See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257388962

Photocatalytic performances in a city tunnel in Rome: NOx monitoring results

Article  in  Construction and Building Materials · February 2012

DOI: 10.1016/j.conbuildmat.2011.07.065

CITATIONS READS

66 652

1 author:

Gian Luca Guerrini

Consultant
58 PUBLICATIONS   514 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Photopaq View project

Fotocatalisi e sanificazione - Photocatalysis and Sanitation View project

All content following this page was uploaded by Gian Luca Guerrini on 14 January 2020.

The user has requested enhancement of the downloaded file.

 Author's personal copy

Construction and Building Materials 27 (2012) 165–175

Contents lists available at SciVerse ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Photocatalytic performances in a city tunnel in Rome: NOx monitoring results

Gian Luca Guerrini ⇑
Italcementi - Italcementi Group, Bergamo, Italy

a r t i c l e i n f o a b s t r a c t

Article history: This paper deals with the results of two environmental monitoring campaigns entailing the analysis of
Received 19 October 2009 NOx levels registered before and after the renovation work of the ‘‘Umberto I’’ tunnel in Rome, Italy.
Received in revised form 26 July 2011 The results of the data collection highlighted an efficient abatement of pollutants by the photocatalytic
Accepted 28 July 2011
treatment of the tunnel vault with a cement-based paint, as evidenced through the lower concentrations
Available online 30 August 2011
found after the renovation. A NOx reduction of about 20% was calculated in the center of the tunnel (as an
absolute comparison between the two periods), whereas a NOx reduction of over 50% was calculated in
Keywords:
the same position, considering the reference data registered in the official stations of the city of Rome
Depollution
Lighting system
(Official Environmental Agency – ARPA).
Monitoring The duration of the monitoring campaigns performed seems to be considered significant from the sta-
NOx tistical point of view, unlike all previous experiments which were carried out on the photocatalytic mate-
Photocatalysis rials.
Tunnel This renovation project can prove out to be a very useful reference for the wide program for a sustain-
able upgrading of tunnels in Europe, according to the Directive 2004/54/EC on road tunnel safety.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction
Photocatalysis has been applied for over a decade to various
materials – among which cementitious binders – to obtain a ‘‘self
cleaning’’ and, more recently, ‘‘depolluting’’ effects. In particular,
on-going research shows that cement-based materials containing
TiO2 have a good potential in urban pollution control [1–3]. Exam-
ples of pollutants which can be eliminated by the photocatalytic
cementitious products are NOx, SOx, NH3, CO, volatile organic car-
bons (VOCs) such as benzene and toluene, organic chlorides, alde-
hydes, and polycondensated aromatics.
This technique is particularly effective when wide surfaces
(horizontal or vertical) are coated. In this sense, the construction
of pavements for busy streets, roads, parking lots, intersections,
gas stations and toll gates seems to be the best solution in order
to optimize the photocatalytic depollution.
Significant experiences have recently been carried out in Japan,
Italy, France, Belgium and United States (mainly consisting of con-
crete roads and pavements built by means of paving blocks).
Therefore, several solutions can be adopted, for a wider use of
this environmentally friendly technology which can give a signifi-
cant contribution to a healthier life.
Indeed, a number of cementitious materials and related prod-
ucts are by now widely validated in laboratory. However, the
crucial point still seems to be the demonstration of their photocat-
⇑ Tel.: +39 035579325; fax: +39 035570251.
E-mail address: g.guerrini@itcgr.net
alytic properties in real conditions. For this reason, some pilot pro-
jects, including a monitoring programme for demonstrating the
degree of depolluting action (referred to NOx) by means of photo-
catalytic cement-based materials, have already been carried out.
Due to the high investments on equipment or the costs of the mon-
itoring program management, this activity is still limited to rele-
vant projects or to short duration of measurement, which
sometimes means limits the range of data for an acceptable statis-
tical evaluation.
Furthermore, the variability of local weather conditions must
also be considered, so that the complexity of the approach results
to be higher.
Two ‘‘pioneering’’ experiences were carried out in Italy in
2002–2003, when relevant NOx abatement performances were reg-
istered in Segrate, near Milan (a cement-base coating on a bitumi-
nous road) and in Calusco, near Bergamo (industrial site with
paving blocks) [4]. Another relevant experience has recently been
performed in Bergamo, in 2006 (renovation of a main road using
paving blocks) [5]. In this case, a broader monitoring approach
was adopted in order to collect high quantities of data, both before
and after the renovation work: results demonstrated a difference
between photocatalytic pavement and asphalt zone (reference) of
about 20–50%, as a function of the weather conditions. Finally, a
recent project has been completed in Paris – Porte de Vanves,
through the construction of a concrete road (2007).
This report describes the first example of a monitoring approach
adopted for an indoor application of photocatalytic products in a
tunnel.

