presence of fiberglass in two of the four tested mattress
covers. The fiberglass was evident in layers that were accessible beneath outer zipper layers
(Figure 1), presenting a relatively easy route of exposure. The presence of fiberglass
was not disclosed on the labels of the FG-4 mattress, leaving consumers unaware of its
presence. Fiberglass was disclosed on the label of FG-3, though not in the most prominent
flammability section (“law label”), which is required to specify filling materials only [10].
At minimum, any exposure to fiberglass represents a potential dermal irritation, as
well as an eye irritation hazard [
4
,
5
]. However, the potential of fiberglass to additionally
represent an inhalation hazard depends on whether the material is aerosolizable and the
range of potential fiber widths and lengths [
4
]. As described in the Results, both FG-3
and FG-4 contained very long fiberglass fibers, but these were observed to fragment into
smaller pieces when cut or disturbed, particularly in the case of FG-4.
The effective aerodynamic size, d
a
, for fibers in an oriented fluid flow can be calculated
using an assumed density of ρf= 2.5 g/cm3[18], and the following expression [19,20]:
da= W ×(9/4 ×ρf/ρ0×[ln(2L/W) −0.807])0.5 (1)
where W = fiber diameter,
ρ0
= unit density, and L = fiber length. For the smallest observed
fiberglass fragments (50
µ
m long and 5
µ
m wide), d
a
= 30
µ
m. For the commonly observed,
longer fiberglass fragments (2 mm long and 10 µm wide), da= 50 µm.
Standard inhalability curves predict substantial head airway (nose, mouth, pharynx,
and larynx) deposition for particles with approximately da = 10–100 um, while standard
thoracic and respirable curves predict enhanced tracheobronchial and alveolar lung deposi-
tion for d
a
< 10
µ
m and 2.5
µ
m, respectively [
21
]. These curves suggest that the observed
fiberglass fragments from these covers are potentially inhalable into head airways, but are
likely too large to penetrate deeper into the chest or lungs. As such, they are potentially a
nasal and throat irritant and asthma trigger, but unlikely to be a lung hazard. Note that
exposure to extreme environments, solvents, or mechanical damage during manufacture or
product aging may alter fiberglass friability, or further reduce fragment sizes from those
observed here.
FTIR micro-spectroscopy determined that the synthetic fibers in the fiberglass socks
were additional sources of flame-resistant fibers (e.g., modacrylics), and SEM/EDS con-
firmed they also contained halogenated compounds and metal synergists (antimony tri-
oxide). It is not yet known to what degree these flame-retardant additives may represent
additional risks thorough dermal, oral, or inhalation exposure routes [
17
,
22
]. However,
many of these compounds are prohibited under the CertiPUR-US certification of the foam,
which may lead consumers to believe they are not present in the mattress.
Similarly, flammability tags attached to the mattress covers containing fiberglass
and antimony-containing fibers (FG-3 and FG-4) contained statements such as “100%
polyurethane foam pad”, suggesting they were referring to the foam cores only. California
law currently requires the labeling of flame retardants in specific product types if they
represent >0.1% of the product mass [
17
], but exempts some mattress types and their
non-foam components [
10
,
22
]; flame retardants in non-foam mattress components must
be labeled if they are for toddlers or infants [
23
]. Accordingly, the only mattress in this
study that possessed such a label (FG-4) was a crib mattress, but it listed “no added flame
retardant chemicals”, despite the presence of fiberglass and modacrylic fibers in the cover.
In this regard, it is unclear whether fiberglass is considered a flame-retardant chemical
using current guidelines [
24
,
25
]. In general, the possibility that none of these mattress cover
contents are included in current chemical-free certifications suggests that improvements
are still needed in mattress certifications, labeling, and the education of consumers.
This work was limited to four mattress covers obtained from the current marketplace.
Future analyses of other mattress brands and models would be useful to determine whether
these findings apply to mattress covers in general. The observations of inner sock friability

Int. J. Environ. Res. Public Health 2022,19, 1695 11 of 12
during the laboratory analyses are suggestive of differential fiberglass exposures between
brands and sock designs. More systematic experiments of fiberglass shedding under
realistic environments and mechanical stresses are needed, and could be combined with
standard methods to measure fiberglass fragments in air [
26
], dust [
16
,
27
], or eye or nose
mucous samples [
28
,
29
]. This work focused on the presence of fiberglass and other flame-
resistant fibers in mattress covers, but future analyses of the associated foam materials
would provide a useful complement to this information. The extent to which our results
raise a generalizable issue related to fiberglass exposure from other consumer products
that meet flammability standards is unknown, but worthy of investigation.
