ORIGINAL PAPER

Suspension and resuspension of dry soil indoors followingtrack-in on footwear

Andrew Hunt • David L. Johnson

Received: 3 January 2011 / Accepted: 7 July 2011 / Published online: 19 July 2011

� Springer Science+Business Media B.V. 2011

Abstract Contamination of the indoor environment

by tracked-in outdoor soil has the potential to pose a

significant human health threat through exposure

to hazardous soil constituents. The indoor distribution

of (contaminated) soil following ingress is important

when evaluating exposure risk. Here, the time evolu-

tion of size-resolved airborne particulate matter

aerosolized as a result of mechanical (i.e., footfall or

step-on) impacts on a floor surface with a layer of dry

soil was investigated using laser particle counters.

Suspended particle levels were recorded after step-on

impacts that aerosolized soil particles at a single

contact point by the action of a human tester who

followed a pre-determined walking pattern. The

experimental design presumed that the floor area

immediately upon entrance indoors is the location of

maximum deposition of outdoor soil transferred on

footwear. The suspension of soil resulting from the

first step-on floor contact and the subsequent resus-

pension of soil resulting from additional step-on

events were quantified by various arrangements of

four laser particle counters. Step-on impacts produced

a transient increase in particle levels at various lateral

distances and heights from the contact point. Also,

with increasing distance and height from the step-on

contact point, the level of suspended particles after

successive step-on events decreased markedly. The

results suggested that a lateral component of the

dispersion process was more significant than a vertical

one under these experimental conditions. A wall

jet effect created by the impact of the footfalls on

the floor surface was considered responsible for the

apparent greater lateral dispersion of the soil particles.

Keywords Exterior soil � Interior dust � Indoor

walking � Airborne soil � Particle transport

Introduction

Urban geochemical investigations have long focused

on citywide contamination of both soils and indoor

dust and the impact such contamination has on

human health. Increasingly, efforts have been made

to elucidate the nature and extent of exposure to

exterior contaminants in the indoor environment, as it

is recognized that people spend most of their time

indoors (Klepeis et al. 2001). The physical convey-

ance indoors of outdoor soil and dust (by, e.g., on

footwear or by pets) is an important mode of transfer

A. Hunt (&)

Department of Earth and Environmental Sciences,

University of Texas at Arlington, 500 Yates Street,

Box 19049, Arlington, TX 76019-0049, USA

e-mail: hunt@uta.edu

D. L. Johnson

Department of Chemistry, State University of New York

College of Environmental Science and Forestry, Syracuse,

NY, USA

123

Environ Geochem Health (2012) 34:355–363

DOI 10.1007/s10653-011-9400-8

of toxicants to the indoor environment (Lioy et al.

2002). The exposure risk posed by indoor dust

contaminated with outdoor pollutants is of particular

concern for sensitive pediatric populations; this is not

only from a developmental perspective, but also with

respect to age-specific preferential exposure (Roberts

and Dickey 1995). Crawling moves an infant to any

accessible floor spaces where inadvertent ingestion of

deposited dust can take place through hand-to-mouth

transfer (Hunt et al. 2008). Crawling also locates

the young child’s breathing zone proximal to a region

where mechanical resuspension poses a greater

exposure threat than for older children. Clearly, once

settled, indoor dust exists as a reservoir for particle-

bound hazardous substances, and typical indoor

activities, such as cleaning and walking, can result

in significant contaminant resuspension.

The resuspension of entrained dust deposited

indoors is an integral part of the exposure pathway

for outdoor pollutants. This mobilization of settled

dust has been the focus of many investigations (see,

e.g., Nicholson 1988 for a review of early studies).

The resuspension of floor dust as a result of

mechanical disturbance is known to increase the risk

for respirable particulate matter exposure (e.g., Ferro

et al. 2004a). The importance of the transfer flux

of dust indoors and its fate and transport within

the interior environment have been recognized in

efforts to develop models that assess transport into

and within homes. Abt et al. (2000) modeled source

emissions, infiltration rates, and associated decay

rates to define the subsequent distribution of particles

indoors. Layton and Beamer (2009) developed a

model that incorporates the ingress of soil and

airborne particles, the relocation of particulate matter

indoors by deposition and resuspension, and the

removal from the indoor environment by cleaning

and by egress of suspended particles to outdoor air.

