Effectiveness of Rinse Water during In-Place

Cleaning of Stainless Steel Pipe Lines

Mengyuan Fan, David M. Phinney, and Dennis R. Heldman

Abstract: The 1st step of any Clean-In-Place (CIP) operation is a prerinse with water. The purpose of this step is to
remove the bulk of food material remaining in the processing lines after production period has ended. It is known that
this prerinse step can be a very water intensive process. The objective of this investigation was to measure the influence
of CIP parameters (flow characteristics, water temperature, and contact time) on the effectiveness of prerinse water
in removing dairy-based deposits from stainless steel pipe surfaces and to compare the rinse effectiveness of unused to
reused rinse water. A pilot-scale CIP system was operated to rinse 304 stainless steel pipe sections of 3 different pipe
diameters. The velocity of the rinse water was varied from 0.72 to 2.26 m/s. The rinse water temperatures were 22 °C,
45 °C, and 67 °C. The contact times between rinse water and deposited film were 20 and 60 s. Rinse effectiveness was
E: Food Engineering &

expressed as the ratio of the amount of protein residue removed from the pipe surface during rinsing, as compared to
Materials Science

the magnitude of the initial protein deposit. The rinse effectiveness varied from 73.1% to 94.9% for the range of the
CIP parameters investigated. High velocities of rinse water provided a higher level of rinse effectiveness. Increasing the
rinse water temperature from 23 °C to 45 °C improved rinse effectiveness significantly (P < 0.05). This impact was not
significant when the water temperature was increased from 45 °C to 67 °C and at higher rinse water velocities. Similarly,
longer contact time provided less improvement in rinse effectiveness at higher temperatures and velocities as compared
to lower temperatures and velocities. There were no significant differences in rinse effectiveness when comparing reused
and unused water (normal tap water) within the range of velocities evaluated.

Keywords: Clean-In-Place, cleaning effectiveness, food processing, prerinse, protein deposition

Practical Application: The rinse steps are important components of the CIP operation and have direct impact on the
amounts of water and energy used for the entire processing operation. The efficiency of rinse water can be improved
significantly by the selection of appropriate combinations of operating parameters. For example, higher velocities of
rinse water (2.26 m/s) provide significant improvements on rinse effectiveness when compared to current commercial
practice (1.52 m/s). The careful selection of rinse water temperature and velocity can result in overall reductions in water
and energy used for cleaning operations. The reuse of water for a 2nd or 3rd pass provides additional opportunities for
reducing water requirements without influencing effectiveness.

Introduction periods. The success of applying CIP does depend on the degree
Fouling is the accumulation of undesired food materials on the to which it has been integrated into a given processing system,
surfaces of processing equipment and pipelines. The build-up of including the specific food deposit characteristics (Seiberling and
dairy-based deposits during processing is a significant challenge for Hyde 1997). Most CIP systems share 5 common steps: prerinse,
dairy processing operations due to the influence on heat exchange alkaline cleaning, rinse, acid cleaning, and final rinse (Seiberling
efficiency and on product quality. Cleaning is pivotal to any food 1955; Seiberling 1968). In other words, the cleaning process is
processing system since equipment fouling cannot be prevented based on 3 types of operations; rinsing, cleaning, and sanitizing.
(Dunsmore 1983; Bird and Bartlett 1995). Failure to properly The main purpose of prerinse is to remove the loose soil from the
clean processing equipment effectively and efficiently can increase hard surface (Seiberling and Hyde 1997). The remaining rinsing
overall costs of operations, and can lead to postprocessing product steps are used to remove the chemical residues remaining from
contamination (Fryer and others 2013). the alkali and acid steps.
Clean-In-Place (CIP) has been widely used in the dairy industry The high cost of detergent, water, and energy use during CIP
since the 1960s. The key improvements achieved by CIP systems has put escalading economic and environmental stress on industry
are automation and reduced instances of equipment disassembly, (Dif and others 2013). The optimization of CIP processes is
resulting in less down time and fewer unpredictable factors while required both in terms of minimizing the cost of water and
cleaning (Stewart and others 1996). With CIP, liquids are pumped energy and reducing the detergent usage due to its health or
through the processing lines and circulated for various time environmental risk (Wilson 2005).
Most studies related to CIP systems have been focused on the
alkaline cleaning step. The driving force of alkaline cleaning and
MS 20141681 Submitted 10/10/2014, Accepted 4/22/2015. Authors are with
the related kinetic models have been studied on various types of
Dale A. Seiberling Food Engineering Laboratory, The Ohio State Univ., 2015 Fyffe
Road, Columbus, OH 43210 U.S.A. Direct inquiries to author Heldman (E-mail: deposits. According to Schöler and others (2012), mass transfer
[email protected]). of detached foulant was the controlling factor of the cleaning
of a starch-based foulant from stainless steel surface. Le Gentil

C 2015 Institute of Food Technologists

 R

E1490 Journal of Food Science r Vol. 80, Nr. 7, 2015 doi: 10.1111/1750-3841.12914
Further reproduction without permission is prohibited
Effectiveness of rinse water during in-place . . .