5. Conclusions
We determined the basic structures and compositions of select, new mattress covers
using various microspectroscopy analyses. Fiberglass was observed in two of the four
covers, including potentially inhalable fiberglass fragments that pose a health risk if the
covers are opened by consumers. Further, undisclosed chemical additives were observed,
including modacrylics containing antimony and vinyl chloride, and these could present
additional exposure and health risks. Although flame-resistant fibers are used in mattress
covers to pass flammability regulations, their compositions are uncertain on labels that may
describe the foam contents only. The apparent omission of mattress covers from the criteria
for chemical-free certifications of mattresses suggests that improvements are needed in
mattress labeling and the education of consumers.
Supplementary Materials:
The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/ijerph19031695/s1, Table S1: Mattress labels, certifications, and
cover fibers identified; Table S2: Summary of compositions determined from mattress cover analyses;
Figure S1: Additional 10–80
×
images of FG-4 inner sock; Figure S2: PLM images of FG-1.; Figure
S3: PLM images of FG-2; Figure S4: PLM images of FG-3; Figure S5: PLM images of FG-4; Figure S6:
400×PLM images of fiberglass morphology; Figure S7: FTIR from three FG-1 fibers.
Author Contributions:
Conceptualization, T.B., J.F. and J.W.; investigation, T.B., J.F. and J.W.; method-
ology, T.B., J.F. and J.W.; formal analysis, J.W.; resources, T.B., J.F. and J.W.; visualization, J.W.;
writing—original draft preparation, J.W.; writing—review and editing, T.B., J.F. and J.W. All authors
have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement:
All data used to form conclusions are contained within the paper and
Supplementary Materials.
Acknowledgments:
The findings and conclusions in this article are those of the authors and do not
necessarily represent the views or opinions of the California Department of Public Health or the
California Health and Human Services Agency.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
NIOSH. Criteria for a Recommended Standard: Occupational Exposure to Fibrous Glass—DHHS (NIOSH) No. 77. 1997. Available
online: https://www.cdc.gov/niosh/docs/77-152/default.html (accessed on 31 January 2022).
2.
Moitra, S.; Tabrizi, A.F.; Machichi, K.I.; Kamravaei, S.; Miandashti, N.; Henderson, L.; Mukherjee, M.; Khadour, F.; Naseem, M.T.;
Lacy, P.; et al. Non-Malignant Respiratory Illnesses in Association with Occupational Exposure to Asbestos and Other Insulating
Materials: Findings from the Alberta Insulator Cohort. Int. J. Environ. Res. Public Health 2020,17, 7085. [CrossRef] [PubMed]
3.
Fireman, E. Man-made mineral fibers and interstitial lung diseases. Curr. Opin. Pulm. Med.
2014
,20, 194–198. [CrossRef]
[PubMed]
4.
IARC (International Agency for Research on Cancer). Man-Made Vitreous Fibers, Volume 81. Available online: https:
//publications.iarc.fr/Book-And-Report-Series/Iarc-Monographs-On-The-Identification-Of-Carcinogenic-Hazards-To-
Humans/Man-made-Vitreous-Fibres-2002 (accessed on 31 January 2022).

Int. J. Environ. Res. Public Health 2022,19, 1695 12 of 12
5.
Marsh, G.M.; Buchanich, J.M.; Youk, A.O. Fiber glass exposure and human respiratory system cancer risk: Lack of evidence
persists since 2001 IARC re-evaluation. Regul. Toxicol. Pharmacol. 2011,60, 84–92. [CrossRef] [PubMed]
6.
NTP (National Toxicology Program). Report on Carcinogens, 14th ed.; Department of Health and Human Services, Public
Health Service: Research Triangle Park, NC, USA, 2021. Available online: https://ntp.niehs.nih.gov/go/roc14 (accessed on
31 January 2022).
7.
Consumer Product Safety Commission (CPSC). Safer Products Public Search Tool. Available online: https://www.saferproducts.
gov/PublicSearch (accessed on 31 January 2022).
8.
CBS KMOV4. News 4 Investigates: The Danger Lurking in Your Mattress. 2020. Available online: https://www.kmov.com/
news/news-4-investigates-the-danger-lurking-in- your-mattress/article_2bb77cc2-4eb7- 11ea-b4c6- 0319ba20d45f.html (accessed
on 31 January 2022).
9.
Stapleton, H.M.; Klosterhaus, S.; Keller, A.; Ferguson, P.L.; van Bergen, S.; Cooper, E.; Webster, T.F.; Blum, A. Identification of
Flame Retardants in Polyurethane Foam Collected from Baby Products. Environ. Sci. Technol.
2011
,45, 5323–5331. [CrossRef]
[PubMed]
10.
BHGS (California Bureau of Household Goods and Services). Technical Bulletin 117- Residential Upholstered Furniture Standard
Fact Sheet. 2013. Available online: https://bhgs.dca.ca.gov/industry/tb_117_faq_sheet.pdf (accessed on 31 January 2022).
11.