Johnson (2008) developed a model that focused on

the distribution of track-in soil within the interior

environment by running simulations of foot tracking

along major traffic paths within a home.

In this study, we provide empirical data on the

small-scale dispersion of soil indoors at the point of

transition from the outdoor to the indoor environment.

The importance of outdoor soil as a contributor to

indoor dust has been summarized by Paustenbach

et al. (1997) who suggest that approximately 50% of

indoor dust is made up of soil. The amount of soil

introduced indoors is likely quite variable, but is

probably in the range of 50–300 mg/day (Johnson

2008). It is clear that the fate of outdoor soil following

entrance into a residential environment is likely to be

of some importance. We contend that the point of

ingress into the indoor environment is an important

one as it likely represents a site of maximum

deposition of external soil and dust transported by

track-in on footwear. Outdoor materials will tend,

upon initial track-in, to be deposited close to the site

of entrance (Cannell et al. 1987; Hunt et al. 2006).

Other evidence for the importance of this initial

introduction of outdoor soil comes from the work of

Roberts et al. (1990) who showed the amount of lead

(Pb), a frequent contaminant of outdoor soil, could be

reduced by 90% in carpets simply by removing shoes

before entering the home or by employing ‘‘walk-off’’

mats. Similarly, Farfel et al. (2001) demonstrated

the viability of entryway mats as major outdoor

dust collectors. In addition, it has been shown that

dust mass indoors declines as distance from the

entrance increases (Thatcher and Layton 1995),

although subsequent indoor distribution can be exten-

sive (Allott et al. 1992; Thatcher and Layton 1995).

Here, we assess the suspension and resuspension of

track-in dry soil that potentially occurs immediately

indoors following occupant entry. Using a set of laser

particle counters, we investigated the changes in

airborne dust levels following initial deposition and

aerosolization from contact between an adult-sized

shoe with adhering soil and a non-carpeted floor

surface. The effect of this transition from the outdoor to

the indoor environment was simulated using a human

tester tracking soil across vinyl flooring. This method is

not as reproducible as resuspension devices such as the

dropped spherical weight developed by Kildesø et al.

(1998, 1999), the weighted disk employed by Madler

and Koch (1997, 1999), or the artificial foot utilized by

Kivitso and Hakulinen (1981). However, the natural

actions of the tester probably more accurately reflect a

real-world track-in process.

Materials and methods

The dry soil aerosolization tests employed subsam-

ples of a composite surface soil from Syracuse, NY.

We have described this test soil previously, and of

most relevance to this study, we reported on the

356 Environ Geochem Health (2012) 34:355–363

123

varying element percentages in different size frac-

tions of the soil (Table 4 in Hunt et al. 2006).

For example, the lead (Pb) percentage was found to

increase with decreasing particle size, while the

silicon (Si) percentage decreased with decreasing

particle size. As a point of comparison for the

measurements taken in this study, in the 1- to 4-,

4- to 8-, and 8- to 16-lm size fractions in this soil,

the element percentages for Pb were, respectively,

0.13, 0.11, and 0.08 and for Si were, respectively,

23.00, 24.85, and 25.70.