0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2011.07.065
 Author's personal copy

166 G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175

This renovation project could act as a very useful reference for

the wide program for a sustainable upgrading of tunnels in Europe,
according to the Directive 2004/54/EC on road tunnel safety.
It is widely known that in recent years, the authorities respon-
sible for tunnel safety have passed a series of particularly restric-
tive regulations concerning tunnel construction and maintenance,
as well as controls and signals both at the tunnel entrances and in-
side the tunnels. In Italy, one of the procedures to be followed by
tunnel operators, for example with reference to motorway tunnels,
is that the interior walls must be painted white up to a height of 4
meters and kept perfectly clean (Italian regulations of 31/12/2000).
In addition, due to an increase in the number and efficiency of
the lighting systems, both visibility and safety have significantly
improved, thanks to the benefits of reflecting capacity of the large
surfaces along the sides of the roadway. As a consequence, the vis-
ibility conditions inside the tunnels are subject to a quick drop,
thus affecting the reflecting capacity of the walls which is caused
Fig. 1. Painting of the tunnel.
by the dirt and smog generated by vehicles. In order to keep the
walls clean, it should be advisable to wash the interior walls of
the tunnels through high-pressure jets of water, whose employ-
ment however creates significant problems in terms of safety.
The carriageway must be narrowed for these cleaning operations,
therefore the risks for motorists as well as maintenance workers
are extremely high.
The use of the photocatalytic system can improve this situation,
thus reducing the number of maintenance works and, conse-
quently, the corresponding costs.
In the case of city tunnels, a strong improvement in air quality
conditions is expected especially for health pedestrians: the possi-
bility to reduce their exposition to unhealthy level of polluting
agents (especially NO2, CO, polycyclic aromatic hydrocarbons and
particulate matters) is a strong and current concern.
In this particular situation, the exposure time is anyway limited
to about 5 mins, however very high pollution levels could give sig-
nificant airway inflammatory and blood coagulation consequences.
Pollution and weather conditions are less variable than in out-
door environment, thus allowing to evaluate the photocatalytic Fig. 2. Painting of the tunnel.
depolluting action with a reduced number of parameters and mak-
ing the interpretation of collected data results less complex. Fur-
thermore, not only the tunnel geometry, but also the level of
lighting can be considered constant in this project, due to the com-
plete substitution of the lighting system by means of dedicated
lamps, with a high amount of UV-A light.
For the above mentioned reasons, the most relevant parameters
considered in our evaluation will be:

– Level of pollution (strictly connected to the traffic pollution).

– Wind speed.

Before presenting the monitoring results, a description of the

tunnel and the renovation works will be given.

2. Tunnel description
This structure is located in the center of Rome – under the
Quirinale Hill, near the Italian Republic Presidential Building
(Quirinale) – and is by far the most famous tunnel in the Capital,
very relevant for the road system of the whole city center. Its con-
struction started in June 1900 (Fig. 4) and the official opening was
on October 20, 1902. Its dimensions are: length 347.70 m, width:
17 m, height: 8.5 m.
This renovation was absolutely necessary, due to the need of
substituting the lighting system (lamps) and the electrical lines
(in compliance with to the current regulations and safety rules),
and to the dirtiness of the vaults, which were coated by an oily
Fig. 3. The tunnel before re-opening.
thickness of smog. An official decision concerning the tunnel reno-
vation was taken in May 2007 and a monitoring program to eval-
uate the current pollution condition was immediately defined
before the starting of the works (programmed in the summer per-
iod, in order to minimize the traffic problem in that very congested
zone of Rome). According to the plan, renovation works covered
four weeks (August 2007).
The first step of the work included the removal of the existing
lighting system, then an accurate washing/cleaning of the vault
 Author's personal copy

G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175 167

depollution degree which is to be obtained and to the management

costs in terms of power consumption. The lighting system was de-
signed by Disano Group and required the use of experimental data
provided by CTG Laboratories, Fig. 4, in order to determine the ade-
quate level of UV lighting required for the activation of the photo-
catalytic reactions along the tunnel. For this purpose, two types of
lamp were chosen and installed in the tunnel and their position
was defined in accordance with the safety of the pedestrians walk-
ing through the tunnel, Figs. 5 and 6.