AFPF (Alliance for Flexible Polyurethane Foam Inc.). CertiPur-US Technical Guidelines for Molded Foam. 20 November 2020.
Available online: https://certipur.us/technical-guidelines (accessed on 31 January 2022).
12.
U.S. Environmental Protection Agency. Method for the Determination of Asbestos in Bulk Building Materials, U.S. Environ-mental
Protection Agency Method 600/R-93/116; U.S. Environmental Protection Agency: Washington, DC, USA, 1993.
13. McCrone, W.C. Asbestos Identification; McCrone Research Institute: Chicago, IL, USA, 1987.
14.
Wagner, J.; Wang, Z.-M.; Ghosal, S.; Murphy, M.; Wall, S.; Cook, A.-M.; Robberson, W.; Allen, H. Nondestructive Extraction and
Identification of Microplastics from Freshwater Sport Fish Stomachs. Environ. Sci. Technol.
2019
,53, 14496–14506. [CrossRef]
[PubMed]
15.
van Esch, G.J. FRs: A General Introduction. Report No.: Environmental Health Criteria; World Health Organization: Geneva,
Switzerland, 1997.
16.
Wagner, J.; Ghosal, S.; Whitehead, T.; Metayer, C. Morphology, spatial distribution, and concentration of flame retardants in
consumer products and environmental dusts using scanning electron microscopy and Raman micro-spectroscopy. Environ. Int.
2013,59, 16–26. [CrossRef] [PubMed]
17.
Petreas, M.; Gill, R.; Takaku-Pugh, S.; Lytle, E.; Parry, E.; Wang, M.; Quinn, J.; Park, J.-S. Rapid methodology to screen flame
retardants in upholstered furniture for compliance with new California labeling law (SB 1019). Chemosphere
2016
,152, 353–359.
[CrossRef] [PubMed]
18.
NIOSH, Fibrous Glass Dust, Pocket Guide to Chemical Hazards. Available online: https://www.cdc.gov/niosh/npg/npgd0288.
html (accessed on 31 January 2022).
19.
Baron, P. Measurement of asbestos and other fibers. In Aerosol Measurement: Principles, Techniques, and Applications; Willeke, K.,
Baron, P., Eds.; Van Nostrand Reinhold: New York, NY, USA, 1993; pp. 560–590.
20. Cox, R. The motion of long slender bodies in a viscous fluid Part General theory. J. Fluid Mech. 1970,44, 791–810. [CrossRef]
21. Hinds, W. Aerosol Technology; John Wiley and Sons: New York, NY, USA, 1982.
22.
National Research Council Toxicological Risks of Selected Flame-Retardant Chemicals; National Academies Press: Washington, DC,
USA, 2000.
23.
SFDE (San Francisco Department of the Environment). FAQ for Retailers Selling Upholstered Home Furniture in San Francisco,
Including Information about Flame Retardant Chemicals. 2018. Available online: https://sfenvironment.org/sites/default/files/
fliers/files/sfe_th_flame_retardants_faq_for_retailers.pdf (accessed on 31 January 2022).
24.
BHGS. AB 2998 Frequently Asked Questions. 2019. Available online: https://bhgs.dca.ca.gov/forms_pubs/ab2998_faq.pdf
(accessed on 31 January 2022).
25.
ISPA (International Sleep Products Association). Survey of FR Fibers and Materials Used by Mattress Producers to Meet Federal
Flammability Standards. 2020. Available online: https://bhgs.dca.ca.gov/forms_pubs/20200131_frsurveyreport.pdf (accessed
on 31 January 2022).
26.
NIOSH. Method 7400 Issue 3, Asbestos and Other Fibers by PCM. 2019. Available online: https://www.cdc.gov/niosh/nmam/
pdf/7400.pdf (accessed on 31 January 2022).
27.
Webster, T.F.; Harrad, S.; Millette, J.R.; Holbrook, R.D.; Davis, J.M.; Stapleton, H.M.; Allen, J.G.; McClean, M.; Ibarra, C.; Abdallah,
M.A.-E.; et al. Identifying Transfer Mechanisms and Sources of Decabromodiphenyl Ether (BDE 209) in Indoor Environments
Using Environmental Forensic Microscopy. Environ. Sci. Technol. 2009,43, 3067–3072. [CrossRef] [PubMed]
28.
ATSDR (Agency for Toxic Substances and Disease Registry). Toxicological Profile for Synthetic Vitreous Fibers; ATSDR: Atlanta, GA,
USA, 2004.
29.
Paananen, H.; Holopainen, M.; Kalliokoski, P.; Kangas, J.; Kotilainen, M.; Pennanen, S.; Savolainen, H.; Tossavainen, A.; Luoto,
K. Evaluation of Exposure to Man-Made Vitreous Fibers by Nasal Lavage. J. Occup. Environ. Hyg.
2004
,1, 82–87. [CrossRef]
[PubMed]