At the beginning of the study, the test soil was

oven-dried, ground, and screened through an 85-lm

nylon mesh. All tests were conducted by an 86-kg

tester wearing size 11 (US) shoes with a fine ribbed

tread pattern on the sole. The tests were conducted

with a surface constructed of 1200 9 1200 vinyl floor

tiles set out in a 5 by 20 tile rectangle. The tester

walked along the two right-hand files/columns of

tiles while the sampling inlets for the particle

counters were set out on the first rank/row of tiles

of the three left-hand files. All tests used the same

tile set, and each tile was vigorously cleaned (by

wet wiping), rinsed with deionized water, and dried

between experiments. The track-in process involved

the tester pressing the left test shoe (under the

weight of the tester) into a soil reservoir (approx-

imately 10 g of soil evenly spread across a plastic

tray) to acquire a coating of dry soil on the sole and

heel of the shoe. Each test involved stepping on an

initial deposition tile (on the first step from the soil

reservoir) with the tester and then continuing

forward for 10 paces. The tester then remained

stationary for 2 min before retracing the step pattern

and stepping on the deposition tile (with the same

shoe) at the end of the pass. The soil reservoir was

not stepped in again during an individual test. After

another 2 min had elapsed, the tester repeated the

forward pass. This back-and-forth tracking process

was repeated until shoe-to-floor contact was made

with the initial deposition tile six times. Dry soil

was accumulated on the sole of the left shoe only at

the start of each test and was progressively lost from

the sole surface over the course of the back-and-

forth tracking.

The airborne levels of soil particles produced by

the step-on impact events were measured using four

Met One 237 A/B laser particle counters. These

counters use a laser diode light source and collection

optics for continuous particle detection. Sampled

particles scatter the light from the laser diode beam in

the direction of the collection optics. The collection

optics focuses the light onto a photodiode that

converts the bursts of light into electrical impulses.

The pulse height is proportional to the particle size.

Impulses are counted and their intensity is measured

for particle sizing. The instruments operated at a

sampling rate of 0.01 cfm (2.83 L/min) through

isokinetic sampling inlets. Six particle size channels

(0.3, 0.5, 0.7, 1.0, 2.0, and 5.0 lm) were used for

continuous data logging; this provided particle counts

in the size ranges: 0.3–0.5, 0.5–0.7, 0.7–1.0, 1.0–2.0,

2.0–5.0, and 5.0–&20 lm. The counters were started

simultaneously at the start of each test and were set to

count continuously with recording periods of 2 min

separated by a 2 s break. The tests were performed

within 4 months of a factory recalibration of the

counters (using NIST traceable monodispersed poly-

styrene spheres).

Measurements of the lateral changes in airborne

soil levels involved having the isokinetic sampling

inlets of the counters anchored to floor tiles lined up

next to the initial deposition tile. The results from

three test configurations are described here. In the

first configuration, the inlets probes were located at a

height of 10 cm and located at lateral distances of 20,

40, 60, and 80 cm from the initial step-on tile. The

inlets were staggered by a distance of 2 cm between

each inlet providing an unrestricted path to the

deposition tile. The test was subsequently duplicated

with the counters in the same order and then

replicated twice more with the counters switched

(replicated tests produced the same results). Vertical

variations in the airborne soil levels were assessed in

two sets of tests with the sampling inlets placed at

different heights. In the second test configuration, two

sampling inlets were located in parallel at heights of

10 and 25 cm at a distance of 30 cm from the initial

deposition tile, with the other two sampling inlets

lined up in parallel at heights of 10 and 25 cm at a

distance of 60 cm. The two sets of sampling inlets

were again staggered by 2 cm gap. In the third test

configuration (a second set of height comparisons),

the same lateral spread of sampling inlets was

employed; however, pairs of inlets were arranged at

heights of 10 and 35 cm.

Environ Geochem Health (2012) 34:355–363 357

123

Results

Numbers of particles counted in each channel by the

laser particle counters were assessed on a time-

resolved basis. A typical set of data describing the

spatial variation in the time evolution of the sus-

pended particles in the first test configuration is set

out in Fig. 1 (the timing of each footfall impact is

annotated on the graphs with a vertical white line

at the point where the particle count was summed

for the 2 min count period). Each graph plots the

size-resolved particle counts for one of the four

counters. Differences in particle numbers with

increasing lateral distance from the initial deposition

tile are here illustrated by the graph sequence a

through d. There are two components to the particle

count traces. Each trace, for some period, records a

general elevated background particle level (elevated

above the background level present at the start of the

test). Superimposed upon the elevated background

are transient peaks in numbers of particles that

correspond to incidences of ‘‘step-on’’ impact events

Par

ticl

e N

um

ber

s (t

wo

- m

inu

te p

erio

d)