and others (2010) investigated the effect of alkaline cleaning of water, temperature of water, and contact time between water
on removing spores from stainless steel pipeline, during which and the surface to be rinsed will impact the effectiveness of the
velocity was the driving force at the beginning of the cleaning prerinse step (Dunsmore 1983; Stewart and others 1996).
cycle. After 5 min, contact time was more important in removing Compared to the alkaline cleaning step, the prerinse step has
spores. Gillham and others (1999) suggested temperature was been less studied. Models which describe the impact of control
more important than flow rate during the removal of whey parameters on effectiveness of alkaline cleaning are not applicable
protein from stainless steel surface in the alkaline cleaning to the prerinse step. The reactions associated with the alkaline are
step. absent during the prerinse, making the effective distribution be-
It has been widely agreed that both physical and chemical forces tween physical and chemical forces much different. Development
are involved in controlling the rate and the extent of cleaning, thus of appropriate models between the prerinse control parameters
any parameters impacting the flow characteristics or dissolution can be used to ensure prerinse effectiveness while reducing wa-
of deposit will impact cleaning (Grant and others 1997; Fryer ter/energy requirements, leading to establishing optimum cleaning
and others 2006; Weidemann and others 2013). Increasing parameters. The overall objective of this investigation was to quan-
temperature can improve cleaning to a certain extent, however, tify the effectiveness of the prerinse step in the removal of a deposit
the relationship between temperature and rinse effectiveness is not film in a pilot scale CIP system by quantifying the influence of

E: Food Engineering &

linear. Furthermore, it has been shown that raising temperatures various parameters (rinse water temperature, velocity, and contact

Materials Science
may not increase cleaning effectiveness when the temperature is time) and the effect of reuse of rinse water.
above 50 °C (Gillham and others 1999; Goode 2012). Higher flow
rates affect cleaning by increasing shear force that is required to
mechanically remove the food deposit (Gillham and others 2000). Materials and Methods
However, flow rates and temperatures are not the only factors
influencing flow characteristics, the size and shape of the surface CIP system description
as well as fluid properties (that is, viscosity or accumulation of A pilot plant scale CIP system with an experimental test pipeline
soluble solids) are also factors to be considered (Jullien and others manifold was developed for this research. The pilot CIP system
2003; Boxler and others 2013). Finally, the contact time between (Figure 1) includes a water tank, a solution tank, acid and alkaline
the cleaning solution and the surfaces to be cleaned impacts the tanks, test pipeline manifold, connecting pipeline, heater, pump,
cleaning effectiveness as well as water and energy consumption. control panel (not shown in this figure), thermocouples, flow
Longer contact time requires a higher energy input to the meters, and a conductivity sensor. Type 304 stainless steel was used
pump and higher water consumption to the system in single-use to fabricate the test pipeline manifold and connecting pipeline.
CIP systems where unused water is continuously introduced The test pipeline manifold (Figure 2) contains 0.0254 m (1 inch)
while used water is drained without recirculation in the system and 0.0381 m (1.5 inch) outside diameter test pipelines connected
(Cords 2001). to flow concurrently with a 0.0254 m (1 inch) by pass line before
The 1st step in the CIP process—and the focus of this flow recombines through the 0.0508 m (2 inch) test pipeline.
investigationis a prerinse using water to clear the bulk of unbound The inside diameter for 0.0254, 0.0381, and 0.0508 m testing
deposit as well as wetting the deposit surface in preparation for pipes were 0.0222, 0.0349 and 0.0476 m, respectively. Each test
the following detergent step (Guzel-Seydim and others 2000). pipeline contains 7 0.1524 m (6 inch) individual pipe sections. The
The contribution of the prerinse has a significant impact on the pass line was designed to balance flow once the water was diverted
entire cleaning process. Although prerinse may not solely remove to the test pipeline manifold. In this study only the middle and 2
the entire deposit, it does reduce the deposit load carried into end sections of each line were soiled while the remaining sections
the alkaline cleaning step thereby making the following CIP served only as connections. Tap water was used throughout this
steps more efficient (Goode 2012). Parameters such as velocity study. The pH of the water was 7.21 to 7.69 (Model HI 2020,

Figure 1–Illustration of pilot-scale CIP

system with flow meters at locations 1 and
Cold Hot 3, thermocouples sensors at 2 and 4, and a
Testing pipe water water conductivity sensor at location 5.
manifold

Water
tank

Solution Heater
tank
Alkali Acid

Valve

Drain Pump

Vol. 80, Nr. 7, 2015 r Journal of Food Science E1491

 Effectiveness of rinse water during in-place . . .