3. The cement-based paint, used for the coating

Fig. 4. Reference curve for NOx abatement (CTG data).
was carried out before installing the electrical line and the new
lighting system.
Afterwards, a photocatalytic paint was applied by airless spray-
ing technique in two layers on the vault (a gray paint up to 1.8 m of
height and a white paint for the remaining surface), for a total of
9000 m2, Figs. 1 and 2. All these steps were completed in four
weeks.
It is important to point out that the choice of the lighting system
and the equipment to be used for the activation of the photocata-
lytic effect is strictly connected to the tunnel geometry, to the
In order to renovate the tunnel, a photocatalytic cement-based
paint was employed, applying a gray paint on both sides of the tun-
nel (until 1.80 m of height) and a white paint on the remaining sur-
face, Fig. 3.
The photocatalytic property of both products was assessed
according to the NOx gas recirculation method, adopted by CTG
for the quality evaluation of TX Active products, Figs. 7 and 8.
According to the adopted methodology, the best formulations cho-
sen for the tunnel renovation showed a NOx abatement capacity of
88–90% after 60 min, under UV light irradiation (20 W/m2).
Some qualitative tests carried out according to the UNI standard
11259 (colorimetric–Rhodamine B test) showed the maintenance
capacity of the white paints: the a reduction after 26 h was over

Fig. 5. Simulation design of lighting system: visible light irradiation (left) and UV-light irradiation (right).

Fig. 6. Calculation of light irradiation, along the tunnel.

 Author's personal copy

168 G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175

Fig. 7. NOx gas recirculation test.

Fig. 8. Position of analyzers in the tunnel (longitudinal section).

70%, with respect to the initial values of redness (minimum level of Measurements of NOx values, traffic volumes, light conditions
acceptance for standard mortars prepared with photocatalytic and wind speed were carried out during the daytime (period
hydraulic binder: 50%). 8.00–18.00), from Monday to Friday.
Main equipment used for the monitoring campaigns, as follows:

4. The monitoring program – 4 Chemiluminescence NOx analyzers (by ENVIRONNEMENT).

– 1 Calibrating system for NOx analyzers (by ENVIRONNEMENT).
The monitoring approach was defined by CTG in agreement – 1 Thermo-hygrometer (by TESTO).
with some managers of the Municipality of Rome, from the Depart- – 1 Luxmeter (by TESTO).
ment for the Environmental Activities. – 1 UVA/UVB/RAD radiometer (by Delta OHM).
Environmental monitoring was accomplished through two suc- – 1 Anemometer (by Delta OHM).
cessive campaigns, each of them lasting three weeks (15 working
days). The four NOx analyzers were positioned along the tunnel, as de-
The first campaign was carried out in the period: July 3, 2007 – scribed in Fig. 9.
July 20, 2007. The height of gas sampling was approximately 1 m, except for
The second campaign was carried out in the period: September the analyzer referred to the 6 m of height, in the center of the tun-
25, 2007 – October 12, 2007. nel, Fig. 8.
 Author's personal copy
G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175
The analyzers were set to measure air concentration values, in
ppbVol, of NOx, NO and NO2 (measuring frequency: one measure-
ment/minute).
In addition to NOx measurements, the most significant data
which were recorded were humidity, temperature, wind parame-
ters, cloudiness, pressure and atmospheric precipitations, as well
as the number of vehicles travelling at daytime, with an hourly
frequency.
During the monitoring campaign carried out after the renova-
tion works, some measurements of lighting parameters were also
carried out in different points of the tunnel, using a portable
photo-radiometer. These measurements confirmed the mean light
values, previewed in the design phase.
5. Results and discussion
The following data were collected:
– NOx values (NO, NO2 and NOx).
– Weather conditions (T, RH, P, wind speed) and sometimes, light
conditions inside and outside the tunnel (UVA, UVB, RAD, Lux).
– Traffic situation, vehicles/h.
In order to compare the mean values of NOx in different posi-
tions of the tunnel, the official data from ARPA Lazio (Official Envi-
ronmental Protection Agency in Rome) were processed in advance.
5.1. Official pollution data, from ARPA Lazio
Some data which will be included in this report were collected by
some official weather stations (ARPA Lazio) [6]. ARPA Lazio operates
several air quality monitoring sites inside the Region including
Rome. Rome’s monitoring network consists of 13 monitoring sta-
tions classified into four different types: A, B, C and D. Type A stations
are usually located in areas not directly affected by traffic sources
such as parks or green areas: they monitor pollutants such as CO,
SO2, NOx, NO, BTX, PM10 and O3. Types B are located in areas with
heavy traffic conditions and monitor CO, NOx, BTX, PM10 and O3.
Types C are located in residential areas. They monitor CO, NOx and
BTX. Types D are located outside the urban area, not far from the
countryside. They monitor O3 and NOx, and are devoted to the con-
trol of photochemical pollution. The monitoring network receives
concentration data every hour. These data are sent to the Regional
Environmental Protection Agency (ARPA) for validation and deliv-
ered to the Environmental Department of the Municipality of Rome
which is responsible for data collecting, storing and delivering.
In the full list of monitoring stations, three of them were chosen
for reference, with the agreement of the Technical Managers of the
Environmental Department of the Municipality of Rome:
– Villa Ada (‘‘background’’ station) – A Type (gardens).
– Largo Arenula (‘‘traffic’’ station) – B Type (avenue) – nearest
station to the tunnel.
– Fermi (‘‘traffic’’ station) – B Type (avenue).
Daily pollution values in Rome registered for the two reference
periods by means of official (fixed) stations were higher in the per-
Table 1
ARPA official stations - comparison of mean daily values – NOx and NO2.
Values in (lg/m3) NOx
July 2007 September–October 2007
Largo Arenula 114 252
Via Fermi 161 185
Villa Ada 57 92
169
iod of three-week monitoring campaign (September–October
2007), in comparison with the three-week period of July 2007. This
is clearly confirmed in Table 1, referred to Largo Arenula, via Fermi
and Villa Ada stations.
Data are referred to NOx and NO2 as mean daily values, between
8.00 a.m. and 6.00 p.m. h and are expressed as lg/m3.
Crossing the tunnel from via Nazionale to via del Tritone, the
mean daily NOx values increase, probably due to the already men-
tioned windy conditions and to the pollution effects derived from
the other entrance (‘‘via del Tritone’’).
The NOx mean values are very high, but lower than the values
registered in a previous, similar experience in Milan (via Porpora,
2004): daily values of 400–800 ppb (Rome) against 1200–
1400 ppb (Milan).
An increase of NOx values during the afternoon was almost reg-
ularly observed, due to the stationing of tourist buses (especially
for the positions next to the entrance ‘‘via del Tritone’’) and very
high pollution peaks were sometimes registered.
5.2. Traffic
The number of passing vehicles per hour (mean hourly value)
which was registered during the two monitoring campaigns was
similar for both periods (July 2007 and September–October 2007)
and amounted to approximately 1100 vehicle/hr.
The location of the tunnel is in the historical center of Rome, in
the Restricted Access Zone (LTZ, Limited Traffic Zone, ‘‘Zona a Traff-
ico Limitato’’), where daily public traffic is interdicted from
7.00 a.m. to 6.00 p.m. (from Monday to Friday). During evening
and night hours and during week-ends, no restrictions are applied
on public circulation (except Saturday afternoons).
During the LTZ hours, the traffic is mainly consisting of public
transport, tourist buses, taxis (more or less, 50%) and private trans-
port or ‘‘blue cars’’.
Indeed, a large number of tourist buses were observed along the
day, mainly in the range: 10.00 a.m.–4.00 p.m. (both in July and in
September–October period).
5.3. Environmental conditions inside the tunnel
In relation to the semi-confined conditions, the values of tem-
perature, relative humidity and pressure in the tunnel were almost
constant, along the day and in the monitoring weeks.
During the first period (July), the average temperature mea-
sured in daylight period was in the range 28–40 °C (entrance ‘‘via
del Tritone’’), around 30–32 °C (in the tunnel).
In the second period, the average temperature was 20–28 °C
(outside) and 23–25 °C (tunnel).
The average relative humidity was in the range 15–40% during
the first campaign and in the range 40–70% during the second cam-
paign (a few rainy days did not however affect the conditions in
the tunnel).
As to the wind speed, it was influenced by the traffic variations
along the two periods, because of the high speed of vehicles pass-
ing through the tunnel.
However, typical values registered in the entrance ‘‘via del Tri-
tone’’ ranged between 0.10 and 1.70 m/s, corresponding to an aver-
NO2
July 2007 September–October 2007
65 73
95 102
38 49
 Author's personal copy

170 G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175

2000
FIRST CAMPAIGN SECOND CAMPAIGN
1st and 2nd WEEK 1st, 2nd and 3rd WEEK
1800

1600

1400
NOx, microg/m3

1200

1000

800

600

400

200

0
10/7

11/7

12/7

13/7

25/9

26/9

27/9

28/9

1/10

2/10

3/10

4/10

5/10

8/10

9/10

10/10

11/10

12/10
3/7

4/7

5/7

6/7

9/7

NAZIONALE CENTRE TRITONE

Fig. 9. Daily trends of NOx values in the tunnel (from ‘‘Via Nazionale’’ to ‘‘Via del Tritone’’) Values in lg/m3.