Par

ticl

e N

um

ber

s (t

wo

- m

inu

te p

erio

d)

Par

ticl

e N

um

ber

s (t

wo

- m

inu

te p

erio

d)

Par

ticl

e N

um

ber

s (t

wo

- m

inu

te p

erio

d)

5.0- ≈20μm

≈20μμ

1.0-2.0

0.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

0.00 5.46 11.31 17.17 23.02

Elapsed Time (min)

Probe: 10cm high, 20cm lateral

5

50

500

5,000

22,000

26,00026,000

22,000

5,000

500

50

5

0.7-1.01.0-2.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

Elapsed Time (min)

0.00 5.46 11.31 17.17 23.02

Probe: 10cm high, 40cm lateral

1.0-2.0

0.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Pa rticle

Size (µm)

0.00 5.46 11.31 17.17 23.02

Elapsed Time (min)

5

50

500

5,000

24,000

28,000Probe: 10cm high, 60cm lateral

0.00 5.46 11.31 17.17 23.02

0.7-1.0

1.0-2.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

5

50

500

5,000

24,000

28,000

Elapsed Time (min)

Probe: 10cm high, 80cm lateral

a b

c d

5.0- ≈20μm

5.0- ≈20μm 5.0- ≈20μm

Fig. 1 Time- and size-resolved airborne particulate matter

levels at a height of 10 cm and at distances of 20, 40, 60, and

80 cm from a floor tile where repeated resuspension of soil

particles occurred following the impact of a contaminated shoe

sole and floor tile over a period in excess of 20 min

358 Environ Geochem Health (2012) 34:355–363

123

(on the deposition tile). The elevated background

count level decreased over the course of each test as

did the relative heights of the spikes in particle counts

associated with step-on events. In most of the tests, the

highest numbers of counts for a step-on event were

recorded after the first footfall of the test, unsurpris-

ingly when the sole of the shoe was first loaded with

picked-up dry soil. The general case that was observed

for these tests had subsequent step-on events produc-

ing successively smaller spikes in particle counts. We

propose that this is a response to an increasingly

reduced quantity of dry soil on the initial deposition

tile and on the sole of the shoe. In addition, in the

general case, the spikes that represent elevated levels

of resuspended particles are somewhat transient

phenomena. The numbers of airborne particles, gen-

erally, did not return to background levels between

step-on events, but for the most part, the airborne

levels did decline to a concentration lower than that

generated by a subsequent step-on event in the

sequence (three or more minutes later). We hypoth-

esize that either the airborne particles suspended by a

step-on event had been removed from the air by

various deposition processes or the suspended parti-

cles had been transported beyond the sampling inlets.

Interestingly, the data in Fig. 1 illustrate two excep-

tions to the general case. First, there is an instance of a

step-on impact later in the sequence of steps (the fifth)

producing a spike in the counts for all particles in the

size ranges [0.5 lm greater than all the preceding

step-on events except the first one in the sequence.

This variation likely arises as a result of unintentional

variability in the weight of the footfall of the tester.

Although this is indicative of inconsistent reproduc-

ibility on the part of the human tester, it also

demonstrates that under real-world conditions, signif-

icant resuspension events may occur even after

multiple tracking events. Second, this test is an

example of a situation where the maximum number

of particles recorded in the 0.3 to 0.5 lm size range

(the smallest measured) did not coincide with the

initial step-on event. Here, the highest particle levels

in this size range (for most of the sampling locations)

occurred during the latter half of the test. It is unclear

why this should be. Possibly, in this test, for some

reason, the coarser particles ([0.5 lm) were dispersed

early on in the sequence of steps and a greater

proportion of finer particles remained to be aerosol-

ized in the second half of the test.

In the second and third test configurations, sam-

pling inlets were fixed at two different heights.