Hanna instruments, Inc., Woonsocket, R.I., U.S.A.) and tap water Table 1–Protein content of the deposit film in pipe sections lo-
total hardness as CaCO3 ranged from 54 to 82 ppm (Total hardness cated in various positions of test manifold after 20 s contact time
with rinse water at 45 °C and flow rate of 0.00284 m3 /s.
test kit 4482, LaMotte Co., Chestertown, Md., U.S.A).
The range of Reynolds Numbers investigated in this study was Protein content of the deposit film (µg/cm2 )
from 1.6 × 104 to 2.6 × 105 , which suggested the flow was tur- Pipe diameter 1 2 3
bulent. The lowest magnitude occurred at a temperature of 22 °C
at a mean velocity of 0.72 m/s in a 0.0254 m diameter pipe, while 0.0254 m (1 inch) 16.99 ± 2.09a 14.72 ± 1.40a 11.75 ± 1.42a
0.0351m (1.5 inch) 11.39 ± 3.54b 9.42 ± 1.51b 9.58 ± 2.01b
the highest Reynolds Number was achieved at 67 °C, 2.26 m/s in 0.0508 m (2 inch) 11.64 ± 4.47c 11.75 ± 1.42c 13.43 ± 2.62c
a 0.0508 m diameter pipe. The entrance effect on flow character-
istics was not evident, based on the lack of significant differences The reported values correspond to the means ± standard deviation. Statistical analysis
was only compared across pipe sections (rows) and not pipe diameters (columns). Values
(P > 0.05) among the protein content of the residual film (Table 1) followed by different letters (a, b, c) in the same row are significantly different (p < 0.05).
for different pipe sections along a test line of a given diameter
(Figure 2). Due to gravity and other fixed factors of the system Table 2–Rinse water flow rates and velocities in 3 pipe section
set-up, the flow separation was not equalized in the manifold mak- diameters.
ing the volumetric flow rates and velocities in 0.0254 m test line, Outside diameter
E: Food Engineering &

0.0381 m test line, and 0.0254 m pass line differ from each other. System set flow Flow rate
Materials Science

In order to establish the actual flow velocities of each pipe line, an rate ∗ 10-3 (m3 /s) (m) (inch) Vave (m/s) ∗ 10-4(m3 /s)
ultrasonic velocity meter (Dynasonics R
Portable Ultrasonic Flow 1.58 0.0254 1 0.72 2.82
Meter, Racine Federated, Inc., Racine, Wis., U.S.A.) was used 1.58 0.0351 1.5 1.09 10.42
to obtain the maximum velocities in each pipe line under each 1.58 0.0508 2 0.99 17.67
flow rate (0.00158, 0.00284, and 0.00385 m3 /s). Validation of the 1.58 0.0254 1∗ 0.87 3.39
2.84 0.0254 1 1.25 4.84
tested flow velocities was achieved by the following calculations. 2.84 0.0351 1.5 1.87 17.86
As the flow was turbulent (Re > 4500), the equation Vave = 2.84 0.0508 2 1.70 30.29
Vmax × 0.82 was used to calculate the average velocity of each 2.84 0.0254 1∗ 1.72 6.68
pipe line. Nine velocities were achieved in the CIP system using 3.85 0.0254 1 1.67 6.46
3.85 0.0351 1.5 2.49 23.81
3 different flow rates through 3 different pipe diameters (Table 2).
3.85 0.0508 2 2.26 40.39
The volumetric flow rate of each testing pipe line was calculated 3.85 0.0254 1∗ 2.26 8.81
based on the tested velocities by the equation:
1∗ Denotes the 1 inch pass line.

Q = Vave ×A (1)
(3) prerinse procedure, (4) determining the quantity of foulant
where Q is the volumetric flow rate, Vave is the average velocity, remaining after cleaning.
and A is the cross-sectional area of the pipe.
Table 1 shows the flow separation result under variously set flow Soiling of pipe sections
rates. The velocity test results were validated by the equation: The pipeline sections were soiled before installation into the CIP
test pipeline manifold. Nonfat-dry-milk (NFDM) (U.S. Foods,
Q2 inch =Q1 inch +Q1.5 inch +Qpass line (2) Inc., Rosemont, Ill., U.S.A.) was reconstituted to 20% w/w
(72 mg/mL protein by Bradford Assay, Sigma-Aldrich Co., St
The fundamental processes for this study were: (1) soiling of Louis, Mo., U.S.A.) in distilled water at ambient temperature.
pipe sections, (2) determining the quantity of fouling achieved, Seven pipe sections of a given size were connected together and

Figure 2–Detailed illustration of test manifold,

with arrows indicating the flow direction. Numbers
1, 2, and 3 designate test pipe section locations
within each test line, sections numbered 1 are
located at the entrance of flow into each test line.

E1492 Journal of Food Science r Vol. 80, Nr. 7, 2015

Effectiveness of rinse water during in-place . . .