1400

NOx - mean daily values - 8.00-18.00

1200 July 2007 Sept-Oct 2007

1000

1m 6m
NOx, microg/m3

800

600

400

200

0
7

7
/7

/7

/7

/7

/7

/7

/7

/7

/7

/9

/9

/9

/9
10

10

10

10

10

10

10

0
3/

4/

5/

6/

9/

/1

/1

/1
10

11

12

13

16

17

18

19

20

25

26

27

28

1/

2/

3/

4/

5/

8/

9/
10

11

12

Fig. 10. Comparison of NOx values, in the center of the tunnel (1 m and 6 m).

age value of 0.38 m/s. the highest values along the tunnel were with an open space, we can consider lighting conditions to
always detected in the center. be steady. However, due to the large size of the tunnel, the
As to the direction of the wind, it was always the same: from presence of sunlight radiation in daytime combined with
the entrance ‘‘Via Nazionale’’ towards ‘‘Via del Tritone’’. permanent artificial light, could reinforce the photocatalytic
action.
(b) On this basis, the most relevant parameters which could
5.4. NOx monitoring results: preliminary remarks influence the monitoring results are traffic (number of vehi-
cles per hour) and wind speed.
Before considering and discussing the numerical results, some (c) The data collection of analyzers positioned just outside the
preliminary remarks can be summed up, as follows: entrance ‘‘via del Tritone’’ was strongly influenced by local
traffic, particularly by the temporary parking of tourist buses
(a) In relation to the specific application which deals with a tun- (the tunnel is a few hundred meters far from the famous
nel, that is a semi-confined place, and not with a road, nor ‘‘Fontana di Trevi’’).
 Author's personal copy

G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175 171

Fig. 11. NOx values in the center of the tunnel, 1 m.

2000

1800 NOX - mean daily values - 8.00-18.00

1600

1400

1200
microg/m3

1000

800

600

400

200

0
ly

ly

11 ly

12 ly

16 ly

19 ly

ly

28 pte er

O er

ct r

ct r

ct r

ct r

ct r
y
ly

10 ly

13 ly

17 ly

18 ly

20 ly

27 pte er

pt er

ct r

O er

O er

er
2 obe

3 obe

4 obe

5 obe

9 obe

e
8 obe
Ju

Ju

Ju

Ju

Ju

Ju

Ju
l

Se mb

1 mb

12 ctob
Ju

Ju

Ju

Ju

Ju

Ju

Ju

Se mb

Se mb

10 tob

11 ctob

ob
ct
3

e
4

26 pte

c
O

O
O

O
Se
25

villa ada fermi arenula C1m NAZ TRIT

Fig. 12. A comparison of NOx mean daily values (8.00 a.m. - 6.00 p.m.).

This phenomenon is seasonal, occurring from spring period to (e) A large number of tourist buses was recorded along the day,
the end of summer period and is particularly evident in the central mainly in the time range from 10.00 a.m. to 4.00 p.m. (both
part of the day (from 10.00–11.00 a.m. to 4.00–5.00 p.m.). in July and in September–October period).

(d) A number of factors make the tunnel quite windy, the most
significant of them being its large size, the volume of traffic
which is predominant in one direction (from via Nazionale
to via del Tritone), as well as its two lanes in the opposite
direction the circulation is allowed only to public vehicles
(buses, taxi) service vehicles and ‘‘official cars’’ (no private
vehicles). In addition, the tunnel grade is slightly downhill,
from via Nazionale to via del Tritone.
Monitoring data were collected as .txt files from the NOx -me-
ters and then evaluated by means of simple statistical methods.
5.5. Collected NOx data, in the tunnel
With reference to preliminary considerations concerning:
– Daily trends of NO2 values (see previous graph).
 Author's personal copy

172 G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175

500

450 NO2 - mean daily values - 8.00-18.00

400

350
microg/m3

300

250

200

150

100

50

0
ly

ly

ly

ly

10 ly

11 ly

12 ly

13 ly

16 ly

17 ly

18 ly

19 ly

20 ly
ly

27 pte er

28 pte er
pt er

O er

ct r

ct r

ct r

ct r

ct r

c r
O er

O er

O er
er
2 obe

3 obe

4 obe

5 obe

8 obe

9 obe
Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju

Se mb

Se mb

Se mb

1 mb

10 tob

11 ctob

12 ctob

ob
ct

ct
3

O
pt
Se
25

26

villa ada arenula fermi CENTRE 1m NAZIONALE TRITONE

Fig. 13. A comparison of NO2 mean daily values (8.00 a.m.–6.00 p.m.).

Table 2
Mean NOx values in reference zones and in the tunnel, for both periods (lg/m3).