Typical results from these tests (Figs. 2, 3) indicate

that not only do airborne particle levels produced

by step-on impacts on floor deposited soil decrease

with distance after repeated step-on events, but also

decrease with height. Even at modest heights (Fig. 2),

with sampling inlets co-located at the same distance

from the deposition tile, the magnitude of the spikes

in particle counts (after the initial step-on event)

produced by successive step-on impacts became

increasingly smaller (Fig. 2c). The same outcome

was reported by similar co-located sampling inlets at

a greater lateral distance from the deposition tile

(Fig. 2b, d). Unsurprisingly, the same pattern of

airborne particle levels was reported (in the third test

configuration) with co-located sampling inlets where

one sampling inlet was higher than in the second test

configuration. In this case, the data again provided an

example of unintentional variability in the weight of

the footfall applied by the tester. Here, the fourth

step-on impact in the tracking sequence produced

higher levels or airborne particles (in almost all size

ranges) than the preceding step-on impacts at a height

of 10 cm and lateral distances of both 30 and 60 cm

from the deposition tile (Fig. 2a, b). Nevertheless, at

slightly greater elevations (35 cm above ground) at

the 30 cm lateral location, the fourth step-on impact

becomes less significant in all the size ranges

[0.5 lm and is not an apparent contributor to the

particle levels in the 0.3 to 0.5 lm size range

(Fig. 3c). The outcome of most significance for this

sampling inlet configuration was that at the furthest

lateral distance (60 cm) from the deposition tile and

at a height of 35 cm, the spike in particle counts

produced by the fourth step-on event was no more

significant than the spikes in particle counts produced

by other step-on impacts in all size ranges. We

suggest that following this step-on impact event, the

coarser particles available for transport had insuffi-

cient momentum when mobilized to transport them

vertically to the elevated sampling inlet.

Discussion

Interest in the resuspension of settled indoor dust by

various physical activities has generally focused on

aerosolization into the breathing zone and modification

Environ Geochem Health (2012) 34:355–363 359

123

of an individual’s personal cloud. The activities that

tend to raise indoor particle levels most significantly

tend to be short-term activities such as vacuuming,

dusting, dancing, and folding linen (Ferro et al.

2004a). Studies have identified indoor particle

concentrations of 23 and 32 lg m-3 (Long et al.

2000; Ferro et al. 2004b) from dusting and 12 and

15 lg m-3 from walking. Vertical resuspension as a

result of walking is probably limited by the dynamics

of the treading action. Qian and Ferro (2008) found

that ‘‘heavy and fast walking’’ produced more floor

dust resuspension than less active walking and they

attributed this to ‘‘…a combination of increased pace,

increased air swirl velocity, and electrostatic field

effects established by the walking.’’ They also found

that the resuspension rates for particles seeded on a

new level-loop carpet were greater than for particles

on a vinyl floor.

The data presented here highlight the likely

variability associated with the ‘‘real-world’’ resus-

pension of outdoor soil that has been tracked into the

home. The tests revealed various consistent out-

comes; however, there were findings that deviated

from the general results. For example, the potential

for the highest airborne concentrations of the finest

sized particles (measured here as 0.3–0.5 lm) not to

1.0-2.00.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

0.0 5.5 11.3 17.2 23.014.28.42.5

Elapsed Time (min)

3,000

30

300

6,000

19,000

20,000 Probe: 10cm high, 60cm lateral

0.0 5.5 11.3 17.2 23.014.28.42.5

3,000

30

300

6,000

26,000

28,000

1.0-2.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

Elapsed Time (min)0.0 5.5 11.3 17.2 23.014.28.42.5

3,000

30

300

6,000

25,000

26,000

1.0-2.0

0.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

Par

ticl

e N

um

ber

s (t

wo

-min

ute

per

iod

)P

arti

cle

Nu

mb

ers

(tw

o-m

inu

te p

erio

d)

Par

ticl

e N

um

ber

s (t

wo

-min

ute

per

iod

)P

arti

cle

Nu

mb

ers

(tw

o-m

inu

te p

erio

d)

Elapsed Time (min)