Table 3–Initial protein content of film deposit for each pipe di- the mean protein concentrations for pipe sections with 3 different
ameter. pipe diameters, as corrected for pipe diameter. The differences
Pipe diameter (m) Average (µg/cm2 ) in protein concentration (on a µg/cm2 basis) among the various
pipe sections were not significant (P > 0.05). These results
0.0254 268.68a ± 34.64
0.0381 247.41a ± 38.11
confirm that the method used to create the deposit films on the
0.0508 241.87a ± 33.86 interior surfaces of the pipe sections could be reproduced.
Ninety-nine data points were used to calculate the average protein content for each
diameter. The reported values correspond to the means ± standard deviation. Values CIP procedure
followed by the same letter (a) are not significantly different (P > 0.05). Temperature, contact time, and rinse water velocity were the 3
CIP parameters evaluated. The velocities ranged from 0.72 to 2.48
one end of the assembly was capped, and they were filled with m/s in this study, including the industry recommended velocity
the reconstituted NFDM and the opposite end was also capped. of 1.52 m/s (Seiberling and Hyde 1997). Three temperatures
After 5 min one of the caps was removed and the milk was de- (22 °C, 45 °C, and 67 °C) were selected to provide a range
canted. The same milk was then used to rinse the interior surfaces of typical temperatures used during CIP operations. This range
of the pipe section to eliminate air bubbles. The pipe assembly included ambient water temperature, a high temperature as
was then disconnected and each pipe section gasket wheel and

E: Food Engineering &

recommended by Xin and others (2002) and an intermediate

Materials Science
connecting face was wiped clean. Pipe sections were stood on end temperature. Contact time was determined by the length of time
and allowed to drain at room temperature for 30 min then each the rinse water was allowed to pass through the test pipeline
section was placed into a 75 °C oven for 30 min to further dry manifold; 2 levels of contact time were applied in this study:
and increase adhesion to the stainless steel inner pipe surface. In 20 and 60 s. At the beginning of each CIP run the water tank
order to validate the uniformity of the deposit, during the prelim- was rinsed with tap water for 5 s after which the pump was
inary research, fluorescent agent was added to the reconstituted shut down and the tank water drained. After draining, the water
NFDM, and then the formed deposit was checked by naked eye tank was filled with tap water. The system was then flushed to
under UV light, the yellow light spread on the interior uniformly remove air pockets and confirm an unobstructed flow. This was
which confirmed the quality of the deposit formed. accomplished by pumping the tank water through the CIP system
Throughout the study 1 of 7 soiled pipe sections was usually as it was diverted away from the test pipe sections, at the desired
removed for nonuniform film distribution due to bubble forma- flow rate (0.00158, 0.00284, or 0.00385 m3 /s) until the flow
tion. Three of the remaining sections were randomly selected to and temperature was constant. The valves to the test pipeline
determine the initial soiling level while the remaining 3 sections manifold were then opened for the specified contact time. After
were subjected to a CIP prerinse experiment. the prerinse time (20 or 60 s in this study), the pump was stopped
and the entire system was allowed to drain for 5 min.
Determining the quantity of deposit within pipe sections
Manual cleaning of each pipe section, those subjected to CIP Unused and reused rinse water
and those retained for control analysis, was performed using a To compare the efficiency of reused water to unused water
tube brush while rinsing with a 2% (wt/wt) NaOH solution, that during the prerinse step, a study was performed with 45 °C water
was captured in a 50 mL volumetric flask via a glass funnel. The at a flow rate of 0.00284 m3 /s with a 60 s contact time. For the
pipe was cleaned then inverted and cleaned again to ensure all of unused water portion of this study, the valve control remained
the soil was removed. All surfaces, pipe, brush, and funnel, were the same as above. For the reused water portion of this study the
thoroughly rinsed with 2% NaOH into the flask. The volume valve between the solution tank and drain remained closed from
was then brought up to 50 mL with 2% NaOH. The amount of the outset allowing the water to recycle through the test pipe
NaOH used for rinsing the protein away from the pipe section manifold over the duration of contact time.
interior was validated by recleaning the pipe section with fresh
NaOH solution and confirming that the protein concentration of Data treatment
the wash solution was negligible. After each CIP rinse step, the pipe test sections were disassem-
A QuantiProTM Bicinchoninic Acid (BCA) Assay Kit (Sigma- bled and the remaining deposit film was removed from each pipe
Aldrich Co.) was used to quantify the protein concentration section and quantified. The effectiveness of the CIP rinse step was
within the NaOH wash from each pipe section. Assay results determined by comparison to the initial film deposit.
were observed spectrophotometrically as absorbance at 562 nm  
and converted to concentration (µg/mL) using a standard curve C
Percent residual film (RF%) = ×100 (3)
established with bovine serum albumin. This concentration, along C0
with the pipe section dimensions was used to calculate the amount
of protein per cm2 of surface area. The measurement was based Rinse effectiveness = 1 − Percent residual film (4)
on the formation of the purple-blue copper–protein complex.
where C (µg/mL) is the protein concentration after prerinse and
Concentration of the initial protein deposit C0 (µg/mL) is the initial protein concentration. Both C and C0
Given the temperature and the primary component of the were quantified by the BCA Assay mentioned in the previous
reconstituted NFDM, the deposit in this study was a protein-based section.
cohesive solid foulant, which is the type 3 deposit as described
by Goode and others (2013), but not a calcium-based milk stone. Statistical analysis
The amount of protein per unit of pipe section surface area SPSS.21 (IBM, Inc., Chicago, Ill U.S.A.) statistics software was
was determined for 99 different pipe sections using the method used to analyze all data. The reported values correspond to the
described in the previous section. The results in Table 3 present means ± standard deviation. A one-way ANOVA with 95%