Ada July Ada September–October 2007 Arenula Arenula September–October 2007 Tunnel July 2007 Tunnel September–October 2007
2007 July 2007
Mean values 108 175 216 479 667 530
Difference +61% +121% 21%

– Geometry and local conditions of the tunnel.
– Possible origin of pollution in the tunnel (traffic + private
buses).
The data referred to the center of the tunnel (Figs. 9 and 10) will
be considered as the most representative for this evaluation in the
report, in which all registered data were however elaborated and
included.
As expected, the values of polluting gases (NOx) increase, from
one side to the other side of the tunnel: this is because the typical
distribution of pollutants in a tunnel with a natural ventilation,
produces a ‘‘plug effect’’. Gas concentration results to be homoge-
neous in the cross-section of the tunnel, which was confirmed by
the direct comparison of NOx concentrations at the center of the
tunnel, at two different heights (1 m and 6 m). A slight difference
for these values was registered only in occasion of a very windy
day (July 20, 2007), Fig. 10.
As to the values registered in the center of the tunnel, the reduc-
tion of pollution (in absolute value) is evident, Fig. 11.
The reduction of average daily values after the tunnel renova-
tion is also evident, if compared with the corresponding values reg-
istered in the official stations, Figs. 12 and 13.
In particular, values referred to the center of the tunnel and to
the entrance ‘‘via del Tritone’’ clearly decrease, but this reduction
is even more significant for the entrance ‘‘via Nazionale’’ as it has
the same order of magnitude referred to the ‘‘open space’’ official
stations, both for NOx and NO2.
Besides, the values of NO2 in the center of the tunnel are within
the reference limit value of 200 lg/m3, defined for outdoor urban
situations (and these values are certainly lower if referred to the
24 hr period, in relation to the lower volume of traffic during night
hours).
Although the tunnel situation is basically critical, because of the
strong concentration of pollutants which can be registered in this
semi-confined area, after the renovation work better environmen-
tal conditions have definitively been achieved.
Such an achievement has proved to be crucially important for
the pedestrians and bikers crossing the tunnel, as their exposure
level is absolutely more comfortable than the previous situation,
with a reduction of health risks against traffic emissions [7,8].
Table 2 summarizes the mean values in reference zones (official
stations) and in the center of the tunnel, before and after the ren-
ovation phase.
By comparing these values, after the renovation there was a NOx
reduction of 21%: [(667–530)  100]/667.
Indeed, if we consider that there is an increase of pollution lev-
els in Rome, passing from the first to the second period, corre-
sponding to +61% for ‘‘Villa Ada’’ and +121% in ‘‘Largo Arenula’’,
it is reasonable to consider an increase of pollutants in the tunnel,
without any renovation work.
For example, considering the pollution increase in Villa Ada,
‘‘background’’ station (+61%), the real value in the tunnel should
be: 1.61  667 = 1074 g/m3, with a theoretical abatement of
51%: [(1074–530)  100]/(1.61  667).
However, the pollution level in the tunnel should be closer to
the values registered in ‘‘Largo Arenula’’ station, which is not far
from the tunnel, with a similar traffic volume. Assuming the same
pollution increase of ‘‘Largo Arenula’’, we should obtain a pollution
level of 1474 g/m3 (maximum calculated NOx concentration), with
a theoretical abatement of 64%: [(1474–530)  100]/(2.21  667).
This calculation can also be justified if we consider the correla-
tion coefficient which is a dimensionless value determined as a ra-
tio between the covariance and the mean quadratic squares of two
variables x and y:
 Author's personal copy

G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175 173

Table 3 Table 5
Correlation coefficient – July 2007. Variance of average values, NOx.

Villa Ada Largo Arenula Tunnel Villa Largo Tunnel Via Via
Ada Arenula center Nazionale Tritone
Villa Ada 1 0.8342 0.0443
Largo Arenula 0.8342 1 0.2693 July 631 577 51,915 25,897 148,796
Tunnel 0.0443 0.2693 1 September– 1808 3236 7560 3675 40,602
October
Difference (%) +187 +421 85 86 73

P
n
ðxi  xÞðyi  y Þ
i¼1
Table 4 rxy ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð1Þ
Correlation coefficient – September/October 2007. P n P n
ðxi  xÞ2 ðyi  y  Þ2
Villa Ada Largo Arenula Tunnel i¼1 i¼1

Villa Ada 1 0.7215 0.4765

As to the two reference stations (‘‘Largo Arenula’’ and ‘‘Villa
Largo Arenula 0.7215 1 0.5681
Tunnel 0.4765 0.5681 1 Ada’’) and the tunnel, the pollution trend could be evaluated in
relation to the two periods: (July and September–October).