Probe: 10cm high, 30cm lateral

Probe: 25cm high, 30cm lateralProbe: 25cm high, 60cm lateral

23,000

25,000

0.0 5.5 11.3 17.2 23.014.28.42.5

3,000

30

300

6,000

0.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

1.0-2.0

Elapsed Time (min)

a

c

d

b

5.0- ≈20μm 5.0- ≈20μm

5.0- ≈20μm5.0- ≈20μm

Fig. 2 Time- and size-resolved airborne particulate matter

levels at height of 25 cm and at distances of 30 and 60 cm from

a floor tile where repeated resuspension of soil particles

occurred following the impact of a contaminated shoe sole and

floor tile over a period in excess of 20 min

360 Environ Geochem Health (2012) 34:355–363

123

occur when the shoe sole loading was at a maximum

was unexpected. Also, for step-on impacts to produce

higher airborne concentrations of particles when the

amount present on the floor (and shoe sole) was less

than when the shoe sole loading was at a maximum at

the initial step-on impact suggest that even under

conditions with less dust on the floor, a heavy impact

footfall can produce high airborne particle levels.

The data also identified several consistent features

in the resuspension of soil under these experimental

conditions. First, spikes in airborne particle levels

were a relatively short-lived occurrence. We contend

that particles (certainly the coarser sized ones) that

were aerosolized by a step-on impact were, shortly

thereafter, removed either by deposition processes

or by the generated particle cloud moving laterally

beyond the sampling inlets. Second, in general, each

spike in particle counts corresponding to a step-on

impact on the deposition tile had lower counts than

spikes produced by preceding step-on events. We

account for this by a continual reduction in the

reservoir of particulate matter on the deposition tile

(and shoe sole) over the course of a single test. Third,

at greater sampling distances, the magnitude of the

Par

ticl

e N

um

ber

s (t

wo

-min

ute

per

iod

)P

arti

cle

Nu

mb

ers

(tw

o-m

inu

te p

erio

d)

Par

ticl

e N

um

ber

s (t

wo

-min

ute

per

iod

)P

arti

cle

Nu

mb

ers

(tw

o-m

inu

te p

erio

d)

Probe: 10cm high, 30cm lateral

18,000

20,000

900

9

90

9,000

0.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

1.0-2.0

Elapsed Time (min)

0.0 5.5 11.3 17.2 23.014.28.42.5 20.1

Elapsed Time (min)

0.0 5.5 11.3 17.2 23.014.28.42.5 20.1

1.0-2.00.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

3,000

30

300

6,000

18,000

20,000 Probe: 10cm high, 60cm lateral

500

5

50

5,000

25,000

27,000

1.0 -2.0

2.0 -5.0

0.5 -0.7

0.3 -0.5

Particle

Size (µm)

Probe: 35cm high, 30cm lateral

Elapsed Time (min)

0.0 5.5 11.3 17.2 23.014.28.42.5 20.1

Elapsed Time (min)

0.0 5.5 11.3 17.2 23.014.28.42.5 20.1

0.7-1.0

2.0-5.0

0.5-0.7

0.3-0.5

Particle

Size (µm)

Probe: 35cm high, 60cm lateral

500

5

50

5,000

24,000

25,000

a

dc

b

5.0- ≈20μm

5.0- ≈20μm5.0- ≈20μm

5.0- ≈20μm

Fig. 3 Time- and size-resolved airborne particulate matter

levels at height of 35 cm and at distances of 30 and 60 cm from

a floor tile where repeated resuspension of soil particles

occurred following the impact of a contaminated shoe sole and

floor tile over a period in excess of 20 min

Environ Geochem Health (2012) 34:355–363 361

123

spikes in particle counts (after the initial step-on

event) produced by successive step-on impacts

became increasingly smaller. We posit that under

these particle resuspension conditions at scales of

approximately [0.5 m, the lateral spread of resus-

pended dry soil after repeated step-on events becomes

less significant. The exception to this appears to be

aerosolized particles in the 0.3 to 0.5 lm size range

which seem to continue to be elevated over time.

Presumably, once aerosolized, these particles remain

suspended in a cloud that may, or may not, be

contributed to by successive step-on episodes.