Vol. 80, Nr. 7, 2015 r Journal of Food Science E1493

 Effectiveness of rinse water during in-place . . .

confidence intervals followed by least significant difference post hoc shear force is the key factor in breaking down and removing the
test were conducted to evaluate the statistical difference between deposit film when the flow of water is introduced into the pipe
the means of the residual film percentage on the data collected section (Timperley and Smeulders 1987). When other factors,
for the interactive impact of velocity and temperature on prerinse pipe diameter, temperature, and viscosity of the fluid, are fixed,
effectiveness. An independent sample T- test, with a 95% confi- higher velocity delivers a higher shear force. Thus increasing
dence interval, was used to analyze the difference in the means the velocity could provide a better cleaning result. The primary
of residual film percentage on the data collected for contact times mechanism of viscous deposit removal during water rinse is the
and reused rinse water. shear force of the water flow (Yeckel and Middleman 1987).
Gillham and others (2000) also claimed that flow shear force plays
Results and Discussion a key role during the initial seconds of contact, and during the
entire process of removal of whey protein deposits from stainless
Effect of water velocity on rinse effectiveness steel. Besides a higher shear force, high velocity can also provide
The influence of the velocity of 22 °C rinse water on the a higher mass transfer rate. According to Plett (1985), the mass
percent residual film is illustrated in Figure 3. A log-linear transfer rate of the reaction products within the layer between the
relationship, RF% = 41.68 × exp(−0.68 × Vave ), was used to fluid and the deposit increased with greater fluid velocity, which
describe the relationship between the residual film percentage
E: Food Engineering &

also explained a better rinse effectiveness at higher fluid velocity.

Materials Science

(RF%) and the rinse water velocity (Vave ), with a standard error
of 13.4% (P = 0.95). The RF% decreased from 26.9 ± 2.54% to
9.3 ± 1.43% as the velocities of rinse water increased from 0.72 Effect of temperature on rinse effectiveness
to 2.26 m/s. The results indicate that higher velocities provided The influence of rinse water temperature on residual film is
more significant removal of the deposits, which is in agreement shown in Figure 4. As the water temperature was increased, the
with previous research (Timperley and Smeulders 1987). Higher magnitude of the residual film decreased from 26.9 ± 2.54% to

Figure 3–Effect of rinse water velocity (Vave ) on

residual film percentage (RF%) with 22 °C rinse
water, and 20 s contact time; error bars represent the
standard deviation of replicates.

Figure 4–Effect of rinse water temperature and

velocity on the residual film percentage after a
20 s contact time, error bars represent the
standard deviation of replicates.

E1494 Journal of Food Science r Vol. 80, Nr. 7, 2015

Effectiveness of rinse water during in-place . . .

15.6 ± 4.52% when the velocity of the flow is 0.72m/s. This ob-order to investigate the interactions of temperature and velocity
on rinse effectiveness, the water velocities at various temperatures
servation indicates that an increase in the rinse water temperature
from 22 °C to 67 °C improved the rinse effectiveness. This trend
and the water temperatures at various velocities were analyzed.
is in agreement with previous research by other study groups onThe residual film remaining after rinsing at 22 °C, 45 °C, and
whey protein deposits (Xin and others 2002; Mercadé-Prieto and67 °C, velocities of 0.72, 1.25, and 2.26 m/s are presented in
Chen 2006). Water temperatures affected diffusion coefficients,Figure 4. The results show the range of RF% is from 5.1% to 26.9%
and enhance the removal of deposit when all other parameters at the tested combinations of velocities and temperatures, indi-
are fixed (Fryer and Robbins 2005). However, some researchers cating that between 73.1% and 94.9% deposit film was removed
suggested higher temperature did not lead to a higher removal during the rinse step. According to Christian (2004), whey protein
of deposit. Kulkarni and others (1975) suggested that removal deposit formed at 100 °C cannot be removed by rinse water. The
of a protein film by cold (10 °C to 15 °C) water was similar todeposit in this study was formed at a lower temperature (75 °C)
removal of the same film by hot (75 °C to 80 °C) water. Limitedcompared to the temperature used in Christian’s study and the
impact of higher temperature on removal of deposit could be material used in this study is reconstitute NFDM instead of whey
observed if the temperature of water was above 50 °C (Liu and protein, thus a less severe deposit was formed, which might be the
others 2006; Goode and others 2010). This lack of agreement is reason why the rinse effectiveness could be as high as 94.9% in this

E: Food Engineering &

study.
most likely because the magnitudes of fluid velocities and contact

Materials Science
time of deposit to the fluid are different when compared to the Figure 4 also illustrates that the percent residual film decreased
current investigation. as the water velocity increased. Velocity had a significant effect
on the effectiveness of deposit film removal at all 3 temperatures
Interaction of water temperature and velocity on rinse (P = 0.0001, P = 0.0001, P = 0.001 for 22 °C, 45 °C and
effectiveness 67 °C, respectively). At 22 °C, a water velocity of 1.25 m/s reduced
Based on previous results, both high water temperature and percent residual film significantly (P = 0.001) when compared to
high velocity seem to improve the effectiveness of rinsing. In 0.72 m/s. In a similar manner, a water velocity of 2.26 m/s reduced

Figure 5–Effect of rinse water contact time and

velocity on the residual film percentage at a rinse
water temperature of 45 °C, error bars represent
the standard deviation of replicates.