4000
NOX - 1st MONITORING CAMPAIGN - July 2007
3500
3/7 4/7 5/7 6/7 9/7 10/7 11/7 12/7 13/7 16/7 17/7 18/7 19/7 20/7

3000
NOX, microg/m3

2500

2000

1500

1000

500

0
13.22
16.45
9.53
13.16
16.39
10.24
13.47
17.10
10.39
14.02
17.25
10.44
14.07
17.30
11.14
14.37
18.00
11.30
14.53
8.14
11.37
15.00
18.23
11.04
14.27
9.13
12.36
15.59
9.38
13.01
16.24
10.11
13.34
16.57
10.42
14.05
17.28
11.10
14.33
17.56

4000
NOX - 2nd MONITORING CAMPAIGN - Sept-Oct 2007
3500
25/9 26/9 27/9 28/9 1/10 2/10 3/10 4/10 5/10 8/10 9/10 10/10 11/10 12/10

3000
NOX, microg/m3

2500

2000

1500

1000

500

0
11.02
14.20
17.38
11.16
14.34
17.52
11.32
14.50
18.08
11.34
14.52
8.41
11.59
15.17
8.53
12.11
15.29
9.11
12.29
15.47
9.30
12.48
16.06
9.40
12.58
10.37
13.55
17.13
10.56
14.14
17.32
11.14
14.32
17.50
11.08
14.26
17.44
11.31
14.49

Fig. 14. A comparison of the number of NOx peaks, before/after the renovation.
 Author's personal copy

174 G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175

2000
NO2 - 1st MONITORING CAMPAIGN - July 2007
1800

1600
3/7 4/7 5/7 6/7 9/7 10/7 11/7 12/7 13/7 16/7 17/7 18/7 19/7 20/7
1400
NO2, microg/m3

1200

1000

800

600

400

200

0
13.22
16.45
9.53
13.16
16.39
10.24
13.47
17.10
10.39
14.02
17.25
10.44
14.07
17.30
11.14
14.37
18.00
11.30
14.53
8.14
11.37
15.00
18.23
11.04
14.27
9.13
12.36
15.59
9.38
13.01
16.24
10.11
13.34
16.57
10.42
14.05
17.28
11.10
14.33
17.56
2000

NO2 - 2nd MONITORING CAMPAIGN - Sept-Oct 2007

1800

1600

25/9 26/9 27/9 28/9 1/10 2/10 3/10 4/10 5/10 8/10 9/10 10/10 11/10 12/10
1400
NO2, microg/m3

1200

1000

800

600

400

200

0
11.02
14.20
17.38
11.16
14.34
17.52
11.32
14.50
18.08
11.34
14.52
8.41
11.59
15.17
8.53
12.11
15.29
9.11
12.29
15.47
9.30
12.48
16.06
9.40
12.58
10.37
13.55
17.13
10.56
14.14
17.32
11.14
14.32
17.50
11.08
14.26
17.44
11.31
14.49

Fig. 15. A comparison of the number of NO2 peaks, before/after the renovation.