The test results indicate that not only may the

nature of the foot to floor impact process limit the

vertical distribution of resuspended particles, but

also it likely determines the scope of the plume of

particles that extends laterally. The redistribution of

deposited particulate matter by walking is probably

controlled by several features of the step-on action. A

fraction of the dry soil adhering to the sole of the shoe

will detach under gravity both before and after a step-

on event. There also operates a ballistic mechanism

(transferring kinetic energy) from the impacting shoe

to the settled particulate matter by direct contact.

Then, there is a hydrodynamic component, resus-

pending particulate matter by impacting shoe creating

a flow disturbance (Eames and Dalziel 2000). In the

experimental procedure followed here, in the first

step-on impact of a test sequence, the ballistic

mechanism dominates the suspension of particulate

matter as the deposition tile was initially free of soil

(except for a small amount that may have detached

from the shoe sole by gravity prior to impact). In

subsequent step-on events, in addition to any direct

impact effects, resuspension would have taken place

if a sufficient amount of energy was transmitted by

air flow (produced by step-on impact) to the depo-

sition tile particles enabling the particles to overcome

adhesion with the surfaces or with a layer(s) of

intervening particles. Madler and Koch (1997), using

a dropped experimental weight, assumed that surface

particle resuspension is controlled by surface shear

stress and proposed a two-dimensional wall jet

developing beyond the point of contact of the weight

and the surface. A wall jet effect supports our results

that a potentially more significant effect of walking

indoors is the lateral spread of surface dust. We have

shown elsewhere (Hunt et al. 2006) that repeated

tracking of soil on hard surface flooring can lead to

significant lateral deposition of metals onto floor

areas not subject to direct foot traffic. The finding

here of a vertical drop-off in recorded particle levels

is important for infant exposure. For the crawling

child, the breathing zone will be close to ground

level. Under the exposure conditions simulated in this

study, it is apparent that the dust inhalation risk for

pre-walking infants is greater than that for older

walking children. Moreover, infant exposure is likely

to be exacerbated by preferential inhalation of

smaller-sized particles. This would be the case with

this test soil; as we noted earlier, the percentage of Pb

present in the soil increases with decreasing particle

size. From several lines of reasoning, Layton and

Beamer (2009) assert that two basic particle size

classes for indoor dust are important. These are the

B60 and [60–150 lm size fractions, with the finer

fraction being important in terms of adhering to the

hands of young children. From the results presented

here, we would conclude that an important size

fraction of settled dust is the\0.5 lm, which appears

to remain airborne at consistent levels following

walk-on resuspension actions.

References

Abt, E., Suh, H. H., Catalano, P., & Koutrakis, P. (2000).

Relative contribution of outdoor and indoor particle

sources to indoor concentrations. Environmental Scienceand Technology, 34, 3579–3587.

Allott, R. W., Kelly, M., & Hewitt, C. N. (1992). Behavior of

urban dust contaminated by Chernobyl fallout: Environ-

mental half-lives and transfer coefficients. EnvironmentalScience and Technology, 26(11), 2142–2147.

Cannell, R. J., Goddard, A. J. H., & ApSimon, H. M. (1987).

Contamination of dwellings by particulate matter: Ingress

and distribution within the dwelling. Radiation ProtectionDosimetry, 21, 111–116.

Eames, I., & Dalziel, S. B. (2000). Dust resuspension by the

flow around an impacting sphere. Journal of FluidMechanics, 403, 305–328.

Farfel, M. R., Orlova, A. O., Lees, P. S. J., Bowen, C., Elias, R.,

Ashley, P. J., et al. (2001). Comparison of two floor mat lead

dust collection methods and their application in pre-1950

and new urban houses. Environmental Science and Tech-nology, 35(10), 2078–2083.

Ferro, A. R., Kopperrud, R. J., & Hilderman, L. M. (2004a).

Source strengths for indoor human activities that resus-

pend particulate matter. Environmental Science andTechnology, 38(6), 1759–1764.