Figure 6–Effect of unused and reused rinse water

on residual film percentage at a rinse water
temperature of 45 °C after a 60 s contact time at
3 rinse water velocities; error bars represent the
standard deviation of replicates.

Vol. 80, Nr. 7, 2015 r Journal of Food Science E1495

 Effectiveness of rinse water during in-place . . .

the percent residual film significantly compared to 1.25 m/s. The (Jurado-Alameda and others 2011), and strong mechanical force
same patterns were observed for data collected at 45 °C and 67 °C. may play a key role during these initial seconds of contact
The results in Figure 4 also indicate that higher water tem- (Weidemann and others 2013). It is important to note that there
peratures provide a significantly lower residue film percentage, is a point of diminishing return where no further deposit can be
or better rinse effectiveness, at low and medium water velocities removed by water alone and the remaining deposit can only be
(P = 0.001, and P = 0.003 for 0.72 and 1.25 m/s, respectively). At removed after chemical agents are introduced (Goode 2012). At
velocities of 0.72 m/s, a significantly higher (P = 0.005) amount higher velocities, the increased mechanical force can remove all
of deposit film was removed by rinsing at 45 °C as compared to the facile deposit film within a short period of time. Extending
22 °C. Similarly, a significantly higher (P = 0.003) amount of de- contact time in this situation may not affect the rinse effectiveness
posit film was removed by rinsing at 67 °C as compared to 45 °C. since the remaining deposit is strongly attached to the pipe interior
However, the same patterns were not observed for data collected in a manner which requires chemical action to remove. At lower
at higher velocities of rinse water. When the water velocity was velocities, the initial removal caused by flow mechanical force was
2.26 m/s, the percent residual film was significantly lower at 45 °C less significant compared to higher velocities, thus longer contact
and 67 °C compared to 22 °C. This demonstrates that increasing time could affect the rinse effectiveness by providing more time for
the temperature from 45 °C to 67 °C at this higher velocity did the water to remove loose deposit. It could be expected that the
E: Food Engineering &

not significantly increase deposit removal. impact of contact time in rinsing effectiveness had a plateau beyond
Materials Science

The results of this study suggest that the effect of rinse water which no more feeble deposit could be removed by water alone.
velocity on improving rinse effectiveness was always significant
within the tested range of velocities (0.72 to 2.26 m/s) at each Rinse effectiveness of reused rinse water
temperature (22 °C, 45 °C, and 67 °C). Hoffman (1983) found
As indicated previously, rinse effectiveness of water is impacted
a similar trend in an earlier investigation. He suggested that
by the length of contact time between the water and the fouled
4 m/s was the maximum effective velocity, below which an
surface. In order to explore other opportunities for reducing wa-
increase in velocity always caused a noticeable improvement in
ter use, the number of times water can be recycled during the
cleaning effectiveness while further increasing the velocity did not.
rinse step was investigated. Specifically, rinse water was recircu-
Goode (2012) suggested that increasing velocities cannot improve
lated through the pipe sections for 60 s before being drained, while
rinse effectiveness if the water temperature was above 50 °C.
normal rinsing involved single pass of unused water for 60 s. It
However this study found higher velocities to always significantly
was calculated that during the 60 s of recirculation, the rinse water
increase rinse effectiveness, even when the rinse water temper-
completed 3 passes through the entire system at the volumetric
ature was as high as 67 °C, which implies velocity has a higher
flow rate at 0.00284 m3 /s.
impact on cleaning effectiveness than temperature in this study.
As shown in Figure 6, there was no significant difference in
Alternatively, the effect of increasing temperature from 22 °C
rinse effectiveness between reused and unused water for the pa-
to 45 °C, then to 67 °C on the removal of the deposit did not
rameters evaluated. The effectiveness of the reused rinse water can
reveal the same trend. High temperature did not improve cleaning
be explained by the mechanical forces of water becoming the key
when combined with high velocity. A possible explanation of this
factor in impacting prerinse effectiveness. The concentration of
is that the water rinse effectiveness was controlled by mechanical
protein solids in the reused rinse water was not sufficiently high
force. The high velocity used in this study provides a strong me-
to impact mass transfer between the deposit film surface and the
chanical force which can breakup and remove deposits regardless
rinse water stream. Kulkarni and others (1975) concluded that
of the mass transfer rate of the detached deposit which was more
a cleaning solution can be used to accomplish 3 cleaning cycles
impacted by temperature. Although, at high velocities, tempera-
before redeposition of soil was observed. Reuse limits of cleaning
ture still contributed to the solubility of the deposit, this impact of
solutions and/or rinse water is related to the chemical reactions
temperature was masked by the effect of the increased mechanical
needed for breakup of the deposit and the diffusion of chemical
force. Yeckel and Middleman (1987) suggested that mechanical
reactants through the deposit layers. As reuse of cleaning solutions
force induced by the fluid was the primary factor in impacting
is dependent on chemical activity the number of recirculation cy-
removal of viscous film from rigid plates by water. Temperature
cles for a cleaning solution is expected to be much fewer than
seems to play a much more important role in cleaning when de-
rinse water.
tergent was introduced into the cleaning cycle (Goode and others
The results from this investigation clearly illustrate the unique
2013). In summary, the controlling factor during the rinse step of
interactions of rinse water temperature, velocity, and contact time,
CIP seems to be the mechanical force of the rinsing water. Water
when evaluating removal of a deposit film from the surface of a pipe
temperature has less impact on rinsing effectiveness comparing to
section. At higher rinse water temperatures, there is limited benefit
flow velocity. Other researchers have reached similar conclusions
of using higher velocities and longer contact times. Selection of
when using liquids containing detergents (Jensen and others 2005;
moderate rinse water flow rate, temperature and contact time can
Le Gentil and others 2010).
result in significant reductions in the amount of water required
to accomplish effective removal of the deposit during the rinse
Effect of contact time on rinse effectiveness
step. In addition, the amounts of energy required can be reduced
The results in Figure 5 illustrate the influence of water veloc- by using water at a lower temperature. Finally, the “reuse” of
ity on the residual film after different contact times at 45 °C. rinse water may provide additional opportunities to reduce water
The differences in rinse water effectiveness between the shorter requirements.
and longer contact times were significant at low water velocity
(0.72 m/s) while at the higher water velocities (1.25 and 2.26 m/s),
no significant differences were observed. It is believed that dur- Conclusions
ing the 1st seconds of rinsing, the initial removal of deposit The results of this investigation demonstrate the opportunities
is significantly large compared to the remaining contact time for reducing the amounts of water required for the rinse step