Tables 3 and 4 show the correlation coefficients for both

Table 6
periods. Variance of NOx instantaneous values (tunnel).
Indeed, in July a close correlation value between ‘‘Largo Arenula’’
NOx, Mean value (g/m3) Variance
and ‘‘Villa Ada’’ stations is observed (0.8342), whereas the correla-
tion between the tunnel (semi-confined area)and the official sta- July 681 327,028
September–October 529 82,138
tions (open spaces) is not acceptable (coefficients: 0.0443 and
0.2693). In September–October period the correlation between
the official stations shows the same trend (0.7215), whereas the sit-
uation in the tunnel is more similar to the surrounding open spaces The variance (Table 5) has increased for the two open stations,
thus obtaining higher correlation coefficients (0.4765 and 0.5681). due to the increase of pollution and of the number of peaks in
Other statistical considerations can be made in terms of vari- September–October.
ance and peak reduction. On the contrary, the variance for the tunnel is high in July
The variance is a measure of the average distance between each (because of the influence of semi-confined environment on the
of a set of data points and their mean value, in other words an eval- number of pollution peaks) but decreases dramatically due to the
uation of the ‘‘data similarity’’: presence of photocatalytic paint, Table 5.
Pn The same considerations can be made for instantaneous NOx
i¼1 ðxi  xÞ2 and NO2 values, registered in the tunnel, whose decrease in num-
r2 ¼ ð2Þ
n ber is clearly illustrated in Figs. 14 and 15.
 Author's personal copy
G.L. Guerrini / Construction and Building Materials 27 (2012) 165–175
For example, the variance determined on instantaneous values
of NOx decreases of 75% (passing from 327,028 to 82,138), Table 6.
The same results were obtained for NO and NO2 instantaneous
values.
6. Conclusions
The results obtained from the monitoring program have con-
firmed the high efficiency of photocatalytic paint used for the
‘‘Umberto I’’ tunnel renovation in Rome, the first significant exam-
ple of an indoor application for depollution abatement. The dura-
tion of the data collection can be considered significant from the
statistical point of view.
The high performances of photocatalytic cement-based prod-
ucts have already been determined in other pilot projects, through
outdoor applications (e.g. Via Borgo Palazzo, in Bergamo, Italy).
Indeed, the complexity of statistical analysis is lower, in com-
parison with outdoor conditions: in the tunnel, the main parame-
ters to be considered for the variation of polluting level are traffic
volumes and wind speed. Besides, light irradiation can be consid-
ered constant along the two monitoring periods, thus simplifying
the evaluation of results.
Another interesting result was obtained comparing tunnel pol-
lution levels with official stations set up near the tunnel: this infor-
mation have been very useful for an adequate evaluation of the
obtained monitoring results.
Considering the absolute values numerically calculated, a
reduction of NOx values over 20% was achieved, particularly, in
the center of the tunnel:
– A 25% reduction of NO values.
– A 23% reduction of NOx values.
– A 19% reduction of NO2 values.
However, the outdoor environmental pollution values regis-
tered during the second campaign (September–October 2007) in
Rome were higher than the corresponding values in July 2007,
we can conclude that the real depolluting effect induced by the
system (paint + lighting system) is relatively higher than the above
mentioned values (20–25%). Accordingly, the reduction of pollu-
tion level in the center of the tunnel results to be higher than
50% (calculated abatements range from 51% until 64%) bon ac-
count of the statistical approach applied to the data evaluation.
View publication stats
175
Another relevant effect derived form the photocatalytic degra-
dation of polluting gases is the reduction of pollution peaks (indi-
vidual values) observed for all NOx gases (NO and NO2), also
confirmed by the statistical calculation of the variance for data
population.
The photocatalytic treatment of the vault is really effective and
enables reduce the pollution level up to nearly outdoor conditions.
On account of this treatment, the tunnel has now been totally
renovated in terms of pollution conditions, lighting system and
visibility.
Acknowledgements
This renovation project was carried out with the financial con-
tribution of Italcementi S.p.A. and C.I.M. (Calci Idrate Marcellina),
and after the approval of the Municipality of Rome.
Lighting system design was carried out by Disano Group.
The author acknowledges dr. E. Borgarello, R&D Manager of CTG
Italcementi Group for his encouraging support, as well as E. Peccati,
M. Plebani, G. Cividini, M. Lamera, V. Spreafico (CTG) and M. Ber-
nardoni, G. Faro, D. Gerani, and A. Broggi (C.I.M.) for their assis-
tance in the monitoring campaigns.
References
[1] Cassar L. Cementitious materials and photocatalysis. Betonwerk + Tech
2005;1:10–6.
[2] Cassar L, Beeldens A, Pimpinelli N, Guerrini GL. Photocatalysis of cementitious
materials. In: Baglioni P, Cassar L, (editors). RILEM International symposium on
photocatalysis ‘environment and construction materials’, (Florence/Italy, 8–9
October 2007), TDP 2007, (Florence/Italy, 8–9 October 2007) Rilem Proc. PRO55.
p. 131–146.
[3] <www.picada-project.com>.
[4] Plassais A, Guillot L. De-polluting activity assessment of photocatalytic cement-
based materials: from laboratory to real scale testing’, In: Tenth international
symposium on concrete roads (Brussels/Belgium, 18–22 September 2006). p.
13.
[5] Guerrini GL, Peccati E. Photocatalytic cementitious roads for de-pollution. In:
Baglioni P, Cassar L, editors. RILEM international symposium on photocatalysis
‘environment and construction materials’, (Florence/Italy, 8–9 October 2007),
TDP 2007, (Florence/ltaly, 8–9 October 2007) Rilem Proc. PRO55. p. 179–86.
[6] <http://www.arpalazio.net>.
[7] Svartengren M, Strand V, Bylin G, Jarup L, Pershagen G. Short-term exposure to
air pollution in a road tunnel enhances the asthmatic response to allergen. Eur
Respir J 2000;15:716–24.
[8] Larsson B-M, Sehlstedt M, Grunewald J, Sköld CM, Lundin A, Blomberg A, et al.
Road tunnel air pollution induces bronchoalveolar inflammation in healthy
subjects. Eur Respir J 2007;29:699–705.