Ferro, A. R., Kopperrud, R. J., & Hilderman, L. M. (2004b).

Elevated personal exposure to particulate matter from

362 Environ Geochem Health (2012) 34:355–363

123

human activities in a residence. Journal of ExposureAnalysis and Environmental Epidemiology, 14, S34–S40.

Hunt, A., Johnson, D. L., Brooks, J., & Griffith, D. A. (2008).

Risk remaining from fine particle contaminants after

vacuum cleaning of hard floor surfaces. EnvironmentalGeochemistry and Health, 30, 97–611.

Hunt, A., Johnson, D. L., & Griffith, D. A. (2006). Mass

transfer of soil indoors by track-in on footwear. Science ofthe Total Environment, 370, 360–371.

Johnson, D. L. (2008). A first generation dynamic ingress,

redistribution and transport model of soil track in: DIRT.

Environmental Geochemistry and Health, 30, 589–596.

Kildesø, J., Vinzents, P., & Schneider, T. (1998). Measuring

the potential resuspension of dust from carpets. AerosolScience, 29(Suppl. 1), S287–S288.

Kildesø, J., Vinzents, P., & Schneider, T. (1999). A simple

method for measuring the potential resuspension of dust

from carpets in the indoor environment. Textile ResearchJournal, 69(3), 169–175.

Kivitso, T., & Hakulinen, J. (1981). Der staubgehalt der luft

in raumen mit textilen fussboden-belagen. Staub-Rein-halt.Luft, 41, 357–358.

Klepeis, N. E., Nelson, W. C., Ott, W. R., Robinson, J. P.,

Tsang, A. M., Switzer, P., et al. (2001). The national

human activity pattern survey (NHAPS): A resource for

assessing exposure to environmental pollutants. Journal ofExposure Analysis and Environmental Epidemiology,11(3), 231–252.

Layton, D. W., & Beamer, P. I. (2009). Migration of con-

taminated soil and airborne particulates to indoor dust.

Environmental Science and Technology, 43, 8199–8205.

Lioy, P. J., Freeman, N. C. G., & Millette, J. R. (2002). Dust: A

metric for use in residential and building exposure

assessment and source characterization. EnvironmentalHealth Perspectives, 110(10), 969–983.

Long, C. M., Suh, H. H., & Koutrakis, P. (2000). Character-

ization of indoor particle sources using continuous mass

and size monitors. JAWMA, 50, 1236–1250.

Madler, L., & Koch, W. (1997). Particle resuspension from

surfaces by impacting objects. Journal of Aerosol Science,28(Suppl. 1), S85–S86.

Madler, L., & Koch, W. (1999). Particle resuspension burst

from dust layers induced by impacting objects. Journal ofAerosol Science, 30(Suppl. 1), S731–S732.

Nicholson, K. W. (1988). A review or particle resuspension.

Atmospheric Environment, 22, 2639–2651.

Paustenbach, D. J., Finley, B. L., & Long, T. F. (1997). The

critical role of house dust in understanding the hazards

posed by contaminated soils. International Journal ofToxicology, 16, 339–362.

Qian, J., & Ferro, A. R. (2008). Resuspension of dust particles

in a chamber and associated environmental factors.

Aerosol Science Technology, 42, 566–578.

Roberts, J. W., Camann, D. E., & Spittler, T. M. (1990).

Monitoring and controlling lead in house dust in older

homes. In D. S. Walkinshaw (Ed.) Indoor air ‘90: Pro-

ceedings of the 5th International conference on indoor air

quality and climate, Vol. 2. Toronto, Canada, 29 July–3

Aug 1990. Canada Mortgage and Housing Corp., Ottawa.

Roberts, J. W., & Dickey, P. (1995). Exposure of children to

pollutants in house dust and indoor air. Reviews of Envi-ronmental Contamination and Toxicology, 143, 59–78.

Thatcher, T. L., & Layton, D. W. (1995). Deposition, resus-

pension, and penetration of particles within a residence.

Atmospheric Environment, 13, 1487–1497.

Environ Geochem Health (2012) 34:355–363 363

123