E1496 Journal of Food Science r Vol. 80, Nr. 7, 2015

Effectiveness of rinse water during in-place . . .

during CIP operations. The following are specific conclusions Dunsmore DG. 1983. The incidence and implications of residues of detergents and sanitizers in
dairy products. Residue reviews. New York: Springer. p 1–63.
based on results and analysis completed: Fryer PJ, Christian GK, Liu W. 2006. How hygiene happens: physics and chemistry of cleaning.
r A procedure for establishing a uniform and reproducible film
Int J Dairy Technol 59(2):76–84.
Fryer PJ, Robbins PT. 2005. Heat transfer in food processing: ensuring product quality and
safety. Appl Therm Eng 25(16):2499–2510.
on the interior surface of a pipe section has been developed Fryer PJ, Robbins PT, Asteriadou IK. 2013. Current knowledge in hygienic design: can we
and demonstrated. The magnitudes of the deposit were uni- minimise fouling and speed cleaning? Advances in food process engineering research and
applications. New York: Springer. p. 209–227.
form, and differences among various pipe diameters were not Gillham CR, Fryer PJ, Hasting APM, Wilson DI. 1999. Cleaning-in-place of whey protein
significant (P > 0.05).
r
fouling deposits: mechanisms controlling cleaning. Food and Bioprod Process 77(2):127–36.
Gillham CR, Fryer PJ, Hasting APM, Wilson DI. 2000. Enhanced cleaning of whey protein
The rinse effectiveness decreased from 94.9% to 73.1% as rinse soils using pulsed flows. J Food Eng 46(3):199–209.
water velocity and temperature were decreased.
r
Goode KR. 2012. Characterising the cleaning behaviour of brewery foulants-to minimise the
The rinse water effectiveness was improved significantly cost of cleaning in place operations. [PhD thesis]. Birmingham, U.K.: School of Chemical
Engineering, Univ. of Birmingham. 282p. Avaliable from: EthOS.
(P < 0.05) by increasing the temperature of rinse water from Goode KR, Asteriadou K, Fryer PJ, Picksley M, Robbins PT. 2010. Characterising the cleaning
22 °C to 45 °C. The improvement was not significant when mechanisms of yeast and the implications for Cleaning In Place (CIP). Food and Bioprod
Process 88(4):365–74.
the temperature was increased to 67 °C, and higher velocity Goode KR, Asteriadou K, Robbins PT, Fryer PJ. 2013. Fouling and cleaning studies in the
was used. food and beverage industry classified by cleaning type. Compr Rev Food Sci Food Safety
r 12(2):121–43.

E: Food Engineering &

Increasing the contact time between the deposit film and rinse Grant CS, Webb GE, Jeon YW. 1997. A noninvasive study of milk cleaning processes: calcium

Materials Science
water did not improve rinse water effectiveness at high temper- phosphate removal. J Food Process Eng 20(3):197–230.
Guzel-Seydim ZB, Wyffels JT, Greene AK, Bodine AB. 2000. Removal of dairy soil from heated
atures and water velocities. stainless steel surfaces: use of ozonated water as a prerinse. J Dairy Sci 83(8):1887–91.
Hoffman W. 1983. CIP of straight pipes as a function of surface roughness and other factors.
In general, the results of this investigation describe approaches [PhD thesis]. Staffordshire, Germany: Kiel Univ.
Jensen BBB, Friis A, Bénézech T, Legentilhomme P, Lelievre C. 2005. Local wall shear stress
to reducing the water and energy requirements for CIP operations variations predicted by computational fluid dynamics for hygienic design. Food and Bioprod
in food processing facilities. The potential for reuse of rinse water Process 83(1):53–60.
Jurad-Alameda E, Bravo-Rodriguez V, Bailon-Moreno R, Nunez-Olea J, Vaz DA. 2011. Fatty
provides additional opportunities for reduction of water require- soils removal from hard surfaces in a Clean In Place system. J Food Process Eng 34(4):1053–70.
ments without influencing rinse effectiveness. Jullien C, Bénézech T, Carpentier B, Lebret V, Faille, C. 2003. Identification of surface char-
acteristics relevant to the hygienic status of stainless steel for the food industry. J Food Eng
56(1):77–87.
Acknowledgement Kulkarni SM, Arnold RG, Maxcy RB. 1975. Reuse limits and regeneration of solutions for
cleaning dairy equipment. J Dairy Sci 58(8):1095–100.
The authors wish to acknowledge the generous donations from Le Gentil C, Sylla Y, Faille C. 2010. Bacterial re-contamination of surfaces of food processing
Dale A. Seiberling to the Food Engineering Research Laboratory lines during cleaning in place procedures. J Food Eng 96(1):37–42.
Liu W, Zhang Z, Fryer PJ. 2006. Identification and modelling of different removal modes in the
at The Ohio State Univ. cleaning of a model food deposit. Chem Eng Sci 61(22):7528–34.
Mercadé-Prieto R, Chen XD. 2006. Dissolution of whey protein concentrate gels in alkali.
AIChE journal 52(2):792–803.
Author Contributions Plett EA. 1985. Relevant mass transfer mechanisms during rinsing. Fouling and cleaning in food
Mengyuan Fan collected test data, interpreted the results and processing. Madison: Univ. of Wisconsin. p 395–409.
Schöler M, Föste H, Helbig M, Gottwald A, Friedrichs J, Werner C, Majschak JP. 2012.
drafted the manuscript. David Phinney designed the cleaning sys- Local analysis of cleaning mechanisms in CIP processes. Food and Bioprod Process 90(4):
tem used in this study, as well as assisted in experimental design, 858–66.
Seiberling DA. 1955. Engineering for CIP. Milk Plant Monthly, August.
application and data analysis. Dennis R. Heldman had overall re- Seiberling DA. 1968. Equipment and process design related to mechanical/chemical cleaning
sponsibility for the research and content of the manuscript. procedures. Proc Chem Eng 64(68):94–104.
Seiberling DA, Hyde JM. 1997. Pharmaceutical process design criteria for validatable CIP
cleaning. Cleaning validation. FL: Inst. of Validation Technology, Royal Palm Beach
References Stewart JC, Seiberling DA, Chowdhury J. 1996. The secret’s out clean in place. Chem Engi-
neering 103(1):72–9.
Bird MR, Bartlett M. 1995. CIP optimisation for the food industry: relationships between Timperley DA, Smeulders CNM. 1987. Cleaning of dairy HTST plate heat exchangers: com-
detergent concentration, temperature and cleaning time. Food and Bioprod Process 73(2):63– parison single and two stage procedures. Int J Dairy Technol 40(1):4–7.
70. Weidemann C, Stahl S, Nirschl H. 2013. Development of a qualitative test method for the
Boxler C, Augustin W, Scholl S. 2013. Cleaning of whey protein and milk salts soiled on DLC cleanability of polymer woven filter media. Food and Bioprod Process 91(4):515–24.
coated surfaces at high-temperature. J Food Eng 114(1):29–38. Wilson DI, 2005, Challenges in cleaning: recent developments and future prospects. Heat Transf
Christian GK. 2004. Cleaning of carbohydrate and dairy protein deposits. [PhD thesis]. Birm- Eng 26:51–9.
ingham, U.K.: School of Chemical Engineering, Univ. of Birmingham. 263 p. Avaliable from: Xin H, Chen XD, Özkan N. 2002. Cleaning rate in the uniform cleaning stage for whey protein
EthOS. gel deposits. Food and Bioprod Process 80(4):240–6.
Cords B. 2001. Cleaning and sanitizing in milk production and processing. In: Marth EH, Steele Yeckel A, Middleman S. 1987. Removal of a viscous film from a rigid plane surface by an
J, editors. Applied dairy microbiology. 2nd ed. CRC Press London. impinging liquid jet. Chem Eng Commun 50(1-6):165–75.
Dif M, Blel W, Tastayre G, Lendormi T, Sire O. 2013. Identification of transfer mechanisms
involved in soiled CIP solutions regeneration at extreme pH and high temperature. J Food
Eng 114(4):477–85.

Vol. 80, Nr. 7, 2015 r Journal of Food Science E1497